«By Nathan B. Goodale A dissertation submitted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY WASHINGTON STATE ...»
curves actually increase (the operator adds balloons slightly faster than you remove them, but eventually your removal rate catches up with his adding rate, because the density of balloons on the board increases). One of the curves stays level, at the value a, because the addition rate happens to just balance the removal rate for that curve.
k Figure 7.22.
Balloon populations vs. time, with a cheating operator. Only the lowermost curve has no interference from the operator; all of the other curves have the operator adding balloons at some constant rate (in an archaeological sense the cheating operator would act as a nondestructive force).
171 Figure 7.23. (a) the population growth rates presented earlier scaled to compare to the sneaky operator simulation. Note that the archaeological data take the form most similar to the second line from the top indicating a fast sneaky operator or in our case, slow taphonomic destruction. This line also takes the form most similar to this line at the end of the sequence where there is minimal logistical growth. (b) Balloon populations vs. time, with a cheating operator, as simulated on the computer. The board started with 216 balloons. The lowermost curve has a balloon added every 100 throws; the uppermost curve has a balloon added every throw.
One can also simulate this situation. For every time step, the dart hit the board in a random spot, similar to a site being randomly destroyed by some event (the board
was removed. The sneaky operator would add a dart every 100, 50, or 10, throws. The total number of balloons was tracked for several different addition rates; the result is shown in Figure 7.23. In each case, the curves level out to the value predicted in the
In this case, the parallels with taphonomy in archaeology should be apparent, except that the cheating operator in the balloon game are our friends in prehistory; as people add sites, natural processes destroy sites. The trick here is if sites are getting added at a faster rate than they are getting destroyed, growth rates should produce very noisy plots, such as that in Figure 7.16 for our estimates of growth rates in the southern Levant. Not only should the plots be very noisy, they should also have very little logistic growth towards the end of the sequence, also similar to Figure 7.16.
Figure 7.23 contains only random taphonomic destruction (which works at a well-defined rate) and an addition to the population (which also works at a welldefined and very constant rate), yet they compete to give pretty jagged curves.
Most likely, the conclusion that we can draw from this simulation is that the addition and destruction rates going on in the archaeological data are much less controlled than here, and so one would expect even spikier curves if taphonomic bias was not a significant influencing factor. Moreover, the results of population growth during the interval of 22,000-8650 years ago are very jagged presented in Figure 7.23a. These are comparable to the simulated effects of one balloon (archaeological site) added at a high rate, similar to a fast rather than a slow sneaky operator. In other words, the data
even though a jagged signal suggests that destruction rates should be low. This simulation and observation lends additional support that the population growth model reflects actual population trends through the transition to agriculture and while not completely free of taphonomic bias, we can at least suggest that it is an insignificant factor.
Summary This chapter has accomplished several tasks. First I have argued that the variables utilized here as population proxies do not appear to be largely affected by tamphonomic bias. This was first accomplished by comparing the exponential decay curves of those variables that are the most prone to taphonomic bias to those variables that should not be impacted by taphonomic bias. Second, a simulation was conducted to examine the destruction of sites through time in relation to how fast they are added to the record. Both of these exercises suggest that the data used here as population proxies are not significantly impacted by taphonomic bias.
Through cross-correlation analysis I have argued that the data are all tracking and monitoring the same phenomenon (population growth) in a similar manner.
Additionally, all of the datasets have a statistically significant positive correlation with time. Because of the results of both the cross-correlation and regression analyses, I constructed a population growth rate model with all of the variables to examine population growth rates from 22,000 to 8,650 cal BP in the Near East. This
number of interesting facets of the transition including: the relatively few but rapid population growth rate increase and decrease in under 50 years; the seemingly little effect that technological inventions had on increasing fertility until storage was adopted; the first substantial population growth rate increase appears to co-occur with the first signs of intensive food storage in the PPNA; the subsequent population growth rates co-occurring with the addition of other resources bases; and finally, detecting the potential mass migration of people during the LPPNB.
CONVERGENCE IN THE NEOLITHIC: HUMAN POPULATION GROWTH
AT THE DAWN OF AGRICULTUREIt is widely held that the origins of agriculture incorporated a significant demographic transition in the form of a substantial increase in human numbers (Bocquet-Appel 2002). The results of this study support this argument. I believe, however, that the question that is necessary to ask is why did population growth rates mimic zero-growth throughout most of the southern Levantine prehistory, only to increase at a specific point in time? By posing this question we can begin to understand the potential factors that combined to produce this change when it happened, and not before. Further, by understanding the influencing factors, specifically the diachronic relationship between zero-growth rates for much of human history punctuated by periods of exponential growth, will aid in our future endeavors to model human population history, ultimately leading to a better appreciation of significant transitions in our evolutionary past.
Evolution and the NDT The case study presented here highlights the flexibility of human nature to negotiate the constraints of resources (or lack thereof) and the ability of humans to shift food economies and adjust not only to bearing and investing in more offspring, but also the potential to shift resource focus to enhance fertility. However, this shift does not occur overnight; instead it was part of a long and gradual process. The
interact with their social and natural environments.
