Correcting temporal frequency distributions for taphonomic bias

In this paper, we expand upon a prior study [Surovell, T.A., Brantingham, P.J., 2007. A note on the use of temporal frequency distributions in studies of prehistoric demography. Journal of Archaeological Science 34, 1868–1877.] that explored the problem of taphonomic bias. Taphonomic bias refers to the tendency for younger things to be over-represented relative to older things in the archaeological record due to the operation of destructive processes like erosion and weathering. Using a database of radiocarbon dated volcanic deposits from Bryson, R.U., Bryson, R.A., Ruter, A. [2006. A calibrated radiocarbon database of late Quaternary volcanic eruptions. Earth Discussions 1, 123–124.], we develop an empirical model of taphonomic bias. In contrast to our prior study in which we modeled taphonomic bias as an exponential function wherein the likelihood of site loss remains constant through time, we argue that the probability of site destruction actually decreases with site age. We further demonstrate how this model can be used to correct temporal frequency distributions and extract demographic histories. We illustrate this approach using databases of radiocarbon dates from rockshelter and open-air sites in the Bighorn Basin, Wyoming, USA, and mammoths and humans in Siberia, Russia.     

Prehistoric demography of North America inferred from radiocarbon data

Large radiocarbon datasets are increasingly used as a paleodemographic proxy, although potential sources of bias in such records are poorly understood. In this paper, we use more than 25,000 radiocarbon dates extracted from the Canadian Archaeological Radiocarbon Database (CARD) to estimate long-term population trends in North America, while critically examining biases in such records. The frequency distribution of CARD dates shows a positive curvilinear pattern, such that older dates exist in lower numbers than more recent dates, which in part reflects the removal of cultural carbon from the archaeological record through processes such as erosion and dissolution. The average annual growth rate of radiocarbon dates in CARD was calculated and used to derive estimates of the population of North America from the Paleo-Indian to the Contact Periods. While taphonomic bias has likely affected the CARD data, other factors, such as the overrepresentation of Paleo-Indian and Archaic radiocarbon dates, may have offset any bias due to taphonomy. A quantitative reconstruction of Native American population shows that population increased rapidly around 2000 cal yr BP, reaching a maximum of 2,500,000 people by ～AD 1150. From this time until European contact, the population declined, possibly due to the effects of increased sedentism and population density.     

Temporal frequency distributions of alluvium in the American Southwest: taphonomic,paleohydraulic, and demographic implications

The use of radiocarbon frequency distributions to reconstruct prehistoric human and animal populations must account for taphonomic loss and other factors that affect the archaeological and paleontological records. Surovell et al. (JAS, 36, 1715–1724) have recently proposed a correction for “taphonomic bias” that is based on the radiocarbon frequency of a global sample of volcanic deposits. Analysis of 717 radiocarbon dates sampled from the alluvium of the San Pedro and Santa Cruz rivers and their tributaries in southeastern Arizona shows that discovery and scientific biases also play an important role in the creation of radiocarbon frequency distributions, and that the rate of “taphonomic bias” in prehistory is not predicted by the radiocarbon frequency of volcanic deposits. The latter principle is further argued using a sample of 123 Pliocene to Clovis-age proboscideans from the San Pedro Valley. We propose an alternative model that is based on the nature of the stratigraphic record, with discovery bias, scientific bias, taphonomic loss, and the shape of the calibration curve all operating to influence the temporal frequency distribution of radiocarbon-dated prehistoric phenomena.     

ISSUES RELATED TO DETERMINING BURIAL CHRONOLOGY BY FLUORIDE ANALYSIS OF BONE FROM THE MAYA ARCHAEOLOGICAL SITE OF CHAU HIIX,BELIZE*

Fluoride percentage of bone is often used to determine relative dates of burials in archaeological contexts. An analysis of prehistoric Maya burials at Chau Hiix, Belize, identifies taphonomic factors, including soil composition, grave depth, soil disturbance, age at death and cortical bone thickness, that cause variations in the rate of fluoride absorption by altering the access of water to bone. This example demonstrates that fluoride dating is unlikely to be useful at most prehistoric Maya sites because of the inconsistent taphonomy due to typically diverse mortuary practices, and it also serves as a cautionary tale for applications of this methodology in other contexts.     

