OLYMPIC The Evolution and Diversification of Native Land Use Systems on the Olympic Peninsula A Research Design

Introduction

The aboriginal cultures of the Northwest Coast have long fascinated anthropologists interested in understanding the determinants of variation in hunter gatherer societies (Benedict 1934; Harris 1968:302-313; Lee 1968). The Northwest Coast has also attracted much attention from archaeologists interested in the development of organizational complexity and maritime adaptations among hunter gatherers (Fladmark 1975, 1982; Schalk 1977, 1978, 1981; Burley 1979; Ames 1981, 1985; Aikens et al. 1985). There can be no doubt that Northwest cultures observed in the early 19th century were unusual in comparison to hunter-gatherers in other areas of the world and scientific inquiry has a special interest in the unusual because exceptional phenomena provide the greatest challenges to explanations that are truly general It is not surprising that a major research problem for archaeologists working in the Northwest has been to explain how these unusual adaptations evolved, diversified, and, in certain cases such as the Makah, became fully maritime.

It has been widely recognized that the distinctive character of the Indian cultures of the Northwest Coast owes much to the area's highly productive and concentrated marine resources. Though quite variable from region to region along the coast, salmon, halibut, herring, a variety of other fish resources, and sea mammals provided the major food staples in the subsistence of these societies. Given the productivity and reliability of marine resources, the greatest enigma in Northwest Coast archaeology has been that good evidence for systematic and intensive use of marine resources extends little more than three to five millenia back into an archaeological record with a time depth of more than 10,000 years. Explicitly or implicitly, virtually all explanations for the evolution of the maritime adaptations of the Northwest Coast attempt to come to grips with this paradox.

From these observations, it would appear that there are two questions that are central to much of what has been written about the archaeology of the Northwest and which remain as primary questions for future research. These questions are:

How did the highly distinctive Northwest aboriginal cultures described historically evolve?

Why does evidence for intensive marine resource use only occur in the later portions of the archaeological record in the Northwest?

The remainder of this chapter is an effort to establish the theoretical context in which these kinds of questions can be addressed. Throughout the discussion that follows, there is a focus on land use systems and, before proceeding, it is necessary to define the usage of this term. The concept of land use system refers to the strategies by which essential resources are acquired by cultural systems and by which consumers are matched to those resources. Land use amounts to the combination of a subsistence system and a settlement system. A land use system could also be thought of as the spatial organization of subsistence (Schalk 1981:56).

The Bipolar Nature of Hunter-Gatherer Land Use Systems

If one views cultural adaptations as thermodynamic systems, then variations in the ways in which they process energy are of fundamental importance. Although terms such as "cultural complexity", "sociocultural complexity", and "organizational complexity" are frequently used in discussions of variability in hunter-gatherer systems, these terms are not often explicitly defined. Even less often are these terms linked explicitly to measurable characteristics of land use systems or the structure of energy flow through cultural systems. In an effort to develop an operational definition of "organizational complexity" for hunter gatherer land use, I have suggested the utility of focusing on the relationship between resource procurement and resource consumption:

Any economic system could be ranked along a scale of increasing distance in space-time between energy capture and consumption; organizational complexity might be said to increase as this distance increases... In this sense, the simplest imaginable foraging system would be one in which resource consumption takes place at the time and location of resource procurement. In such a hypothetical system, there would be no delayed consumption (food storage) and no transporting of resources....At the other end of the proposed spectrum of organizational complexity would be those modern industrial systems in which resources are typically transported over great distances and delayed consumption is the norm (Schalk 1981:55; emphasis added).

This view of complexity draws attention to two dimensions of a land use system that are of paramount importance—the degree of dependence upon food storage and the extent to which resources are transported from locations of procurement to places of consumption. These are the principal means by which human subsistence systems convey energy from the place/time of its capture to a place/time of consumption.

