Diffusionism in Archaeological Theory: The Good, The Bad, and The Ugly

Diffusionism in Archaeological Theory 2 The Good, the Bad, and the Ugly A L IC E A . S T OR EY A N D T ER R Y L . J O N ES Diffused culture material often contains concrete or specific elements by which the fact of diffusion can be subsequently recognized even in the absence of a record of the event. In some cases it happens that the diffusion is definitely piecemeal; only fragments of a larger complex or system reach the affected culture or are accepted by it. In this event, the fragments or isolated items may be put into an entirely new context in the culture which they enter. —KROEBER (1940:1) T his volume deals with archaeological, linguistic, and bio- logical evidence for contact between the Pacific islands and the Americas and thus relies heavily on migration, interaction, and direct cultural diffusion as explanatory variables. In the 1970s diffusion as a concept was generally marginalized by the “New Archaeology,” which tried to distance itself from earlier types of explanation, particularly migra- tion and diffusion. For New Archaeologists, to employ these concepts was contra a scientific approach to archaeological interpretation (Ucko 1995). Then as now, however, basic culture history remained a necessary compo- nent of archaeological research, and it is clear that there is still a need for development and refinement of cultural chronologies that rely fundamen- tally on notions of diffusion of goods and ideas between groups of people. Hyperdiffusion, in which diffusion is used to explain all phenomena 7 8 ALICE A. STOREY AND TERRY L. JONES observed in the archaeological record, implies that independent invention was the property of only a select few in prehistory; it is this extreme view that tarnished early diffusionists and led mainstream archaeologists away from the concept. This volume includes contributions from archaeologists and linguists working in both the Pacific and the Americas, who acknowl- edge that the prehistoric people they focus their studies on were sophisti- cated, innovative, intelligent, and also interactive. It is this interaction that we are particularly interested in, and we will explore a range of evidence and theories that address the possibility of cross-cultural contact events. Each of the chapters in this volume presents hypotheses developed using the most up-to-date archaeological, linguistic, ethnographic, and genetic information, which in turn will require further investigations and testing as the study of prehistoric contact moves forward. The purpose of this chapter is to present a brief history of archaeologi- cal thought on the principle of diffusion, particularly as it applies to the Pacific and the Americas. The work strives to set a tone for the volume by outlining the ways in which the concept came to be so heavily disfavored in the 1970s. In arguing that diffusion still has value as an explanatory principle, we also hope to separate ourselves from publications that employ diffusion in racist or ethnocentric frameworks in order to justify religious or political ideas of culture origins and change. In this chapter, we set out specifically to explore the definitions of diffusion and the history of dif- fusionist thought from the 19th century to the present, particularly as it relates to Polynesian and American archaeology. We then explore criteria that may be used to distinguish artifacts of diffusion in the archaeological record in order to develop a rigorous set of criteria to establish contacts in prehistory. Defining Diffusion Diffusion has been defined as “transmission by contact, and thus the spread of norms from one person to another and more broadly from people of one local tradition to another” (Rouse 1986:1). Rogers (2003:5–11) defined diffusion as “the process in which an innovation is communicated through certain channels over time among members of a social system; emphasizing the importance of novel concepts or technology.” A key characteristic of diffusion is interaction, being contact among individuals and social groups and the mechanism by which cultural traits diffuse both through internal and external contacts (Rouse 1986:9). The areas in which these sorts of exchanges take place, either physically or cognitively, are called interaction  DIFFUSIONISM IN ARCHAEOLOGICAL THEORY 9 spheres (Rouse 1986:9). These spheres are thought to be made up of rep- resentatives from two or more culture areas. Kroeber (1962:15) suggested that the term culture area contained an implicit developmental history, as did taxonomy in biology. There are a variety of models for how cultural diffusion takes place. One model views diffusion as epidemic or contagious. This means that cul- tural behaviors spread through contact between peoples. These behaviors emerge in central hearths and then spread out in wave-like patterns from the center (Dickson 1988; Morrill et al. 1988). This could also be described as spatial diffusion (Morrill et al. 1988). Later assessments, particularly by geographers, found that not all groups in the path of the wave adopt a diffused phenomenon, and thus an exploration of barriers and resistance was required (Hudson 1972:22). The wave may be disrupted by rejection of the innovation by some groups or the effects of social or physical barri- ers (Hudson 1972). Innovations may also be spread from several different places at different times, and thus spread in several independent waves. Diffusion can also occur hierarchically, when innovations spread between large centers first and then out to smaller towns and villages (Morrill et al. 1988). This may also provide a good alternative explanation when phe- nomena do not spread in a wave-like pattern. However, hierarchical dif- fusion is hampered when the agents of change do not spread innovation to centers first (Hudson 1972). If agents of introduction are local, they may go through known political channels to introduce new things, such as the presentation of chocolate by Maya dignitaries to Prince