Estimating parent population of pottery vessels from a sample of fragments: a case study from inter-tidal surface collections,Roviana Lagoon,Solomon Islands

Archaeologists frequently interpret the duration and type of prehistoric occupation at a location/site from the relative abundance of prehistoric materials (accumulations research). Taphonomic factors have the potential to alter substantially the archaeological record leading to inaccurate assumptions about the nature of prehistoric settlement. Where vessel families of potsherds can be established, we provide a general method to estimate the parent population from which the sample of sherds derives, through analysis of vessel completeness of the sample. This provides a basis for archaeological interpretation of the scale of discard represented at a location. The approach used has broad applicability to many archaeological settings where behavioural inferences are made from poorly preserved ceramic samples. Statistical analysis of a ceramic assemblage from Roviana Lagoon stilt villages indicates that most of the ceramic assemblage has been destroyed over time. An unexpectedly severe taphonomic regime is inferred for this relatively sheltered landlocked lagoon setting. Rather than marking an ephemeral occupation, estimating a parent ceramic population indicates instead that early stilt village sites in the Solomon Islands region were probably permanently occupied for several centuries. Results also suggest that the relative absence of recorded Lapita sites in the main Solomon Islands, and possibly elsewhere in Near Oceania may be a result of a harsh taphonomic regime for the remains of stilt village occupation.     

The use of summed radiocarbon probability distributions in archaeology: a review of methods

Using a database of Australian archaeological radiocarbon dates (n = 2996), this paper explores three key methodological issues associated with the use of summed probability plots of radiocarbon data: 1) the minimum sample size needed for a statistically reliable plot; 2) the effect of radiocarbon calibration on the structure of these plots; and 3) the application of a taphonomic correction to such time-series data. The results identify several protocols, recommended as best-practice when using summed probability plots: 1) a minimum sample size of 500 radiocarbon dates should be used, and the sample size and the mean of the standard deviations of the radiocarbon dates (ΔT) in the sample should both be reported; 2) a moving average trendline of 500-800 years should be used to offset the effects of the calibration process; and 3) Surovell et al. “Correcting temporal frequency distributions for taphonomic bias” [Journal of Archaeological Science 36 (2009) 1715-1724] is explored, with modifications and temporal limits (<25,000 cal years BP) proposed. Correction of time-series data using theoretical taphonomic correction curves is useful as a heuristic tool but can obscure real trends if applied routinely. Comparison between summed probability plots and other occupation data is presented and shows good correlation. However it is recommended that plots are supplemented by other archaeological indices and the cross-comparison of such multiple proxies will strengthen identification of occupation trends.     

A SPATIO‐TEMPORAL KERNEL METHOD FOR MAPPING CHANGES IN PREHISTORIC LAND‐USE PATTERNS

Archaeologists are accustomed to considering both the spatial distributions of sites and the temporal distributions of dates as means of analysing the dynamics of prehistoric societies. However, spatial and temporal approaches have thus far remained largely separate, rather than being combined within a single, unified framework. A formal methodology is outlined that combines univariate kernel density estimation based on radiocarbon dates with bivariate kernel density estimation based on spatial site coordinates; the approach allows archaeologists to arrive at reconstructed land‐use distributions through time that not only correct for the problematic issue of site contemporaneity, but also reflect the continuous nature of the archaeological record. The model is implemented using as a data set a series of sites from the Mesolithic of Atlantic Iberia; the results demonstrate that it is capable of providing key insights into changing patterns of land use that are not apparent from either the temporal or the spatial perspective alone.     

The Definition of Spatial Units in Middle Palaeolithic Sites: The Hearth-Related Assemblages

This paper deals with hearth-related accumulations in Middle Palaeolithic sites. Spatial behaviour is a key to understanding the cultural capabilities of ancient hominids. Ethnoarchaeological and archaeological research has shown that hearth-related assemblages are a basic feature in the spatial behaviour of modern and prehistoric hunter–gatherers. In this paper, we propose a methodology for analysing archaeological accumulations and study a series of hearth-related assemblages from the Abric Romanı&#x0301; Middle Palaeolithic site (Capellades, Spain). We also analyse the degree of contemporaneity between the different activity episodes documented at each archaeological accumulation. Our results suggest that the characteristics of domestic areas in this site are not substantially different from those observed in ethnoarchaeological contexts and Upper Palaeolithic sites.     