In many recent discussions of global patterning in hunter-gatherer adaptations, there is a dichotomous partitioning of the variability. In such discussions, there has been considerable attention to the role of food storage (c.f. Woodburn 1980; Binford 1980; Testart 1982; Ingold 1983; Kelly 1983) When recent and contemporary hunter-gatherers are examined, there is not a continuous spectrum of variability. Instead, variability is discontinuous and strongly correlated with the presence or absence of food storage:

To classify hunting and gathering societies into those with economy and social organisation based on immediate return, and those with economy and social organisation based on delayed return, is surprisingly easy. The polarity is nearly always marked. In each type there are a whole range of phenomena which appear to be closely dependent on each other (Woodburn 1980:98; emphasis added).

That there is such a marked polarity in hunting and gathering societies is consistent with the suggestion that the emergence of land use systems based upon systematic and intensive food storage represents an evolutionary threshold (Schalk 1977). Relative to immediate return cultural systems, those based upon food storage tend to require greater degrees of sedentism, larger investments in technology, more complex division of labor, and more complex social systems. Where bulk storage is based upon resources that are highly clumped in their spatial distribution, there also tend to be marked increases in local group sizes compared to those commonly associated with mobile, non-storing hunter gatherers. The increased sedentism necessitated by food storage, in turn, favors mobility strategies in which task groups radiate out and back from a permanent or seasonally permanent central place. In this way, mobility becomes more logistically organized in proportion to dependence upon delayed consumption. Therefore, a direct connection exists between increases in delayed consumption and increases in the amount of energy that flows along spatial pathways through a cultural system.

Up to this point, attention has been focused upon identifying the critical variables that must be central to any effective model of hunter gatherer land use strategies. In the next section, one particular land use model that has been widely applied in archaeological studies in recent years is examined.

The Forager-Collector Model

The most comprehensive model of hunter-gatherer land use systems that has been advanced to date is the "forager-collector continuum" presented by Binford (1980). In this bipolar model, "foragers" and "collectors" are the terms applied to the two very different storage/mobility strategies. Foragers are characterized by high degrees of residential mobility and minimal dependence upon food storage. In Woodburn's (1980) terminology, these are immediate return systems. Foragers tend to subsist on relatively dispersed resources procured in quantities sufficient to meet daily food requirements. They move their place of residence frequently as resources are exhausted within a day-radius of camp. Collectors, on the other hand, depend extensively upon logistic mobility—the deployment of task groups to and from a central base to obtain specific resources. They also tend to amass quantities of stored food to be consumed during intervals of low environmental productivity, usually the winter in the temperate and higher latitudes. The collector strategy is associated with clumped resource distributions but Binford (1980:17) emphasizes that any condition that inhibit residential mobility will promote increased reliance upon logistic procurement.

The temporal and spatial distribution of resources at higher latitudes tends to favor greater dependence upon logistic procurement; moreover, food storage itself promotes greater reliance upon logistic mobility during the interval of consumption from stores (Schalk 1977, 1981; Binford 1980:16). Among recent hunter gatherers, the structuring of resources is such that foraging systems are most prevalent in equatorial and low latitude settings while the collector strategy is more common in the temperate and higher latitudes (Binford 1980; Kelly 1983). It must be emphasized, however, that this generalization is based upon "modern" or recent hunter-gatherers and is questionable as a generalization that applies to the variability reflected in the archaeological record. In a later chapter it is argued that the early inhabitants of the Olympic Peninsula and the Northwest practiced a foraging type of settlement system.

A key aspect of the forager-collector model is that these two strategies result in the generation of different kinds of archaeological records (Binford 1980). The forager strategy involves the production of two basic site types—the residential bases where both producers and consumers live and locations which are the loci of resource procurement where primary extractive activities occur. The collector strategy involves these two site types as well as field camps, stations, and caches. Field camps are the locations where task groups live while on logistic forays out of a residential base. They are essentially satellite locations positioned to exploit specific resources that are beyond the distance ordinarily travelled to and from on a daily base a residential base. Stations are places used for information gathering (e.g. game locations and movement) and caches are places where resources are temporarily stored for subsequent consumption at another location. In general then, collector systems generate more site types and more intersite variability than foraging systems.

