The Journal of Island and Coastal Archaeology
ISSN: 1556-4894 (Print) 1556-1828 (Online) Journal homepage: http://www.tandfonline.com/loi/uica20
Pre-Columbian Settlement Trajectories in Eastern Dominica: Report on Initial Radiocarbon Age Estimates Isaac Shearn To cite this article: Isaac Shearn (2017): Pre-Columbian Settlement Trajectories in Eastern Dominica: Report on Initial Radiocarbon Age Estimates, The Journal of Island and Coastal Archaeology, DOI: 10.1080/15564894.2017.1316798 To link to this article: http://dx.doi.org/10.1080/15564894.2017.1316798
Published online: 27 Apr 2017.
Submit your article to this journal
Article views: 4
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=uica20 Download by: [Isaac Shearn]
Date: 04 May 2017, At: 19:01
The Journal of Island and Coastal Archaeology, 0:1–15, 2017 C Taylor & Francis Group, LLC Copyright ISSN: 1556-4894 print / 1556-1828 online DOI: 10.1080/15564894.2017.1316798
Report Pre-Columbian Settlement Trajectories in Eastern Dominica: Report on Initial Radiocarbon Age Estimates Isaac Shearn Department of Sociology and Anthropology, Community College of Baltimore County, Baltimore, Maryland, USA
ABSTRACT Radiocarbon age estimates from a micro-regional survey of Eastern Dominica provide a chronological framework for Ceramic Age (ca. 400 BC to AD 1492) settlements and population movements in relation to inter-island communities of the Windward Islands. This report covers the design of the survey, the recovery context of the radiocarbon samples, and the implications for the future of research involving Dominica. Although only a small number of radiocarbon age estimates are reported, these are among the first published from systematically excavated pre-Columbian contexts in Dominica. Keywords Caribbean, chronology, settlement pattern
INTRODUCTION
ogy of Dominica, home to nine active volcanoes, dense tropical rainforests, rugged highlands, and shear coastal cliffs, all of which present barriers to archaeologists and other outsiders that have been exploited for protection by Dominica’s Indigenous populations for thousands of years (Honychurch 1995; Trouillot 1988). Fifteen months of archaeological fieldwork were conducted between 2009 and 2013 as part of doctoral dissertation research at the
This report presents results from a microregional survey of pre-Columbian settlements in Dominica, which provide an initial set of radiocarbon age estimates and a chronological framework for the island long underrepresented in Caribbean archaeology (Bright 2011; Evans 1968; Giovas and Fitzpatrick 2014). This underrepresentation is at least in part due to the ecol-
Received 4 November 2016; accepted 4 April 2017. Address correspondence to Isaac Shearn, Department of Sociology and Anthropology, Community College of Baltimore County, Essex Campus, 7201 Rossville Boulevard, Baltimore, MD 21237-3855, USA. E-mail:
[email protected] Color versions of one or more figures in this article are available online at www.tandfonline.com/uica.
1
Isaac Shearn
University of Florida (Shearn 2014). A comprehensive analysis of all recovered materials, including approximately 800 ceramic vessels, formed the baseline for comparison between sites, and micro-regions. These analyses were then evaluated in light of radiocarbon age estimates from six samples recovered from systematically excavated Amerindian contexts. This paper focuses on these age estimates, along with the context of their recovery, and what they suggest about settlement patterns during the Ceramic Age (ca. 400 BC to AD 1492) in Dominica. Regional interactivity and inter-island relationships have become the focus of much archaeological research in the Caribbean with the recognition that the sea functioned more like a highway than a boundary during the Ceramic Age (Cooper and Boothroyd 2011; Hofman et al. 2007, 2010). A corollary to the focus on regionality is the recognition that modern geopolitical boundaries obfuscate any past social, political, or demographic boundaries that may have shaped pre-Columbian settlements. Furthermore, if spheres of interaction were oriented toward the sea, there may have been greater social distance between communities separated by landmasses than those separated by the sea, and differences may reflect differential involvement in regional communities of interaction. To evaluate this in Dominica, three micro-regions spaced along the length of the windward coast were chosen for survey (Figure 1), and comparisons were constructed both within and between micro-regions to assess the degree of interaction among communities in the study area (Shearn 2013, 2014). Following Yaeger and Canuto (2000:10), the micro-region is defined as “an area larger than an individual site but smaller than a settlement region, [which] delimits a mid-level scale of analysis that includes both diverse material remains found within an area and intra-site spaces.” The micro-region is particularly well suited for the Dominican case because preliminary research (Shearn 2010) indicated that small to medium sized sites were distributed throughout watersheds, as opposed to
2