Spatial and temporal variety of prehistoric human settlement and its influencing factors in the upper Yellow River valley,Qinghai Province,China

We report an investigation of forty-two prehistoric sites in the upper Yellow River valley located in Guide, Jianzha, Hualong, Xunha and Minhe, Qinghai Province, China, including forty-four new radiocarbon dates from twenty-four sites, review published archaeological surveys and radiocarbon dates, analyze landform features and discuss the reasons for spatial and temporal variety of regional prehistoric human settlement. Our results suggest that people occupied the upper Yellow River valley during the early Holocene, but a continuous archaeological record begins only after about 5500 Cal yr BP. Thereafter, the intensity of prehistoric human settlement in the area is varied. Neolithic human settlement spread northwestward to high altitude areas of the upper Yellow River valley during early-mid Majiayao and Qijia periods, but moved southeastward to lower locations in the area during late Majiayao period. During the Bronze period, two coeval archaeological cultures, the Kayue and Xindian, occupied the upper Yellow River Region. Kayue archaeological sites are numerous and widely distributed in the high areas of the upper Yellow River, while Xindian sites are restricted to the lowest basin of the region. Variation in site density and location are likely the result of changing paleoclimate and technology.     

Kana Tamata or Feasts of Men: An Interdisciplinary Approach for Identifying Cannibalism in Prehistoric Fiji

By integrating osteological, taphonomic, archaeological and stable isotopic data, we test for cannibalism in the Lau Group, Fiji and discuss the potential underlying cause(s) and context(s) of this behaviour. First, we compare taphonomic and element representations of human skeletal material from two contexts in Fiji, examining human bone fragments from archaeological sites, including middens and burials in the Lau Island Group. Fourteen sites produced human remains. Only two of those sites included distinct human burial contexts, but in the remaining 12 sites, the human bone was recovered from middens or contexts where midden was mixed with possible secondary burials. A total of 262 number of identified specimens per species, representing an estimated 15 minimum number of individuals make up the Lau human assemblage. Second, we analysed bones contained in 20 individual human burials from four different sites that are housed at the Fiji Museum for comparative purposes. Third, we examine previously published stable isotopic (δ¹³C, δ¹⁵N) analysis of bone collagen to gauge protein consumption of likely cannibalised humans in midden contexts and potential cannibals from primary burials. We model a cannibalistic diet category within the context of isotopically measured Pacific Islands food groups and apply an isotopic mixing model to gauge plausible dietary contributions from six sources including human flesh. Isotopic mixing models of the Lauan samples illustrate a high diversity in reconstructed diets. The percent contribution of human flesh is low for all individual Lauans. We conclude that mortuary rituals evidenced by sharp‐force trauma may suggest non‐nutritive and non‐violent practices that may have included the consumption of small amounts of human flesh. Copyright © 2012 John Wiley & Sons, Ltd.     

Morphometric and taphonomic analysis of granular ark (Anadara granosa) dominated shell deposits of Blue Mud Bay,northern Australia

Metrical analyses of complete mollusc shells have commonly been used to examine aspects of prehistoric mollusc demography, and by extension to investigate potential human exploitation of a variety of species. Recently it has been highlighted that several taphonomic processes, such as fragmentation, have the potential to introduce size bias into these archaeological samples, particularly towards smaller size classes, and thus the potential to significantly skew results of these metrical analyses. This is investigated here by deriving morphometric equations for the sand/mudflat bivalve Anadara granosa, a species prevalent in archaeological sites across northern Australia. A linear regression model based on a series of valve measurements, resulting in high coefficients of determination, demonstrates strong predictive power for size reconstruction. The effects of fragmentation on assemblages from two case study sites are then examined as a means of reviewing issues of differential size preservation, variation in mean valve size, and evaluating the strength of previous interpretations of intensive human exploitation in Blue Mud Bay. Based on these analyses it appears that A. granosa mean size is not significantly affected by fragmentation as other species described in the literature, and previous interpretations retain their validity.     