Philip of Spain, which was facilitated by Dominican friars familiar with the established hierarchy (Weinberg and Bealer 2001:55). Yet accidental contact between different cultural groups is not likely to proceed through official channels, or if it does, it is not necessarily going to occur in the large urban centers. Other types of transmission of goods or ideas include relocation diffusion, which is achieved through migration and intrusion into new areas spread- ing culture or cultural traits with waves of migrants instead of through communication channels (Dickson 1988:65). Stimulus diffusion occurs when one group copies the material culture of another without any tech- nical knowledge of how the trait was initially produced, generally leading to an innovative new type (Jett 1971:45). In this case there will be no evidence for developmental stages in the new culture area. This is distinct from parallel evolution, in which two cultures develop the same complex trait through a series of independent but similar developmental steps from a different starting point (Jett 1971:145). This results in an analogous trait, which may be mistaken as evidence for diffusion. 10 ALICE A. STOREY AND TERRY L. JONES Homologous and Analogous Traits In biology homologous traits typically refer to any character that is simi- lar due to shared ancestry, such as the fact that the bones found in the wings of a bat and the arm of a human are the same through a common mammalian ancestor. In anthropology the application of homologous traits can be extended to any technological or behavioral attribute that shows a degree of structural similarity or correspondence due to his- torical or genetic connections. Not all similar behaviors are the result of relatedness or diffusion; similar traits may evolve independently due to convergence (Eerkens and Lipo 2007). Analogous traits are those that in biology are said to be functionally similar but come about as a response to environmental pressures, not through evolutionary relation- ships. An example of such a trait are the wings of birds, bees, and bats; each provides the ability to fly, but they are the result of very different evolutionary trajectories. In anthropology analogous traits are behaviors or technology that are developed by multiple groups independently due to similar environmental constraints, such as agriculture, once thought to be a homologous trait diffused from one center to the rest of the world. Boas (1911) came to favor convergence as an explanation for indepen- dent developments. The true challenge in archaeology is defining a trait as analogous or homologous and in the case of homology finding the required technological or behavioral connection in the archaeological record of the cultures sharing the trait. The desire to see every trait as homologous was the problem with many of the early hyperdiffusionists and became an impediment to continued use of the concept of close- neighbor diffusion in contemporary paradigms. A History of Diffusionist Thought Diffusion, transoceanic diffusion in particular, was a dominant theme in the earliest thinking about the prehistory of the Americas (see chap- ter 4). In the 18th and 19th centuries in the Americas, questions about the identity of the moundbuilders resulted in hyperdiffusionist argu- ments, due to racist ideas about the abilities of the native Americans. Conventional wisdom among American academics up to the 1890s was that the mounds were the works of a lost tribe of Israel, Phoenicians, or the Norse—anyone except American Indians. Such theories implied transoceanic diffusion, and so the earliest applications of this principle  DIFFUSIONISM IN ARCHAEOLOGICAL THEORY 11 in American archaeology were linked with racism and ethnocentrism. The moundbuilder controversy was resolved in 1894 based on the ar- chaeological and ethnohistorical findings of Cyrus Thomas, who showed unequivocally that American Indians had constructed and used mounds (Thomas 1985 [1894]). From this point onward it was no longer nec- essary or proper to invoke transoceanic diffusion to explain American Indian cultural patterns. With the emergence of the cultural historical paradigm in North American archaeology at the beginning of the 20th century, diffusion remained a viable explanatory variable, but transoce- anic diffusion was no longer seen in a positive light. In a very real sense, American archaeology established itself as a legitimate scientific discipline largely by debunking transoceanic diffusion. Recognition that American Indians had indeed built the mounds was also an important step toward eliminating racist paradigms in the social sciences. The concepts of migration and diffusion emerged as the dominant ex- planations for culture change at the beginning of the 20th century largely as a reaction against evolutionism, which had dominated anthropological thinking in the late 19th century. Particularly problematic were the openly racist and ethnocentric ideas embodied by social Darwinism (Trigger 2006:217). In the United States the idea of diffusion was advanced largely by the highly influential Franz Boas, who used the idea in tandem with the closely related concepts of cultural relativism and historical particularism. The latter idea was the fundamental concept underlying the development of cultural chronologies in American archaeology as they were developed by archaeologists with intellectual lineages tied to Boas. The applications of diffusion developed by Boas and his students were decidedly and conscien- tiously nonracist and nonethnocentric. Still, some early diffusionist theories remained racist in their interpre- tations of the past. This was particularly true in the writing of the most influential hyperdiffusionist, Sir Elliot Grafton Smith, who argued that the concept of civilization was invented only once, independently (in prehis- toric Egypt), and from there it diffused to all other civilized places in the world, including prehistoric Mesoamerica and South America. Egypt was not only the cradle of civilization