The outstanding strengths of the forager-collector model lie in its simplicity and its capacity to anticipate so much important archaeological patterning:

The emphasis [of the forager-collector model] on mobility and storage as an adaptive mechanism suggests a number of archaeologically observable implications: specific site patterning in geographic space; degree of microstratigraphic integrity of specific site types; long term positioning and land use strategies; approaches to economic zonation; patterns of faunal transport and discard; staging, damage, and discard of lithics; long-term implications for sedentism; implications for population growth and intensification of resource exploitation; long-term potential of given strategies across varying landscapes (Thomas 1983:11).

This capacity for generating a wide variety of expectations about archaeological patterning makes the forager/collector model particularly useful in studies like the present one that encompass large areas and diverse archaeological resources. This model, therefore, underpins a number of discussions and other models developed in other chapters of this report. While the utility of the forager/collector model is clear, it is also important to recognize some of its limitations or weaknesses. These limitations are discussed in a later section but, at this point, there is one particular subject that is not explicitly addressed in Binford's presentation of the forager/collector model (Binford 1980). This is the role of demography as a determinant of land use strategies.

Demography and Land Use Intensification

Most archaeological studies on the Northwest Coast view human population as being a dependent variable that is responsive to either climatically driven changes in resource productivity or technological changes. Proponents of this view for the Northwest typically view population surges as a result of increased productivity of resources following upon changing sea levels (Fladmark 1975; Aikens et al. 1986), or technological innovation such as new fishing techniques or the development of storage techniques (Aikens et al. 1986). The unstated assumption in these approaches is that the cultural systems we know from the historic period represent developmental "climaxes" in the use of resources that were inevitable once the ecosystems achieved their modern forms and once appropriate technology was available. To account for discrepancies between the archaeological record and the ethnographic pattern, the proponents of this view attempt to identify factors that would have inhibited the development of a climax.

This dominant view of demography on the Northwest Coast runs counter to the theories of population as independent variable and prime-mover in culture change. The latter view was shown to be a useful model for explanation of the evolution of agricultural land use strategies in Europe (Boserup 1965) and also has been proposed as an explanation for many of the major trends recognizable in the prehistoric archaeological record around the world (see especially Cohen 1977, 1981). According to this theory, population increase is considered the normal condition of human populations and population control mechanisms are sought to account for those cases where populations are not increasing. Population-induced resource stresses are viewed as major sources of pressure favoring changes in adaptive strategies.

In the "big picture" of cultural evolution, this second theory seems to account for a great deal. At the smaller scale of particular places and times, there is often a "chicken and egg" relationship between population and culture change and the direction of the causal arrow at this level can point in either direction. It may be that demography is the ultimate cause of most major cultural changes but that in the proximate sense, other factors are not only important but can influence population. There are, in other words, feedback relationships in evolution which often make simple linear causality models inappropriate.

In later chapters, it is argued that the land use systems of the early Holocene in the Northwest involved low density, highly mobile populations whose winter subsistence came primarily from hunting large mammals for immediate consumption. This type of land use assumes very rigorous physical demands on women and, perhaps more importantly, implies a diet consisting of little else but lean meat for several months of the year. Such diets are linked to reduced fecundity in women (Frisch 1978; Speth and Spellman 1983). It is postulated that very slow rates of population growth through the first several millennia of the Holocene would have eventually brought about a systemic change in land use. This change is seen as involving the development of winter sedentary land use systems based upon systematic food storage. This shift in land use implies major changes in mobility and diet, both of which are known to be implicated in human fecundity. It is postulated that rates of population growth after the initiation of winter-sedentary, storage-based land use systems would have been significantly higher than for the mobile winter hunting land use systems of the early Holocene. In this example, it is evident that population growth may cause changes in a land use system but those changes result in a positive feedback relationship that results in more rapid rates of population increase, thereby creating new stresses which may require still further cultural changes.

Population growth is viewed as having one inevitable consequence—stress to increase the supply of essential resources. This is the essence of "intensification" and, therefore, it should be clear that demography has a very central place in any theoretical treatment of the evolution of land use strategies.