Radiocarbon dates as data: quantitative strategies for estimating colonization front speeds and event densities

Archaeological analysis of large-scale prehistoric population history requires us to estimate rates of spatial spread during dispersals, and rates and magnitudes of temporal contraction during crashes. Using OxCal’s MCMC sampling routine, I introduce and demonstrate a simple and easily implemented method of estimating front speeds that takes due account of the uncertainty in the archaeological data (in both dates and distances), and argue that this method is more appropriate than those most often used in front speed estimation at present. I also propose a simple and easily implemented method of estimating event densities as a demographic proxy, as an alternative to summed calibrated probability distributions. I argue that this alternative is a significantly better technique, and show that its use also enables us to identify individual archaeological dates that are exerting particularly strong influence on the results, so that we can efficiently allocate our attention when assessing the possible effects of exogenous sampling uncertainty. To illustrate these methods I re-analyse two published datasets relating to the early Paleoindian colonization of North America. My results with the new technique indicate that even with a very noisy dataset, there was clear evidence in the framework of the INTCAL04 calibration curve for a drastic reduction in archaeological event densities following the Younger Dryas onset, followed by a prolonged period of reduced human activity, and a possible renewed phase of rapid growth after the Younger Dryas termination and onset of the Holocene. However, the revised estimate of the Younger Dryas marine reservoir offset in the INTCAL09 calibration curve for ∼12,550–12,900 cal BP changes the picture significantly, by flattening the peak in Clovis-age events and pushing it forward in time into the early Younger Dryas itself.     

Bone chemistry and the reconstruction of diet: Strontium discrimination in white-tailed deer

Studies of the relationship between the strontium content of human bone and past diet are still in an experimental stage. Because of the low frequency of carnivore remains at most archaeological sites and the absence of information on local strontium levels, it has been difficult to (1) estimate dietary intake for a prehistoric population, and (2) to compare prehistoric populations from spatially disparate areas. However, examination of strontium/calcium ratios in a modern herbivore may help to alleviate these problems. Bones from modern white-tailed deer (Odocoileus virginianus) were compared with deer diet in Wisconsin. An observed ratio between deer bone and diet in terms of a strontium/calcium index was calculated at 0204, which compares favourably with observed ratios reported for other mammals. Prehistoric white-tailed deer are used in the analysis of two Late Archaic sites in the midwestern U.S.A. The strontium levels in deer can be used as a baseline for comparison between sites. The prehistoric subsistence patterns are generally comparable, with hunting accounting for the bulk of the diet.     

An Experimental Study of the Anatomical Distribution of Cut Marks Created by Filleting and Disarticulation on Long Bone Ends

Analogical frameworks created through experimentation are a vital part of taphonomic studies for interpreting the archaeological record. Understanding the anatomical location of cut marks is crucial for interpreting the butchery behaviour of humans in the past, as well as for indirectly inferring the subsistence and economic function of archaeological sites. Two experimental/ethnoarchaeological studies have provided taphonomists with analogues to interpret filleting and disarticulation butchery behaviours from archaeofaunal assemblages. However, these analogues were made with limited control and both involved the use of metal knives. The present work provides the first systematic and controlled study of cut mark distribution on long bones made with stone tools, aimed at differentiating cut marks created by filleting or defleshing from those inflicted during disarticulation. It also studies the variability of cut mark distribution according to stone tool type (simple flakes, retouched flakes and handaxes). The results show some differences with previous studies made with metal tools and offer an updated analogue to interpret butchery (filleting, dismembering and skinning) from prehistoric contexts.     

Neotaphonomic measures of carnivore serial predation at Ngamo Pan as an analog for interpreting open-air faunal assemblages

Here I present a neotaphonomic account of natural bone accumulations that have resulted from carnivore serial predation at Ngamo Pan, a vast complex of seasonal water holes located in Hwange National Park, Zimbabwe. Reconstructing the taphonomic histories of faunal assemblages from open-air archaeological sites is burdened with a host of interpretive complications, and this study is offered as a referential means for evaluating instances where stone tools and bone accumulations are associated in apparent archaeological contexts. While the presence of stone tools implies some involvement on the part of humans, open-air sites near water would also have served as prime locations for serial predation by large carnivores to ambush prey—a situation that, over time, can mimic archaeological bone accumulations. The taphonomic and zooarchaeological signatures of carnivore serial predation at Ngamo Pan show marked similarities with the open-air faunal accumulation from Kalkbank, a late Pleistocene site in Limpopo Province, South Africa, located along the margins of a relict pan. Many potential archaeological sites within the interior of southern Africa dating to the Middle Stone Age are known from open-air settings near permanent or ephemeral bodies of water, and the ability to decipher between hominin and non-hominin carnivore involvement with bone accumulations is paramount in determining the hunting and scavenging behaviors of our early ancestors. As much of our understanding of hominin subsistence during the Middle Stone Age is drawn from coastal cave locations, this study is intended to encourage a broadening of our perspective on the taphonomic histories of faunal accumulations dating to the Middle Stone Age by incorporating supplementary evidence provided by these open-air sites.     