but also of all innovations related to civilization, and it was the migrations of the Egyptians that served to spread knowledge around the world (Smith 1933). He also believed that civiliza- tions would decline when cut off from their parent society. Others, who subscribed to the Viennese school of anthropology developed in the late 1800s, believed that Asia was the source of all innovation and that the level 12 ALICE A. STOREY AND TERRY L. JONES of advancement achieved by a civilization could be directly correlated to the degree of exposure they had to the parent culture (Trigger 2006:219). Regardless of which culture was identified as mother of all others, anthropology of the late 19th and early 20th centuries was of the view that diffusion, migration, and invasion were the main processes that af- fected change in prehistory (Politis 1995:212). Early diffusionist paradigms also suffered from the idea that transmitted features could be identified intuitively (Graebner 1911). It was this aspect of these theories that later provoked the ire of the New Archaeologists. Whether overtly racist or not, migration and diffusion were certainly at the core in the early cultural historical paradigm, as the movement of people was most often cited as the mechanism for the transfer of cultural traits. However, there were differences between large-scale migrations and small-scale contact events. Major changes that appeared as archaeologically instantaneous were thought to be the products of migration. Petrie (1939) saw migration as the most important factor in culture change, and he be- lieved that the transfer of culture and technology from one group to an- other would also result in biological changes (Trigger 2006:221–222). This mechanism was adopted in part by V. Gordon Childe, who also viewed diffusion as the spread of functionally beneficial traits or more attractive styles from one culture to another, whereas migration resulted in the re- placement of one group by another or cultural mixing (Trigger 2006:246). While Childe himself did not maintain his emphasis on diffusion for the whole of his career, the principle became synonymous with the cultural historical paradigm. By the 1920s, fieldwork and artifact analyses revealed significant dif- ferences between the Old and New Worlds that contradicted most hyperdiffusionist views, discrediting the radical fringe of the discipline (Trigger 2006:221). Hyperdiffusionism did not entirely disappear, but it was generally abandoned by mainstream academia. Much more reasonable constraints were being imposed on archaeological interpretations of dif- fusion. For example, Julian Steward (1929) suggested that when cultural traits are shared by two cultures the probability of independent invention is directly proportional to the difficulty of communication between the two places and the uniqueness of the element. More importantly, by the 1950s cultural anthropology had developed the two theories that would undermine diffusionism: Steward’s cultural ecology and Leslie White’s cultural evolutionism.  DIFFUSIONISM IN ARCHAEOLOGICAL THEORY 13 Diffusionism and South American Archaeology The history of anthropological thought in South American archaeology, as in most places in the world, can be seen as connected with political and economic trends (Politis 1995). During the late 1800s politicians in a variety of South American nations were very keen on taking advantage of foreign investment, capital, and culture. In this climate the vast majority of archae- ologists working in South America were foreigners who did not publish in Spanish, tended to focus on cultural evolution, and largely ignored any evidence for migration or diffusion (Politis 1995:198). In the early 1900s more and more immigrants began moving into South America, and this led archaeologists to think more critically about the effects of migration and the diffusion of cultural traits. Imbelloni, of the Vienna school, pub- lished papers in Spanish, which made the diffusionist perspective accessible at a time when such ideas were on the minds of Spanish-speaking South Americans. Imbelloni published extensively on evidence for Polynesian contact with the Americas, focusing both on material culture (Imbelloni 1925, 1928a, 1931, 1953) and linguistics (Imbelloni 1928b, 1929a, 1929b). He further suggested that cranial measurements from Peru indicated some gene flow between Oceania and the New World (Imbelloni 1943). Also influential in developing diffusionist paradigms for the settlement and enculturation of the Americas was Rivet. Not limited to contacts between only Polynesia and America, Rivet considered cultural contacts between the Americas and the wider Pacific, including Australia (Rivet 1909, 1925, 1926, 1927, 1956). Métraux, who studied with Rivet, also encouraged a cultural historical approach to understanding the prehistory of Argentina (Politis 1995). He was, however, very critical of prehistoric contacts between Polynesia and the Americas (Métraux 1940:416; 1951). Instead, he offered evidence for parallels between South America and ancient Greece (Métraux 1925). Meanwhile, Uhle brought a diffusionist perspective to his work in Peru, Chile, and Ecuador (Politis 1995). He of- ten pointed to China/Asia as the source of Central American high cultures (Uhle 1930, 1942a, 1942b). Even after the Second World War, due largely to the influence of Menghin (1960, 1967) and Imbelloni, archaeology in Argentina retained a strong diffusionist bent. This was also true in Ecuador with the work of Meggers, Evans, and Estrada (Meggers et al. 1965; Politis 1995). In several South American countries, politicians retained a conservative outlook and wanted archaeologists to emphasize the Spanish colonial past and down- 14 ALICE A. STOREY AND TERRY L. JONES play the unique and important accomplishments of the indigenous peoples (Politis 1995). This was often accomplished by ousting archaeologists who were seen as giving too much credit to ancient peoples and, by extension, their indigenous ancestors. The