There are two general means by which resource production can be increased. One response is to expand the productive area of the biosphere from which subsistence is drawn. A second response is to increase levels of production from the same amount of land. Although both of these responses might be thought of as examples of "intensification", this term is most often applied to the process by which production is increased on a given amount of land.

In the context of regional populations experiencing resource stress from population growth it can reasonably be assumed that territorial expansion would generally imply conflict with neighbors. Warfare involving actual appropriation of land occurred on the Northwest Coast (Drucker 1951). It can be argued, however, that there is an even more fundamental means for expanding the area of the biosphere exploited on the Northwest Coast. For populations that formerly relied mainly upon locally produced terrestrial resources, expansion of the resource base to include locally unearned marine resources was just such an expansion. The major marine resource staples in the economies of Northwest natives at the time of contact are species that are energetically produced over vast areas of the Pacific Ocean. Intensified use of these resources effectively increases the territory exploited even though the land area used by humans may have remained unchanged. Expansion of land use strategies to include systematic and intensive use of marine resources during the Holocene was, in an energetic sense, analogous to colonizing a previously uninhabited region.

The second type of response to population/resource stress, development of strategies by which more people can be supported on less land, is also arguably reflected on the Northwest Coast as well. One example here, of course, would be the expansion of the diet to include food resources that offer lower rates of return (e.g. calories) for a given labor investment. Shellfish have been identified as one example of low-return resources in coastal settings such as the Northwest Coast Osborn (1977a, 1977b).

Another example of intensification responses which increase the productivity of a finite land area is the actual modification of ecosystems to enhance the availability of energy that can be exploited by humans. The deliberate use of fire to maintain early seral vegetation is one of the more obvious examples of this type of intensification. Extensive prairie areas are known from the southern Northwest Coast and these areas provided important forage for ungulate resources but were also important for the plant resources that they produced for humans as well: Although prairie areas were of limited size in the study area relative to those in other regions of the Northwest, it is suggested in later chapters that these may have been of particular importance as places where carbohydrate rich plant resources could be obtained.

Some General Expectations of the Forager-Collector Continuum

A fundamental assumption of the preceding discussion is that land use systems are variable in their organization and that these organizational differences will produce different kinds of archaeological patterns. At this point it is appropriate to examine some of the kinds of archaeological patterns that are predictable from the forager-collector model. The effort to identify archaeological correlates of land use strategies is but one of the linkages that must be established between behavioral theory and the archaeological record. The connection of the general theoretical domain with empirical phenomena deductively is a necessary step and the focus of this section, this step is not by itself a sufficient one. The inductive component of archaeological research—mid-range theory (sensu Binford 1977)—must also be considered. This point is expanded upon in a later section of this chapter but at this point the discussion is directed at illustrating the kinds of general patterns about archaeological patterning that can be deduced from the forager/collector model.

The forager-collector model predicts archaeological variability of several kinds. These would include intersite and intrasite assemblage variability, and the distributional structure of archaeological remains on the landscape. It is not possible or necessary to exhaustively consider all of the archaeological patterning that the model predicts here. Expectations of the forager-collector model have been developed elsewhere (see Binford 1980, 1982; Thomas 1983) and the intent here is simply to list a few examples which emphasize the contrasts between these two system types. The discussion is limited to contrasts between residential sites of foragers and collectors.

For foragers who tend to shift residence frequently and forage for food as they move through the environment, technology is generally expected to be relatively simple, portable, multi-purpose or generalized. These characteristics are necessary for food procurement strategies that are brought into play frequently on an opportunistic or encounter basis rather than as highly planned, resource specific procurement tactics. These characteristics of the toolkit also tend to favor the use of raw materials that are widely available and which can be employed for tool manufacture on an opportunistic basis.