Feeding the scavengers. Actualistic taphonomy in the Jordan Valley,Israel

This paper describes the results of a 3‐year project carried out in the Jordan Valley, Israel, in which the influence of specific local environmental conditions, such as scavenging, weathering and scattering of carcasses, were related to the rate of their bone survival. During the 3 year period, the taphonomy of 16 whole animal carcasses, deposited in a nature reserve and agricultural cultivated area were compared. The exact age, date of death and weight of all animals was known. Results of animal scavenging activities on the carcasses, such as disappearance of bones and damage, such as tooth marks, breakage, gnawing to the leftover bones, were recorded at the end of the initial, intensive scavenging period (Phase I), and at later dates. Bone scattering maps were drawn at the end of Phase I and 2.5 and 3 years later. Weathering of surviving bones at the sites is described. Results show that in the Jordan Valley Nature Reserve, scavenging is the most important taphonomic agent causing total disappearance of all bones of young dairy calves, and reducing amount of bones remaining of more mature animals to a minimum. These small number of leftover bones, after only 2–3 years are the only candidates for eventual burial and fossilization. The loss of most of the bones through scavenging in the nature reserve might cause serious bias to conclusions about subsistence behaviour of prehistoric man. This is a taphonomic actualistic analogue to eventual conditions in prehistoric times. Copyright © 2000 John Wiley & Sons, Ltd.     

The Importance of Multiple 14C Dates from Significant Archaeological Contexts

Radiocarbon (¹⁴C) dates are the most important means for determining the age of Holocene archaeological deposits. The theoretical physical basis of this method is by now unassailable, having been consistently tested and refined over two generations. However, the means by which this method has been applied and the interpretation of these results remain as key issues, particularly for complex archaeological discoveries that substantially affect our understanding of world prehistory and social evolution. Many factors can produce uncertainty or variation in the ¹⁴C concentrations of samples, even those that have been selected from the same archaeological context or event. A number of recent studies have also addressed the ways in which ambiguities and irregularities in the ¹⁴C calibration curve can affect the interpretation of archaeological dates and temporal patterns. Of greatest concern, however, is a growing practice of using only one or two samples to date a significant prehistoric context or event. The date of these events, usually relative to other human activities, often holds important theoretical implications for evolutionary anthropology and related disciplines. In this article, we demonstrate that such a practice is rarely adequate or acceptable. Rather, proper procedure requires a suite of dates that permit statistical verification that the deposit or event itself is being correctly dated. We present a detailed case study that highlights the importance of analyzing multiple samples of ¹⁴C from significant archaeological contexts.     

Assessing the distribution of African Palaeolithic sites: a predictive model of collagen degradation

In order to assess the distribution of African Palaeolithic sites and hominin remains, we need an understanding of the factors that may affect the preservation of archaeological material in the fossil record. One of the main factors that is thought to affect seriously the global distribution of fossil remains is taphonomy, but it would seem that taphonomic processes may themselves be overprinted by collagen degradation.The rate of collagen loss is highly temperature dependent, and it has been shown that bone is more likely to be preserved in areas of lower temperature and so lower collagen loss. Hence, it would be expected that fossil fauna remains would be distributed in areas of low collagen loss, i.e. low temperature. Conversely, lithics will not be subject to a temperature bias in their preservation. We have tested this hypothesis through the use of archaeological material.Our results show that the distributions of both lithic and faunal assemblages are non-random across the African continent, with all archaeological sites being found in areas of relatively low collagen degradation. This implies that bone collagen degradation is not the only factor affecting the distribution of African Palaeolithic sites and hominin remains. We suggest that the site distribution is not a reflection of bias in excavation history, but is a real phenomenon reflecting hominin habitat choice.     

RADIOCARBON TRENDS AND THE EAST TEXAS CADDO TRADITION (CA. A.D. 800–1680)

Abstract

Through the employment of radiocarbon (¹⁴C) dates as data, we use the date combination process to refine site-specific summed probability distributions for 555 dates from Caddo sites (n = 19) in East Texas with 10 or more ¹⁴C dates. Summed probability distributions are then contrasted across river basins and natural regions with the remainder of the East Texas Caddo Radiocarbon Database (n = 338 dates from 132 other Caddo sites), highlighting the temporal and spatial character of Caddo archaeological sites throughout East Texas.