Anti-Diffusion/Culture History Backlash in American Archaeology Anthropological hypotheses based on migration and interaction were deposed in much the same way they came into vogue, with a massive backlash ending in a temporary ban on their application in archaeologi- cal explanations. Relying on cultural ecology, cultural materialism, and cultural evolution, the New Archaeology of the 1970s shunned culture history, regarding it as “unscientific” and thus unworthy of further review (Adams 1978). Certainly, it did not help that several of the world’s most prominent archaeological scholars had by this time openly denounced dif- fusionism in no uncertain terms. Childe (1962:209), for example, stated that “discussions of diffusion are apt to degenerate into combats wherein only dust is diffused or else to ascend into an ether so diffuse that the disputants are left balancing probabilities inherently unsusceptible of sta- tistical treatment,” while John Rowe (1966:334) stated that “diffusionism is a menace to the development of sound archaeological theory based on comparative studies; it distracts archaeologists from such studies and seeks to destroy the basis of comparison. Diffusionist arguments are defective in principle.” The alternative view that emerged in the 1970s, influenced heavily by the writings of Lewis Binford, was that culture change could, in most cases, be explained as an in situ response to local population and climate variation. As suggested in Binford’s influential 1968 paper, demographic pressure was seen as an especially powerful influence over cultural innova- tion. Prehistory could be best understood at the local or regional level with the aid of contextual studies that provided information on changing en- vironmental variables. Cohen (1977) articulated the demographic pressure model of subsistence change more broadly. In these cases external stimulus was simply not an important factor in bringing about cultural innovations; demographic stress precipitated by climate change, growing human num- bers, or some combination thereof was seen as the key variable. This view was exceptionally well represented in California by Chartkoff and Chart- koff (1984), who portrayed virtually all of the technical and sociopolitical developments over fourteen thousand years of hunter-gatherer prehistory  DIFFUSIONISM IN ARCHAEOLOGICAL THEORY 15 as responses to population growth and pressure. In so doing, these authors generally ignored California’s complex linguistic mosaic, long interpreted as the product of a series of migrations and population replacements (see Heizer 1949a; Moratto 1984), and portrayed the region’s prehistory exclu- sively in terms of demography and ecology. Some saw this as a welcome break from a long history of obsession with migration and diffusion, which had often posited overly simple correlations between material culture and ethnic identities (Hughes 1992). Culture History, Historical Anthropology, Interaction, Integration, and Borrowing: A Pacific Perspective An in-depth discussion of the developments and application of theory in Pacific archaeology is not appropriate in the theoretical discussions of diffusionism that are the focus of this chapter. Those interested in an overview of the major theoretical tenets that have been applied in Pacific archaeological studies are referred to Kirch (2000a) for a review. However, a brief discussion of the distinct trajectory that Pacific studies have taken is definitely warranted in this case. During the early 20th century, the Pacific was not without its hyperdiffusionists, including Handy (1927), who saw parallels between Polynesian religions and those of India, China, and the Northwest Coast of America, and Linton (1936), who believed that no culture owed more than 10 percent of its culture to independent inven- tion. Yet despite ill informed and racist paradigms that sought outside influ- ences as the source of most Pacific cultural elements, serious archaeologists were selective in ignoring hyperdiffusionist arguments without rejecting all potential diffusion events, including ephemeral prehistoric contacts and interactions with the Americas, as represented by the transference of the sweet potato into Polynesia. This may be because Pacific archaeology was firmly rooted in a multidisciplinary type of culture history that was due in large part to the organization of the Bayard Dominick Expeditions by the Bishop Museum in the 1920s. These expeditions were equipped not only with archaeologists but also ethnologists and guidelines for gathering physical information from modern populations including height, weight, and even nostril width in order to document biological diversity (see Sul- livan 1921 and Sullivan et al. 1922 for examples). Kirch (2000a) referred to this as the first modern holistic anthropological study, a label that has become important to more contemporary theoretical debates. However, the limitations of archaeological dating at the time led to an emphasis on 16 ALICE A. STOREY AND TERRY L. JONES the ethnographic aspects of the expeditions and a general feeling that the Pacific was a region without history (Kirch 2000a). The advent of radiocarbon dating and increasingly sophisticated tech- niques for historical linguistics brought to an end the “dominance of comparative ethnology and the diffusionists” (Kirch 2000a:29). However, this new era of dating led to the development of artifact-focused culture histories for the Pacific, which was and is still useful in understanding prehistory but has its limits for rich reconstructions of prehistory (Lyman et al. 1997). Roger Green proceeded to apply archaeological techniques and theories that focused more on settlement patterns than portable arti- facts (Green 1970). This shifted the focus from a subset of archaeological assemblages to all aspects of a site, including faunal and floral assemblages and structural remains. The New Archaeology, which created a huge paradigm shift in the Americas, was less dramatic in the Pacific. New ideas from this model were