In the absence of dependence upon delayed consumption, food inputs to a local group of foragers must be regular but need not be large. This implies a resource harvest technology that is effective at taking individual animals rather than for mass harvest. Similarly, small but regular food inputs rather than large intermittent ones implies the absence of archaeological features associated with bulk processing. Facilities for mass processing (e.g. drying racks) and food storage are, of course, the hallmarks of collector systems although these are by no means easy to identify archaeologically. In a region where below-ground storage is uncommon, the identification of food storage facilities can be difficult if there is not exceptional preservation (e.g. wooden storage boxes, elevated wooden cache structures, etc.).

High mobility and the lack of reliance upon food storage tends to discourage heavy labor investments in many of the facilities that are so important to collectors. Perhaps the most dramatic example here would be housing and it is anticipated that foragers will opt for simple, portable structures in favor of the more costly structures that might be preferred in a collector system. For collectors, on the other hand, the fact that they may reside in a single place long enough to exhaust the supply of fuel in the vicinity of a site may favor more costly but better insulated housing that reduces the demand for fuel.

For both foragers and collectors the residential base or base camp is where men, women, and children live and where the most diverse range of material remains is to be expected. All of the day to-day activities that result in the production of archaeological preservable residues may potentially occur at base camps. These activities would include food processing, eating, sleeping, garbage disposal, recreation, ritual, and manufacture of tools, clothing, shelter. Because forager systems tend to be less internally differentiated socially, it is expected that intrasite structure will tend to be less differentiated as well. Residential sites of collectors, on the other hand, should be internally differentiated to the degree that the division of labor is itself differentiated. For foragers, the brevity of occupation in most cases would also tend to reduce the problem of refuse disposal that more sedentary populations must come to grips with. Because foragers tend to live on the "fat of the land" and practice non-intensive patterns of land use/resource exploitation, it is expected that labor-intensive food-processing features (e.g. large earth ovens) will also be scarce in the archaeological sites these systems produce.

Discussion

The theoretical perspective on hunter gatherer land use strategies outlined in this chapter could be broadly characterized as evolutionary ecology. In this approach, ecosystem structure and demography are viewed as the independent variables that condition cultural variability and that operate as the driving forces in evolution. The dependent variables include the various dimensions of a land use system which articulate the regional human population with the spatio-temporal structuring of resources in that region. These dependent variables include food storage strategy, mobility strategy, technology, resource mix, home range (size and shape), grouping strategy, division of labor, and social organization. In a theoretical sense then, this study builds upon several years of research focused on hunter-gatherer land use systems in the Pacific Northwest (Schalk 1977, 1978, 1980, 1981, 1982, 1983, 1984b, 1987, 1988; Schalk and Cleveland 1983; Schalk and Mierendorf 1983). [1]

A land use perspective is a particularly useful theoretical framework for archaeological research and management where large regions with diverse cultural resources are involved. This perspective makes use of emergent theory of hunter gatherer adaptations and offers a powerful means for generating a wide variety of expectations for the archaeological record. In contrast to site-focused studies, the land use perspective is inevitably associated with a regional view of hunter gatherer adaptations which, in turn, facilitates integration and synthesis at the geographic scale well beyond individual sites. As this perspective has been outlined in this chapter, it provides a sort of "off-the-shelf" research context for even the smallest compliance projects which typically have lacked any such context. This approach differs in its most basic assumptions about the nature of culture and the causes of culture change from the cultural historical perspective that prevails in much of Northwest archaeology.

I have emphasized the bipolar character of organizational variability among recent hunter gatherers and the implication of this for Northwest Coast archaeology is intriguing. If truly intermediate land use systems do not exist and, for a variety of reasons can not exist, then it is unrealistic to expect to find evidence of intermediate strategies in the archaeological record of this region. The fascinating implication of the proposed transformation from foraging to collecting systems for the archaeological record of the Northwest Coast is that it implies an abrupt or saltatory change rather than a gradual one. Stated somewhat differently, there are good reasons why major discontinuities in the archaeological record of the Northwest Coast may be expected without invoking dramatic environmental changes or population replacements.

The forager-collector model provides a wide variety of contrastive expectations for the archaeological patterning created by hunters and gatherers. This model is useful for describing geographic/global variability in hunter-gatherer organization and for predicting in a general way archaeological patterning from a knowledge of resource structure in a particular region. The forager/collector model does not, however, provide the generative principles by which foragers become collectors.

