The analysis of more than 1,300 flotation samples from thousands of excavated cultural features dating to the San Pedro phase (1220–730 BC) at the Las Capas site in southern Arizona provides evidence of long-term continuity and change in plant cultivation and collection practices in response to environmental disturbances during the Late Archaic period (2100 BC–AD 50). Although preceramic foodways in the region are widely considered to have been stable for roughly 2,500 years following the introduction of maize prior to 2100 BC, analyses of macrobotanical data reveal that moderate-intensity flood events during the Middle San Pedro phase (930–800 BC) preceded the greatest richness and diversity of harvested plants, while reliance on maize was reduced. In contrast, in periods with little environmental disturbance maize was more dominant, with less diversity in other cultivated and foraged plants. Novel cultivation, processing, and foraging practices were initiated in response to disturbance but persisted after floodplain conditions stabilized. It is argued that the reciprocal relationship between disturbance and botanical diversity is integral for understanding the long-term resilience of Late Archaic foodways and that this relationship is best modeled using ecological theories termed the Intermediate Disturbance Hypothesis and the Intermediate Productivity Hypothesis.

Projections of Paleoindian range mobility in the late Pleistocene are typically inferred from straight-line distances between toolstone sources and sites where artifacts of these raw materials have been found. Often, however, these sourcing assessments are not based on geologic analysis, raising the issue of correct source ascription. If sites of similar age can be linked to a toolstone source through geologic study, and direct procurement of toolstone can be inferred, geographic information systems (GIS) modeling of travel routes between the source and those sites can reveal route segments of annual rounds and aspects of landscape use. In the Hudson Valley of eastern New York, Paleoindian peoples exploited Normanskill chert outcrops for toolstone during the late Pleistocene. Here, we combine X-ray fluorescence sourcing results that link Normanskill chert artifacts at Paleoindian sites to the West Athens Hill source outcrop in the Hudson Valley with GIS least cost path analysis to model seasonal pathways of late Pleistocene peoples in northeastern North America.

Archaeologists have long recognized an important relationship between climate change and the trajectory of the Mississippian polity at Cahokia, with twelfth- and thirteenth-century droughts playing a key role in transforming social relationships and the pace of monument construction. This environmental transition may have spurred emigration from Cahokia and surrounding farming communities. This raises the questions: What was the nature of environmental change and cultural transformations on the Mississippian peripheries and where did these Mississippian emigrants go? This paper provides a case study from the Middle Ohio Valley that brings together spatiotemporal patterns in moisture availability between AD 1000 and AD 1300 and new archaeological data from Fort Ancient villages located in southeast Indiana and southwest Ohio that were occupied during this same temporal interval. We suggest that droughts in the American Bottom region pushed Mississippians to less drought-stricken areas such as the Middle Ohio Valley, which experienced concurrent periods of wetness. This pattern builds on a growing body of data suggesting that the movement of individuals and communities played a large role in the process of Mississippianization throughout the midcontinental and southeastern United States.

The best method for quantifying the marine reservoir effect (MRE) using the global IntCal Marine13 calibration curve remains unresolved. Archaeologists frequently quantify uncertainty on MRE values as errors computed from single pairs of marine-terrestrial radiocarbon ages, which we argue significantly overstates their accuracy and precision. Here, we review the assumptions, methods, and applications of estimating MRE via an estimate of the additional regional offset between the marine and terrestrial calibration curves (ΔR) for the Prince Rupert Harbour (PRH) region of British Columbia, Canada. We acknowledge the influence on ΔR of MRE variation as (1) a dynamic oceanographic process, (2) its variable expression in biochemical and geochemical pathways, and (3) compounding errors in sample selection, measurement, and calculation. We examine a large set of marine-terrestrial pairs (n = 63) from PRH to compare a common archaeological practice of estimating uncertainty from means that generate an uncertainty value of ±49 years with a revised, more appropriate estimate of error of ± 230 years. However, we argue that the use of multiple-pair samples estimates the PRH ΔR as 273 ± 38 years for the last 5,000 years. Calculations of error that do not consider these issues may generate inaccurate age estimates with unjustifiable precision.

We report major new insights from recent research at the Powars II Paleoindian red ocher quarry (48PL330). We salvaged more than 7,000 artifacts from Powars II between 2014 and 2016 by screening redeposited sediment from the talus slope below the intact portion of the site. Clovis artifacts dominate the diagnostic artifact assemblage, including 53 Clovis points, 33 preforms, and artifacts associated with a previously unrecognized blade core industry. We report the first radiocarbon dates from the site, determined from dating bone tools, which indicate Cody-aged use (ca. >10,000 cal BP). Further, salvage efforts discovered a previously unknown toolstone source from which many of the Clovis artifacts were produced. The Powars II Clovis points most resemble early Paleoindian points from the far Northern Plains and were likely both produced and discarded in the red ocher quarry after hunting, as evidenced by preform production and the presence of impact fractures on many used points. Given these production and discard patterns, Powars II holds some of the best evidence archaeologists currently have for Paleoindian ritualism related to hunting.