incorporated within island-specific studies but did not substantially shift theoretical discussions of the wider settlement of the Pacific. In this broader context diffusion continued to be discussed, although more often in terms of borrowing and interaction (e.g., Allen and Gosden 1996; Weisler 1997). As time progressed, there were those who called for greater emphasis on ecology and less attention to basic culture history in Pacific studies (e.g., Howard 1979), referring to culture historians as “diffusionists.” Bellwood (1980) rejected this characterization and distinguished well-grounded cul- tural historical explanations that acknowledged diffusion from poorly con- ceived theoretical models that defied cultural historical realities. Influenced by Dunnell (1980) and others, several Pacific archaeologists began adopting aspects of evolutionary theory, but with a primary focus on human ecology (e.g., Allen 2004a; Kirch and Hunt 1997; Morrison and Hunt 2007), as well as the integration of principles of evolutionary ecology (e.g., Aswani and Graves 1998; Morrison and Hunt 2007). There was an increasing focus on the adaptations of specific populations to specific island environments. In more recent studies, this has further expanded to discus- sions of human ecodynamics (e.g., Kirch 2007). Back in the evolutionary realm of Pacific anthropology, the phylo- genetic model proposed by Kirch and Green (1987) for Polynesia was described by Dunnell (1987) as Lamarckian. However, Kirch and Green (1987) argued that new cultural patterns were transmitted not only gen- erationally but also laterally and between genetically unrelated groups. They felt this could account for homologies between groups, which arose by transmission from a single ancestral group that may or may not have  DIFFUSIONISM IN ARCHAEOLOGICAL THEORY 17 been genetically related to both groups. This allowed for cultural diffusion between culturally diverse groups. These theories were not universally em- braced in the Pacific (e.g., Gosden 1987; Terrell 1987, 2004; Terrell and Welsch 1990) but continued to be developed by Kirch and Green (2001) as part of their holistic archaeology, which included information from linguistics, biological anthropology, and oral traditions. In this way ties to the culture history roots of Pacific archaeology persisted and Green (2000e) felt these were essential for the development of contemporary holistic anthropological and archaeological studies. However, the connotations of culture history and its ties to racist diffusionism led to a rebranding of this “New Culture History” as historical anthropology or the anthropology of long-term history (Green 2000e). The influence of interaction in the historical development of Pacific cultures was also highlighted by Green in a series of papers about the “Triple I model” for Lapita origins (Green 1991, 2000a, 2003). This model of Intrusion/Innovation/Integration was proposed to account for the pat- terns observed in the archaeological record of New Oceania circa 3,500 BP, when new artifacts appeared, including pottery, representing a new cultural complex called Lapita. This theory employed a type of diffusion that was cooperative and not the sort that Allen and Gosden (1996:183) had disavowed when they stated, “our main argument with the Out-of- Asia model is the notion of movement from a higher centre of culture bringing advances in subsistence, technology and society to less progressive regions.” Instead, the Triple I model contrasted the equally important con- tributions of independent invention, indigenous knowledge, and cultural transmission, generally through borrowing. It was a sort of cooperative dif- fusion, which allowed new arrivals to Near Oceania to incorporate aspects from their travels through Sulawesi, Sabah, the Bismarck Archipelago, and even coastal New Guinea. As Green described it: Not surprisingly, and from the start, ensuing transformation involved the integration (into its cultural core) of a set of elements that were derived from a range of populations with other cultures already resident, within either the previous or new homeland zone, probably on the larger adjacent islands. (2003:112) The interaction between unrelated cultures is not only observed in Near Oceania. It has been described between prehistoric residents of the Solomon Islands and Vanuatu with people in the Caroline Islands (Athens 1990; Ayres 1990; Intoh 1997, 1998; Rainbird 2004). It is also impli- cated in the development of hybrid strains of breadfruit, which started a 18 ALICE A. STOREY AND TERRY L. JONES “breadfruit revolution” in the Pacific (Petersen 2006). The same sorts of behaviors outlined in Green’s Near Oceanic Triple I event have been postulated for contacts between Micronesia and more southerly parts of Remote Oceania by Addison and Matisoo-Smith (2010). Perhaps Kirch sums this section up best with his observation: Pacific archaeologists have been studying interaction among island societies for a long time and both isolation and interaction are fundamental concepts for interpretation and understanding of culture change in Oceania. (2000a:41) Processualism Plus: Politics and Evolution Back in the Americas, theoretical developments in the 1990s were a little more forgiving of the concept of diffusion than those of the processual rev- olution. While there is no consensus on exactly where the discipline stands as a result of developments of the 1990s, the decade was clearly highlighted by the postmodernist critique of processual archaeology. Among other things, postprocessualists pointed to the limitations of environment as the prime mover in explanations of past human behavior (see Brumfiel 1992), and they sought to expand the range of factors considered in explanatory models while at the same time they also encouraged greater awareness of the relationships between modern-day social issues (e.g., gender, ethnic- ity, inequality) and archaeological interpretations. Many advocated greater political conscientiousness among archaeologists and a more politicized