One of the limitations of the forager-collector model as presented by Binford (1980) is that it does not directly consider how demography relates to these two strategies. In fact, by drawing attention to general environmental properties that result in a patterned global distribution of the two strategies among recent hunter gatherers, there is an impression that these two strategies are determined entirely by structural properties of the environment. But the term "resource" is defined by the adaptation in question; what are resources to one system are not necessarily so for a different type of system. This point, of course, is central to most discussions of subsistence intensification. Ecologists also recognize that resource mixes of animal species are not dictated altogether by the environment but depend also on population density of that species and whether or not competition exists (e.g. Pianka 1974:192). Human population density, as it impinges on foraging costs is arguably a key factor conditioning what plant and animal species are included in the resource mix of a particular subsistence strategy.

I have suggested that the forager/collector model as proposed by Binford (1980) does not adequately address the subject of how foraging systems may be transformed into collecting systems. [2] A key variable that must be considered in this context is population. It is postulated here that along a time trajectory of population growth, there is a density threshold beyond which foraging systems can evolve into collector systems. Such transformation, of course, assumes the availability of food resources that can be efficiently harvested and stored in bulk. It is postulated that population growth eventually forced the adoption of intensive food storage of seasonally aggregated resources such as anadromous fish on the Northwest Coast, thereby transforming foraging systems into collecting systems.

Application of the forager-collector model has emphasized its potential as a source of deductions about archaeological patterning. But it is also necessary to consider the inductive component in which meaning is assigned to empirical observations—the so-called "mid-range theory" which deals with how the archaeological record is formed (Binford 1977; Schiffer 1976). Although mid-range theory has received minimal attention in Northwest archaeology (but see Ham 1982) and is not emphasized within the scope of the present study, the critical role of this component of archaeological research is undeniable. If, for example, one is interested in comparing archaeological assemblages from before and after the establishment of winter villages, one must use some conventions for identifying the appearance of winter villages in the archaeological record. A number of archaeologists (see especially Fladmark 1975) have equated the appearance of shellmiddens with the emergence of cultural systems basicly like those of the ethnographic record. In this example, the meaning that is assigned to shellmiddens is that they are the material by-products of a certain kind of system. Clearly, models for the origin of "Northwest Coast Culture" can only be as reliable as such inferential linkages between those models and the archaeological record.

A substantial number of archaeological predictions can be deduced from the forager collector model. But these predictions are environmentally generic—they are not linked to any particular ecosystem. While great generality is obviously one of the theoretical virtues of the forager/collector model, it can also be an obvious limitation if it focuses attention exclusively upon the two idealized system types. The model does not deal specifically with the sizable variations that can occur within systems that are labelled "collectors". Although the protohistoric societies of the Northwest Coast were all "collectors", there was an enormous amount of variation in their subsistence and settlement systems.

In general, the major food resources tend to occur at fewer and more concentrated locations in a northward direction along the coast (Suttles 1968). Associated with these resource distributional gradients are correlated gradients in home range size, mobility strategies, and local group size (Schalk 1978, 1979). The amount of variability in these dimensions of land use systems along the coast was considerable despite the fact that all Northwest Coast societies described ethnographically fall under the "collector" rubric. But what about all of the important and interesting variability that occurs within such broad evolutionary system types as collectors? Concern with this finer-grained variability necessitates an examination of the protohistoric land use systems of the Olympic Peninsula and consideration of the structure of food resources in this region. These are the subjects of the next two chapters.

ENDNOTES

¹The present study could be considered an element in a long-term "Research Programme" as discussed by Osborn 1987.

²The general evolutionary trend throughout broad areas of the earth's temperate zone seems to be for collecting systems to replace foraging system. Under some environmental conditions, however, the reverse process would be expected. Environmental changes leading to the loss of seasonally aggregated storable resources (e.g. landslides blocking salmon migrations) or large scale reductions in regional human populations (e.g. due to contagious disease) are examples of changes which could produce this reversal in the general pattern.