Although it is now commonplace for archaeologists to study use-alteration patterns on ceramics, the same cannot be said of one of the most ubiquitous classes of hunter-gatherer artifacts, fire-cracked rocks (FCR). It can be shown, however, that many of the same methods and theories applied to the study of cooking ceramics are also relevant to the investigation of rocks used as heating elements. Because use alteration analyses of FCR are so scarce, I describe a range of attributes with the goal of helping researchers identify use alterations (e.g., sooting, reddening, various fracturing patterns) on lithic artifacts from sites worldwide and evaluate their potential function in various cultural practices. These attributes are also outlined in order to create a standardized terminology for describing FCR use-alteration patterns. I discuss my analysis of FCR from three Late Archaic sites (Duck Lake, 913, and 914) on Grand Island in Michigan's Upper Peninsula, followed by an interpretation of their cooking contexts, as a case study. The results indicate great intersite variability among FCR characteristics, cooking methods, and cooking facilities (earth oven, stone boiling, and rock griddle). This use alteration analysis can be applied in archaeological contexts worldwide where similar materials are recovered.

Bayesian chronological modeling is used to investigate the chronology for a large-scale human depopulation event during the Mississippian period (AD 1000–1600) known as the Vacant Quarter phenomenon. The Middle Cumberland region (MCR) of Tennessee is within the Vacant Quarter area, and six villages from the final phase of Mississippian activity in the MCR have been subjected to radiocarbon dating. Complete radiocarbon datasets from these sites are presented within an interpretative Bayesian statistical framework. The results provide a unique history of each settlement and demonstrate that Mississippian occupations at each site likely terminated in the mid- to late fifteenth and possibly early sixteenth centuries AD, which is 50 to 100 years later than the most recent estimate for the timing of the Vacant Quarter. Mississippian abandonment in the MCR was relatively quick, likely occurring over less than a century. The exact reasons for abandonment are not entirely clear but appear to be linked to climate change. A radiocarbon simulation experiment indicates that future robust radiocarbon dating with well-selected samples could greatly improve the chronological precision for this late Mississippian activity. More broadly, this example demonstrates that model building with radiocarbon simulations can be used to address regional-scale chronological issues within the American Southeast and beyond.

Turquoise has played an important role in the Southwest, both today and in the distant past. Increasingly, archaeologists are coming to appreciate that the mineral was likely valued for its symbolism, rather than its chemical properties or economic worth. Thus, the color blue-green and a variety of blue-green things may have been conceptually analogous, together referencing and petitioning moisture. J. J. Brody recognized that additional symbols, while not themselves blue-green, may have likewise belonged to this blue-green complex. Over a decade ago, and while testing Brody's hypothesis, Stephen Plog convincingly argued that black-on-white hachure in Gallup-Dogoszhi pottery served as a proxy for blue-green. Here, we ask whether Mimbres artists incorporated the same symbolism. Findings suggest that Mimbres hachure was likely representative of color but not necessarily blue-green. In fact, it may have referenced yellow. Yellow and blue are often paired among the Pueblos, and interregional differences in the meaning of hachure may relate to interregional complementarity.

We present a revised chronology for the Kolomoki site (9ER1) in Georgia, occupied primarily during the Middle and Late Woodland periods (ca. 200 BC to AD 1050). The considerable extent of the site has been noted for more than a century but came into sharper focus with the archaeological investigations by Sears (1956) and Pluckhahn (2003). The site includes at least nine mounds, a large central plaza, and a discontinuous habitation area nearly a kilometer in diameter. Previous interpretations assumed gradual and incremental changes in the community plan. We present a greatly revised chronology, based on new investigations in some of the lesser-known portions of the site and a doubling of the number of absolute dates. Bayesian modeling of these and previous dates reveals that, far from the gradualist assumption of previous work, the community at Kolomoki was dynamically transformed several times in its history, reaching its greatest spatial extent and formal complexity in two relatively short-lived phases. In these intervals, the village incorporated permanent residents and visitors into a single community in which daily face-to-face interactions were minimized even as communal identity was celebrated.