large nucleated settlements, which sometimes characterize Early Ceramic Age (ca. 400 BC–AD 600/800) settlements (Petersen 1996; Versteeg and Schinkel 1992). Carbon samples were excavated from pre-Columbian contexts within each microregion investigated, and six specimens were submitted to Beta Analytic for AMS radiocarbon dating. An analysis of settlement patterns and artifact assemblages was conducted prior to obtaining the age estimates, which were intended to serve as an additional line of evidence to evaluate the likelihood of hypothesized settlement relationships (Shearn 2014). The sequence can be used to evaluate possible settlement trajectories for communities along the windward coast, but moreover, positions Dominica within established chronological frameworks in the Caribbean (Fitzpatrick 2006; Giovas and Fitzpatrick 2014; Petersen et al. 2004), which should promote increased participation for Dominica in regional syntheses. METHODS After preliminary fieldwork in 2009, in which potential survey areas were assessed, a GIS-based predictive model was developed integrating ecological and other geographic data such as the amount of flat land, proximity to stable river systems, and access to sheltered coastal bays (Shearn 2014). This helped to identify habitable micro-regions sharing these structural similarities, which could then be evaluated in light of post-depositional disturbance, modern development, and accessibility. Stark ecological differences between the two coasts were among the first variables considered, and this research was focused on the eastern, windward coast. Many who have worked in the Lesser Antilles have noted a distinction in settlement patterns between the windward and leeward coasts, often attributing this variation to ecological differences (Bradford 2001; Callaghan 2007; Haviser 1997). Variability in climatic conditions between the east and west coast are extremely pronounced,
VOLUME 0 • ISSUE 0 • 2017
Pre-Columbian Settlement Trajectories in Eastern Dominica
Figure 1. Map of Dominica with approximate survey boundaries demarcated. Insets show the Caribbean region and sites within the three micro-regions investigated.
owing in large part to the intensity of the rain shadow created by Dominica’s steep interior mountain range. As prevailing winds bring clouds and moisture to the east coast, the mountains force those clouds upwards where they cool in the atmosphere, causing them to drop their moisture on the eastern slopes. This results in a common scenario in which it is cold and raining on the east coast and in the mountains, while at the same time it is sunny, dry, and hot on the west coast. The published annual rainfall
average for the interior is ∼7,500 mm, on the east coast it is ∼4,000 mm, whereas the average on the west coast is closer to ∼1,250 mm (Drigo 2001), but as Trouillot (1988:29) points out, “The yearly averages conceal the specificity of miniature ecological niches. … Altitude or even mere proximity to the highlands determines actual variation within any small enclave.” This also means that the flora and fauna of the two coasts are quite distinct from one another, as well as from the interior
THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY
3
Isaac Shearn
mountainous areas. It was assumed that this variability would have had a pronounced effect on horticultural groups occupying Dominica in the past, as it continues to today. By only comparing micro-regions from one coast, it was assumed that variability among micro regions was more likely to be attributable to social, demographic, or chronological factors. The integration of data from future research on the west coast will help to evaluate this relationship. The terrain and coastal dynamics further differentiate the two coasts, with the Caribbean (leeward) coast given to gently sloping hills and large, flat headlands, and the Atlantic (windward) coast featuring more rugged, dynamic terrain and rapid elevation changes (Figure 2). The variability impacts the availability of coastal and offshore resources, and the scarcity of protected bays featuring calm waters and coastal access along the windward coast may have made certain micro-regions more desirable as settlement areas. These would likely have acted as gathering locations for canoe-based interactions, and were thus considered high probability micro-regions for investigation. Visibility was another criterion considered in the development of the study area. Following the theory that visibility between islands could play a large role in social relations between islands (Cooper 2010; Hofman et al. 2007; Torres and Rodríguez Ramos 2008), the selected northern microregion features views of Guadeloupe and Marie-Galante, while the southern microregion selected has a view of Martinique. Guided by the predictive model and the assessment of several high probability areas during preliminary field seasons, the decision was made to focus research efforts on three micro-regions spaced evenly along the east coast: Delices, Hampstead, and Castle Bruce. Within each micro-region, locales in a variety of positions relative to the coast and the floodplain were prioritized for investigation, utilizing interviews, surface inspections, and 15–25 cm diameter shovelexcavated auger tests to establish the presence of cultural materials before proceed-
4
ing to systematic subsurface sampling. This included the excavation of 50 × 50-cm test pits and 1 × 1-m test units in arbitrary 10cm levels within natural strata until reaching 10–20 cm of sterile soil below cultural deposits. All soil was screened with ¼-inch mesh. Test pits were arrayed along gridded transects at 10-, 20- or 40-m intervals depending on the size of the landform being tested. Sites were initially defined as contiguous landforms yielding cultural remains, bounded either by geography or the absence of archaeological materials. In later analyses, some of these sites were argued to truly represent loci of larger discontinuous settlements (Shearn 2014).