research process. As in the Pacific, North America also witnessed the rise of a school of neo-Darwinian or evolutionary theory in the 1990s. Focused on optimal foraging and other applications of human behavioral ecology, neo-evolu- tionary theorists sought to build on the successes of cultural ecology; thus their models tend to emphasize local and regional adaptation and to dis- courage the importance of external influences. Dunnell (1986:38–39) sug- gested that the shift from culture history to the New Archaeology shifted the focus of archaeological research away from homologous traits to stress analogous similarities. Hegmon (2003) coined the phrase processualism plus to refer to the state of archaeological theory in the new millennium. Whether or not her char- acterization is accurate, it is clear that archaeology in the 2000s features a wider range of explanatory variables than the simple ecology of the pro- cessual era. This broader range should have been beneficial to recognizing cases of diffusion; however, the overriding concern for modern-day politi- cal identities that also emerged at this time has generally mitigated any such  DIFFUSIONISM IN ARCHAEOLOGICAL THEORY 19 possibility. Some postmodernists have even criticized simple migration studies in archaeology. Ucko (1995:12), for example, stated that culture history “at its most insidious (when coupled with theories of migration) has been able to deprive whole peoples of any legitimate past.” An even clearer expression of the influence of politics on postmod- ernist thinking in archaeology and its impact on perceptions of diffusion is seen in Arnold’s (2007) criticism of Jones and Klar’s (2005) case for a Polynesian contact event in southern California. The article’s title, “Credit Where Credit Is Due: The History of the Chumash Oceangoing Plank Canoes,” speaks to the author’s concern for respecting modern-day ethnic groups in interpretations of the past; her main point is that in suggesting that Chumash sewn-plank boat technology originated from contact with Polynesians, Jones and Klar were discrediting the technical and creative capabilities of Chumash peoples. For Arnold, her definition of modern-day ethnic identities is of paramount importance; she implies that transoceanic diffusion is inherently ethnocentric or racist despite the fact that Jones and Klar stated repeatedly that such was not the case. While there are unmistakable elements of racism in early hyperdiffu- sionist models, this does not mean that the entire idea of culture contact and exchange should be dismissed. Such models for the introduction of technology and ideas do not necessarily “conflict with attempts by indig- enous groups to reaffirm political and ethnic identities through connection with the archaeological record” (Jones and Klar 2005:457). It is true that politics and nationalist attitudes led to the all-out rejection of diffusionist ideas (Kauffman Doig 1980), but learning from the misuse of diffusion in the past should allow us to examine instances of diffusion and culture contact within well-established indigenous histories. Migration and Diffusion Revisited While hyperdiffusion and the denunciation of local innovation are ideas that should be left in the annals of anthropological thought, denying that migration and diffusion ever took place is to ignore certain key aspects of human behavior and history. Intelligent and innovative humans would certainly never have been averse to acquiring subsistence items or tech- nologies that could enhance their lives, nor would they have been likely to refuse gifts or exotic items from visitors. Thus while hyperdiffusion should be simply discarded, it remains reasonable to try to establish criteria for recognizing episodes of diffusion of ideas, technology, or subsistence items that resulted from cultural contacts—even those of relatively brief 20 ALICE A. STOREY AND TERRY L. JONES duration. The challenge is to evaluate old theories with new evidence from large, chronologically controlled data sets in order to distinguish cases of diffusion in the archaeological record with a certain degree of confidence. Differentiating archaeological evidence for migration from diffusion and both from trade is what Burmeister (2000:540) has called a “methodologi- cal trilemma.” However, in the case of possible Polynesian contacts with the Americas, this problem is actually less challenging in that the issue at hand is simply whether or not contact of any type actually occurred. Due to the negative connotations of the term diffusion it has been replaced in many cases with other terms for the sharing of ideas or arti- facts such as cultural transmission. This is defined as information exchange through nongenetic mechanism, which may result in a similarity of behav- ior or material culture (Eerkens and Lipo 2007:240). Such exchanges are most likely to be evidenced on an individual level and thus will manifest in single artifacts, households, or mortuary finds rather than large cultural assemblages (Eerkens and Lipo 2007:226). Since the mode of transmission will not be preserved in the archaeological record, this is the most dif- ficult part of applying a cultural transmission paradigm (Eerkens and Lipo 2007:261). Adams (1978) felt that migration and diffusion did not necessarily need to be linked and potential cases should be assessed singularly as opposed to all-or-nothing paradigms. In his model, migration is used to account for site distributions, while diffusion is the main factor behind trait distribu- tions (Adams 1978:3). However, in cases of extreme discontinuity of traits migration was said to be the more likely explanation (Adams 1978:3). Many have suggested that the transfer of biological organisms, such as plants or animals, should be taken as proof positive of diffusion, since outside their natural range most plants and animals can only be dispersed by human agents (Carter 1998; Gade 1998; Jett 1971). Wescott (1998:256) felt that the transfer of domesticates was, in fact, one of the simplest forms