This article presents the results of a program of radiocarbon dating and Bayesian modeling from the precontact Yup'ik site of Nunalleq (GDN-248) in subarctic southwestern Alaska. Nunalleq is deeply stratified, presenting a robust relative chronological framework of well-defined individual house floors abundant in ecofacts suitable for radiocarbon dating. Capitalizing on this potential, we present the results of one of the first applications of Bayesian statistical modeling of radiocarbon data from an archaeological site in the North American Arctic. Using these methods, we demonstrate that it is possible to generate robust, high-resolution chronological models from Arctic archaeology. Radiocarbon dates, procured prior to the program of dating and modeling presented here, suggested an approximately three-century duration of occupation at the site. The results of Bayesian modeling nuance this interpretation. While it is possible that there may have been activity for almost three centuries (beginning in the late fourteenth century), occupation of the dwelling complex, which dominates the site, was more likely to have endured for no more than a century. The results presented here suggest that the occupation of Nunalleq likely encompassed three generations beginning cal AD 1570–1630 before being curtailed by conflict around cal AD 1645–1675.

Short-term hunter-gatherer residential camps have been a central feature of human settlement patterns and social structure for most of human evolutionary history. Recent analyses of ethnohistoric hunter-gatherer data show that across different environments, the average size of hunter-gatherer bands is remarkably constant and that bands are commonly formed by a small number of coresident families. Using ethnoarchaeological data, we examine the relationship between the physical infrastructure of camps and their social organization. We compiled a dataset of 263 ethnoarchaeologically observed hunter-gatherer camps from 13 studies in the literature. We focus on both the scale of camps, or their average size, structure, and composition, and the dynamics that governed their variation. Using a combination of inferential statistics and linear models, we show that the physical infrastructure of camps, measured by the number of household features, reflects the internal social organization of hunter-gatherer bands. Using scaling analyses, we then show that the variation among individual camps is related to a predictable set of dynamics between camp area, infrastructure, the number of occupants, and residence time. Moreover, the scale and dynamics that set the statistical variance in camp sizes are similar across different environments and have important implications for reconstructing prehistoric hunter-gatherer social organization and behavior from the archaeological record.

It is notable how little gender archaeology has been written for the European Neolithic, in contrast to the following Bronze Age. We cannot blame this absence on a lack of empirical data or on archaeologists’ theoretical naïveté. Instead, we argue that this absence reflects the fact that gender in this period was qualitatively different in form from the types of gender that emerged in Europe from about 3000 cal BC onwards; the latter still form the norm in European and American contexts today, and our standard theories and methodologies are designed to uncover this specific form of gender. In Bronze Age gender systems, gender was mostly binary, associated with stable, lifelong identities expressed in recurrent complexes of gendered symbolism. In contrast, Neolithic gender appears to have been less firmly associated with personal identity and more contextually relevant; it slips easily through our methodological nets. In proposing this “contextual gender” model for Neolithic gender, we both open up new understandings of gender in the past and present and pose significant questions for our models of gender more widely.

There is no recorded maize (Zea mays spp. mays) from sites predating circa cal AD 800 in the northern Lake Michigan or Lake Superior basins of the western Great Lakes, despite the presence of maize microbotanicals including phytoliths and starches in Michigan, New York, and Quebec as early as 400 cal BC. To evaluate the potential for an earlier maize presence in the northern Lake Michigan basin, samples of carbonized food residues adhering to 16 ceramic vessels were obtained from the Winter site (20DE17) located on the Garden Peninsula in the northern Lake Michigan basin. Each sample was split and sent to two analysts. Both analysts identified low incidences of maize starch and phytoliths in multiple samples, with overlapping identifications on several. Three direct accelerator mass spectrometry dates on the carbonized residues reveal maize incorporated into the residues as early as the second century cal BC, 800 years before any regional macrobotanical evidence. Although the method of dispersal cannot be determined, these results support the proposition that initial northern dispersal of maize in the region may have been nearly 800 years earlier than macrobotanical evidence would suggest and is consistent with the timing of its introduction to the lower Great Lakes area.

Precolumbian archaeologists traditionally focus on periods of stability rather than change when constructing regional cultural chronologies. However, the advent of large databases of radiocarbon dates and the proliferation of open-source software environments such as R now allow archaeologists to understand technological transitions with greater chronological precision than has been historically possible. In this study, we employ Monte Carlo procedures, Bayes’ Theorem, the R package Bchron, and IntCal13 to address three chronological topics. We calculate the minimum number of dates required on atlatl and bow technologies to robustly date this late Holocene transition in Subarctic northwestern Canada, analyze previously published dates on organic projectile diagnostics to determine whether bows and atlatls overlapped for an observable amount of time, and estimate the years of calendric time that they overlapped. Results indicate that minimum sample sizes of 29 atlatl and 19 bow dates are required to characterize this particular transition in our study area. Actual radiocarbon dates show that bow and atlatl technologies temporally co-occurred within this region for 174 ± 135 (1σ) actual calendar years. Quantitative analyses such as these open the door to testing hypotheses that explain why and how technological transitions occur within and between (pre)historic groups.