DESCRIPTION OF RECOVERY CONTEXTS AND RESULTS Delices (Southern Micro-Region) Delices, located in the southeastern portion of the island, is a wide and highly fertile valley that houses a substantial farming community, with toloma, cassava, and bay leaf being the most common crops. Delices is surrounded by exceptionally steep mountains to the south that separate it from Petite Savanne and the Geneva/Grand Bay enclave. To the north, a series of ridges and narrow valleys separate it from the flatter headlands of La Plaine. The landscape is defined by two rivers that run parallel along the edges of the Delices valley, with approximately 100 hectares of habitable land between. The White River descends from Boiling Lake, high in the mountains, and gets its name from the cloudy white color of the water that derives from the high sulfur content. The other river to the north is a smaller, freshwater stream. Both rivers have gouged deep ravines through the valley, leaving a long gentle slope down to the coast in between that today is under fairly intensive cultivation. The coastline is rocky and the seas are rough over most of the cobble beach, although there are protected areas in the northern corner of the bay that have
VOLUME 0 • ISSUE 0 • 2017
Pre-Columbian Settlement Trajectories in Eastern Dominica
Figure 2. Examples of the coastal terrain of Dominica’s Windward coast as viewed from the sea.
served as boat launching areas in the recent past. When approaching Delices by sea, the low and relatively flat valley and beach stand out considerably from the adjacent coastline to the north and south, which is composed of extremely rugged and steep cliffs that rise abruptly from the sea (Figure 2). From almost any highland position in the valley, Martinique to the south is easily visible across the channel. Delices contains many potentially habitable areas in which extensive pedestrian surveys and auger tests were conducted (Shearn 2013, 2014). Because so much of the central part of the valley was under cultivation, it was easy to rapidly identify artifacts in many places, leaving more time to investigate the marginal, difficultto-access areas in the micro-region. Large tracts of land, particularly along the coast, have been impacted by intense flooding exacerbated by modern activities such as quarrying, grading, and road building. However, owing to colonial history and the location of the historic estate, certain optimal settings within the area remain intact or under cultivation whereas the village itself and the majority of development are located out-
side of these optimal settings. A total of 72 test pits and three test units were excavated among the seven sites identified in Delices (Figure 1), but one of them, DEL-6, consisted of only a historic component. Furthermore, two pairs of sites (DEL-1/DEL-2 and DEL-3/DEL-5) likely represent separate loci of single sites. The test unit at DEL-2 revealed relatively intact stratigraphy and dense deposits of artifacts (Figure 3). The second, and especially the fourth strata had darker soils and higher artifact concentrations relative to the rest of the unit. Below the transitional zone at the interface of the fourth and fifth strata, the basal layer became devoid of artifacts. Scattered throughout the top layers of the excavation there was charcoal in loose association with artifacts. However, as the area had been under recent cultivation, charcoal found close to the plow-zone was disregarded. The darker second and fourth strata were interpreted as anthropogenic, with the intervening third stratum representing the intrusion of colluvium or volcanic material given the large boulders in the matrix and the test unit’s position near the base of the slope bounding
THE JOURNAL OF ISLAND AND COASTAL ARCHAEOLOGY
5
6
VOLUME 0 • ISSUE 0 • 2017
Figure 3. Excavation profiles showing the approximate depth from which dated materials were recovered.
Pre-Columbian Settlement Trajectories in Eastern Dominica
DEL-2 to the north. Below the intrusion, in the artifact-dense fourth stratum, concentrations of charcoal were less abundant, and two were collected and submitted for radiocarbon assays. The first (Beta-366739) was obtained 52 cm below ground surface (cmbs), from the artifact-dense fourth stratum, at the same depth and in horizontal association (