of trade. Unfortunately, there are no “laws” about how a plant will dif- fuse (Carter 1963). A species important to one group of people may be of no interest at all to another; for example, the Maori quickly adopted the potato but were reluctant to plant maize (Best 1925; Yen 1974). In fact, a staple plant may be less likely to diffuse than a medicinal or ornamental plant, because when diffusion occurs between two agricultural societies the donor group may have a suitable alternative already in their botanical complex. Kelley (1971:61) suggested that one of the problems with diffu- sion is that a culture has to reach a certain level of complexity before  DIFFUSIONISM IN ARCHAEOLOGICAL THEORY 21 people have the capacity to adopt innovations from external contacts. A pre-agricultural society is unlikely to adopt the plow, for example. Diffusion is also tempered by the occupation or technical bias of both the donor and the receiver. For example, sailors will carry different cultural and material innovations than merchants or missionaries (Kelley 1971:63). Rogers (2003:36) believes that the perceived characteristics of an innovation determine whether it is adopted by a second group of people. This choice will involve their perception of the item’s relative advantage, compatibility with existing norms, complexity, triability, and observability. Criteria for Recognizing Diffusion As part of the ever-developing pantheon of archaeological theories, diffu- sionism has progressed through a number of developmental stages, which have included the creation of criteria for identifying a diffusion event. Criteria considered previously by Rouse (1958), Kelley (1971) and Meg- gers (1975) are as follows: Material Evidence ■ The identification of clear archaeological horizons in which the artifacts and styles are plainly intrusive to the region in which they occur (Rouse 1958). ■ The intrusive cultural artifacts must appear fully developed, with no evidence of early forms in the archaeological record of the receiving culture (Meggers 1975). ■ The development of intrusive artifacts must have a clear history of invention in the transmitting culture (Kelley 1971; Meggers 1975). ■ Traits are relatively contemporaneous in the archaeological record of two areas (Meggers 1975; Rouse 1958). ■ Traits do not meet the criteria for independent invention in that they are free of evolutionary, environmental, and functional limitations, which would imply they would be invented without need of contact (Meggers 1975). ■ Groups of unrelated innovations appear simultaneously in a single assemblage (Kelley 1971). ■ Tracing the movement of a people or artifact to a place of origin is possible (Rouse 1958). ■ Linguistic linkages confirm a likely source of introduction/ borrowing (Kelley 1971). 22 ALICE A. STOREY AND TERRY L. JONES ■ Identification of suitable cultural and environmental conditions for migration or diffusion (Rouse 1958). ■ The appearance of biologically distinct entities (plants or animals) outside their natural range (Kelley 1971). These criteria were the result of theoretical discussions and were not nec- essarily developed through evaluating available evidence for any specific case of diffusion. Rowe (1966), who was staunchly opposed to hyperdiffusionist argu- ments, nonetheless accepted that some episodes of diffusion did occur, but he cautioned that identification of diffusion is challenging in the archaeo- logical record. Applying these criteria made recognizing actual cultural contacts, such as those between the Vikings and the indigenous people of the east coast of North America much more difficult to detect, since the “influence exerted by the colonists was much less than diffusionist assump- tions call for” (Rowe 1966:335). This sentiment was echoed by Meggers (1998:104), who argued that many of the criteria for identifying cultural contact could not be met with the archaeological record of early European colonies in North America. The L’Anse aux Meadows site in Newfound- land, which provides unequivocal evidence of Norse exploration of the Americas, indeed fails to meet many of the criteria for diffusion listed above because the colony seems to have had little interaction with or effect upon local indigenous societies. Rowe (1966) felt that this rendered the diffu- sionist agenda largely untenable because it suggested that low-level contact events would be essentially unrecognizable in the archaeological record. He suggested that contact can only be proven archaeologically by the identifi- cation of colonies or trading posts and in rare cases by the repeated recovery of trade goods from one area that is in secure archaeological association with a second area. One of the true difficulties in documenting small-scale con- tact with the Americas is that if visitors to the coasts had stayed, they would more than likely be assimilated or absorbed into local populations. Rouse (1986:9) suggested this would make them very difficult to identify in the archaeological record. New methods not available in Rowe’s time such as mtDNA and nuclear DNA analysis and AMS radiocarbon dating, however, have the potential to provide new lines of evidence for interaction. Discussion A legacy of colonial agendas and racism in early studies has rendered the application of diffusion, particularly via transoceanic, cross-cultural con-  DIFFUSIONISM IN ARCHAEOLOGICAL THEORY 23 tact, dangerous for modern-day researchers. In seeking to develop an im- proved paradigm for diffusion, it is important to acknowledge this history and strive to be transparent in the development of new criteria and mod- els. In this book, we deal with evidence for Polynesian contact with the Americas, which has been debated for over a century and a half. As such, it would be a serious mistake to ignore such a substantial body of previous work. Regardless of the conclusions in many of the early hyperdiffusionist studies, some of the evidence presented by South American cultural histo- rians is useful and needs to be critically re-examined, which we