Midpassages are the most recognizable architectural feature associated with the entire Paleoeskimo period (2800 BC–AD 1300) in the Canadian and Greenlandic Arctic. Usually built of stone, midpassages are rectangular-shaped axial structures that run through the center of tent rings and semisubterranean house depressions. However, a unique triangular form of midpassage developed in association with the Late Dorset complex in the Boothia Peninsula area of the Canadian Arctic around AD 400. Unlike the rectangular-shaped varieties that were built contemporaneously across the Arctic, distribution of the triangular form is limited, and occurrences are rare. Initially, construction appears restricted to the Boothia Peninsula region, where the form persisted for the subsequent 400 years. After AD 900, they are found in Inglefield Land, Greenland, and a few other locations in the Canadian Arctic where they continued to be built until around AD 1200, representing an over 800-year period in a remarkably unvarying configuration. Further, these triangular midpassage structures very likely represent a regional architectural variant that moved northward with the Late Dorset diaspora after AD 800.

It has long been assumed that Folsom points are more standardized than Clovis points, although an adequate test of this proposition has yet to be undertaken. Here, we address that deficiency by using data from a sample of Folsom and Clovis points recovered from sites across the western United States. We used geometric morphometric techniques to capture point shape and then conducted statistical analyses of variability associated with Clovis and Folsom point bases and blades. Our results demonstrate that Folsom bases and blades are less variable than those on earlier Clovis points, indicating an increase in point standardization during the Early Paleoindian period. In addition, despite published claims to the contrary, Clovis and Folsom point bases are no more variable than blades. Based on these results, we conducted additional analyses to examine the modularity and size of Clovis and Folsom points. The results suggest Clovis points have more integrated base and blade segments than Folsom points. We suggest that several classes of Clovis points—intended for different functions—might have been in use during the Clovis period and that the later Folsom points might have served only as weapon tips, the shape of which were constrained by the fluting process.

Swan Point in central Alaska contains the oldest recognized human occupation in Alaska (Cultural Zone 4b [CZ4b]), dating to circa 14,000 cal BP. This component consists of a microblade and burin industry with clear technological ties to the Siberian Upper Paleolithic Diuktai Culture. Through the systematic use of the Yubetsu method for the production of microblades, Swan Point is technologically more similar to Siberian microblade sites than to later-age (Denali complex) microblade sites in Alaska. The Yubetsu method is the hallmark of the Diuktai Culture, and in Alaska, Swan Point CZ4b is the only component with systematic production of microblades using the Yubetsu method. Other late Pleistocene and Holocene microblade sites in Alaska have an industry based on Campus-style, conical, or tabular microblade cores. Analysis of the collection furthers our understanding of how CZ4b relates to previous Siberian Diuktai-related assemblages and to later Alaskan Denali-related sites. We interpret the CZ4b component as representing a brief single event that has major cultural and technological implications for the early colonization process of North America.

Modern datasets provide the context necessary for accurate interpretations of isotopic data from archaeological faunal assemblages. In this study, we use the oxygen isotope ratios (δ18O) of modern small mammals from Chaco Canyon, New Mexico, to quantify expected isotopic variation in a local population. The δ18O values of local, modern small mammals encompass a broad range (−6.0‰ to 4.8‰ VPDB), which is expected given the extreme seasonal variation in the δ18O of precipitation on the Colorado Plateau (−11‰ to −3‰ VPDB). Isotopic ratios of small mammals obtained from excavated archaeological sites in Chaco Canyon (ca. AD 800 to 1200) show no significant differences with their modern counterparts, suggesting that there is no difference in the origins of the archaeological small-mammal collection and the modern, local Chaco Canyon small-mammal collection. In contrast, δ18O values of large mammals from Chaco archaeological sites are significantly different from those of modern specimens, reflecting a nonlocal, but also nonspecific, source in the past.

Recent examinations of more than 13,000 disk beads from mortuary contexts determined that macroscopic examination was not always enough to distinguish shell, stone, and fired-clay beads. Using replication experiments and scanning electron microscopy energy dispersive spectrometry (SEM-EDS), we update the 80-year-old conclusions of Emil Haury, who defined features distinctive to bead manufacture. With this renewed confidence in materials identification, we analyzed the distributions of disk beads made from shell, stone, and fired clay among Hohokam inhumations and cremations at the Yuma Wash, Honey Bee Village, and Wetlands sites in the Tucson Basin. Not everyone was buried with disk beads, but all age groups were represented among those who were buried with beads. Some people were buried with only stone, or only shell, or only fired-clay beads, although more were buried with beads of some combination of these three materials. In this article, we consider why fired-clay beads were added to the mix and conclude that they were made as acceptable substitutes for stone beads, not for deceptive reasons concerning wealth or status, but rather in imitation of stone to honor a tradition that could not otherwise be efficiently met.