Natural selection has favored a human phenotype that is behaviorally and cognitively flexible (Flinn 1996) where we are aware of strategies that produce diminishing marginal returns on investment (Kaplan and Lancaster 2000). As a result of these propensities, humans can alternate strategies toward specific goals as social and environmental circumstances fluctuate (Kaplan and Lancaster 2000). The cost– benefit structure of engaging in any economic activity is shaped by the requirements of involvement and competitiveness in a particular context (Kaplan and Lancaster 2000). This structure helps negotiate whether an individual engages in the production of a food resource or spends time and energy in other arenas that may have greater effects on fitness. Linked to this is the availability of resources in the environment, the quality of the resources available, and the number of other individuals already engaged in the enterprise. The balance of these three conditions affects the number of people engaged in harvesting and processing a particular resource into a usable end product. While the traditional focus of Human Behavioral Ecology is on food quantity in relation to some caloric benefit, I have argued that there is a very strong link to quality, specifically in relation to nutrient content.
If competition is intense for high quality and low quantity resources, costs will theoretically be high to engage in the economic activity, which will lead to fewer individuals participating in production and consumption. As a result, population growth should be constrained if other available resources are not associated with
and the resources can have significantly positive impact on fertility, population growth will not be constrained, although this is predicated on the assumption that there are no consistent population regulating practices in place (such as infanticide).
As a result, population should grow and expand across the landscape, producing larger communities and denser populations through time. Linked to this is the integration of additional age brackets into the work force by younger non-dependent offspring (Kramer and Boone 2002), a common characteristic in ethnographic contexts around the world. Since researchers can estimate nutrient content of the foods that were harvested, processed, and eaten during the NDT, we can examine the potential effects of these foods on fertility. While not directly tested in this study, I expect a focus of my future research on this question will directly test this hypothesis.
When resource quality is low, and availability of resources is low, one should expect that the resource will have a low impact on fertility. Poor-quality resources provide little in terms of nutritional value and may be hard to manipulate into their edible form; however, resources of this nature are commonly utilized, for example, sago palms in highland New Guinea (Diamond 1997). Exploitation of a specific resource, of course, is highly dependent upon what other resources are available and if the resource is even targeted.
Strongly linked to resource quality is the duration of available resources, regardless of their nutritional quality in terms of their potential effects on fertility. In other words, if a resource is associated with increased fertility, yet is only consumed
(Chavarro et al. 2008). Thus, under circumstances of inconsistent or seasonally shifting diets, one would expect fertility, and therefore population growth, to be constrained.
Why did Fertility Increase?
Our current understanding of the NDT proposes that there was a substantial increase in human fertility at the start of the transition (Bocquet-Appel 2002). What we lack is a detailed discussion and link to potential explanations as to why there was a significant increase in fertility at this specific point in time. Bocquet-Appel (2002:647) provides one hypothesis that incorporates the change in diet between the Mesolithic and Neolithic in Europe. I have argued here that there is a strong link to this hypothesis; however, there is more to this than just a dietary shift. One would expect that if dietary changes were the sole catalyst for increasing fertility, then populations would have started to grow much earlier, perhaps during the Natufian in the southern Levant as evidenced by a greater reliance on the foods that would ultimately increase fertility millennia after. Yet as the model suggests here, Natufian populations likely remained at the magnitude of those during the Early and Middle Epipaleolithic.
Instead of only a dietary shift influencing increased fertility, I argue the issue is much more complex. In short, increased fertility is likely associated with a stabilized diet of specific foods (Chavarro et al. 2008), something that only occurred
southern Levant for a number of years (at least 23,000 years ago according to Savard et al. 2006), namely cereal grains are strongly linked to increasing fertility (Chavaro et al. 2008).
Why did Population Increase?
While not directly tested by this model other than with providing evidence that populations did increase during the transition to agriculture, I have argued that population increased largely because of the introduction of younger non-dependent offspring into the work force. As frequently documented in the ethnographic record, people practicing cultivation consistently incorporate younger offspring into the labor force than do people practicing hunting and gathering strategies (Kramer and Boone 2002). Kramer and Boone (2002) document that females in groups that cultivate plants provide net subsistence as early as 12 years of age, where hunter-gatherers females produce net subsistence much later, perhaps not until they are 20+ years of age. There is a similar trend in males as well, boys will produce net production in cultivating societies around the age of 17, but in hunter-gatherer societies net production may be as late as 22 (Kramer and Boone 2002).
The link to incorporating younger offspring into the workforce with population growth and cultivation comes down to this simple explanation: just because you can have more children does not mean that parents can or will invest in raising more children. In this case, children were enhancing their parents’
also their younger, dependent siblings, basically underwriting the cost of larger families (Kramer and Boone 2002).
Convergence in the Neolithic Until now we have only been able to speculate why fertility increased at the dawn of agriculture. Bocquet-Appel (2002) suggests that increased fertility is linked to food nutrition, but he does not suggest exactly what that relationship entails. In this study I have argued for a convergence of a number of factors that ultimately laid the foundation for increased fertility. These include the availability of particular foods associated with the fertility diet, human interaction with those resources, with subsequent human actions creating of a series of technological inventions (Figure 8.1) that increased efficiency such as processing and harvesting tools. Although increasing efficiency, tools may help provide more food at certain times of the year;