attempt to do in chapters 3 and 4. Given the intimate knowledge of culture areas and the changes through time in local traditions it is also essential that archae- ologists in both potential donor and receiving areas work together. Due to a variety of environmental and functional needs of particular peoples at particular times, it might be very easy for an archaeologist whose focus is in one area to see parallels in other cultures that are in fact the result of inde- pendent invention or development. In these cases, any similarities between distinct cultures should be investigated independently by researchers who have established expertise in the receiving area. In this way the evolution of artifact types can be assessed in a regional setting, and if the artifact does indeed have no antecedents in the receiving culture then diffusion may be an appropriate explanation for its appearance. Cultural contacts between geographically disparate entities were not the norm in prehistory, nor did they necessarily have a large effect on either society. This point is critical for distinguishing wild-eyed hyperdif- fusionist arguments from the view we endorse in this volume. In no way do we wish to underrepresent the remarkable independent inventions, achievements, and related cultural trajectories of indigenous groups. In- stead, we suggest that occasional cultural contacts did occur in prehistory and while these did not have major ramifications, they nonetheless left archaeologically detectable traces. What is true for archaeology applies as well to linguistics. Linguists with a focus in one language area must resist the impulse to attribute to genetic or diffusional origin every apparent similarity with other languages (cf. Key 1984). However, in evaluating linguistic evidence, one must be able to discern what in the attested record of a language appears anomalous and not amenable to explanation by internal processes of internal development rather than possible borrowing from outside. As with the archaeological record, in the case of Polynesian-American contact, the linguistic record suggests occasional, episodic contacts, but ones that nonetheless left detect- able traces. 24 ALICE A. STOREY AND TERRY L. JONES It is quite possible that Pacific archaeologists are much more amenable to developing models that incorporate long-distance migration and diffu- sion, as the archaeological record of prehistoric Oceania has consistently revealed wide-ranging exploration and settlement as well as established interaction networks (Allen and Gosden 1996; Finney 1998; Green and Kirch 1997; Summerhayes 2000). However, within this localized area developments are recognized and given their due. For example, the de- velopment of the Tongan Maritime Empire and the feats of monolithic architecture on Easter Island are fascinating independent trajectories in- vestigated by archaeologists in the Pacific. Americanists are also starting to re-evaluate the utility of interaction and diffusion in prehistory—interac- tion did happen but local developments were still important and should be recognized alongside diffusion. Diffusion may manifest itself differently in two interacting cultures depending on a myriad of factors, including the nature of interaction (pur- poseful vs. accidental or peaceful vs. aggressive), the purpose of the trip (a canoe on a colonization voyage would have different supplies than one on an exploratory trip), and the wants and needs of each group of people (for example, the Polynesians took sweet potato but not maize from the Americas, due to their system of agriculture being heavily slanted towards tubers and arboriculture). The degree of diffusion is also going to differ. For example, once the sweet potato was introduced to Central Eastern Polynesia it spread to marginal East Polynesia, including Hawai’i, Easter Island, and New Zealand. On the other hand the sewn-plank canoe had a much more limited distribution in the Americas and did not diffuse to other groups of seafaring peoples living nearby. Thus, the modern diffu- sionist paradigm must not only strive to find firm, scientifically acceptable evidence for contact but must also evaluate more difficult evidence such as oral traditions within a complex range of human behaviors. This task is difficult and necessary and is the focus of the remainder of this volume. POLYNESIANS IN AMERICA Pre-Columbian Contacts with the New World EDITED BY TERRY L. JONES, ALICE A. STOREY, ELIZABETH A. MATISOO-SMITH, AND JOSÉ MIGUEL RAMÍREZ-ALIAGA ALTAMIRA PRESS A Division of Rowman & Littlefield Publishers, Inc. Lanham • New York • Toronto • Plymouth, UK Published by AltaMira Press A division of Rowman & Littlefield Publishers, Inc. Rowman & Littlefield Publishers, Inc. A wholly owned subsidiary of The Rowman & Littlefield Publishing Group, Inc. 4501 Forbes Boulevard, Suite 200, Lanham, Maryland 20706 http://www.altamirapress.com Estover Road, Plymouth PL6 7PY, United Kingdom Copyright © 2011 by AltaMira Press All rights reserved. No part of this book may be reproduced in any form or by any electronic or mechanical means, including information storage and retrieval systems, without written permission from the publisher, except by a reviewer who may quote passages in a review. British Library Cataloguing in Publication Information Available Library of Congress Cataloging-in-Publication Data Polynesians in America : pre-Columbian contacts with the New World / edited by Terry L. Jones ... [et al.]. p. cm. Includes bibliographical references and index. ISBN 978-0-7591-2004-4 (cloth : alk. paper) — ISBN 978-0-7591-2006-8 (electronic) 1. America—Discovery and exploration—Polynesian. 2. America--Discovery and exploration—Pre-Columbian. 3. Polynesians—America—History—To 1500. 4. Polynesians—America—Antiquities. 5. America—Antiquities. I. Jones, Terry L. E109.P65P66 2010 970.01—dc22 2010042149 ™ The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences—Permanence of Paper for Printed Library Materials, ANSI/NISO Z39.48-1992. Printed in the United States of America