Beyond the Coastalscapes: Preindustrial Social and Political Networks in East Africa

The archaeological community worldwide now readily recognizes the role and significance of interregional interaction in the development and sustenance of urban societies (e.g., Marcus and Sabloff 2008; Sinclair et al. 2010; Trigger 2003). Over the past two decades, we have carried out a systematic, problem-oriented research program on the Kenyan coast and its hinterland in an effort to understand the ecological and cultural milieu that enabled towns and city-states to develop along the East African coast beginning in the late first millennium CE. Archaeological research complemented with historical sources of the eighteenth and nineteenth centuries shows that different scales of analysis can be used to understand the long-term history of the development of urbanism along this Swahili coast: local, regional, and trans-continental frames of reference each show that Swahili communities were part of numerous networks of interactions. An emerging picture of preindustrial urbanism shows that local, regional, and trans-continental interaction spheres tied coastal towns to the hinterland and to wider Indian Ocean commercial and social networks. Not all of the theorized links between the coastal towns and their local and regional trade and interaction partners will be visible archaeologically. We address the still poorly known elements of preindustrial regional networks of alliance and interaction spheres between urban and rural polities and argue that an integrative approach is necessary to understand the context of coastal urban society.     

Radiocarbon Chronometry of Site QJ-280, Quebrada Jaguay,a Terminal Pleistocene to Early Holocene Fishing Site in Southern Peru

Excavations in 1970, 1996, and 1999 at Site QJ-280, Quebrada Jaguay, in southern Peru, yielded enough dateable terrestrial plant material to establish an extensive radiocarbon chronology for the site. QJ-280 is one of oldest well-dated fishing sites in the Americas: it was occupied from the terminal Pleistocene to the mid-Holocene (about 13,000–8,300 calibrated years BP) based on 42 terrestrial radiocarbon dates, encompassing the Jaguay and Machas Phases of the local archaeological sequence. In addition to the terrestrial dates, radiocarbon measurements on valves of two marine surf clam (Mesodesma donacium) individuals from a single, well-dated mid-Holocene Manos Phase archaeological context have provided insight into marine upwelling conditions during the occupation of Quebrada Jaguay. The marine reservoir age varied between 130 and 730 ¹⁴C years during the brief lives of the two clams (up to 5 years each), and varied by up to 530 ¹⁴C years within an individual valve, suggesting strong and variable deep marine upwelling; conditions broadly similar to those that exist in coastal Peru today. These rapid variations in marine radiocarbon age suggest that marine radiocarbon dates from environments with variable upwelling could be skewed by up to hundreds of years.     

New radiocarbon dating of the transition from the Middle to the Upper Paleolithic in Kebara Cave,Israel

The Middle to Upper Paleolithic transition (MP-UP transition) is considered a major technological and cultural threshold, at the time when modern humans spread “out of Africa”, expanded from the Levant into Europe and possibly into central and northern Asia. The dating of this techno-cultural transition has proved to be extremely difficult because it occurred sometime before 40,000 radiocarbon years before present (¹⁴C years BP), which is close to the end of the effective dating range of radiocarbon. Other dating methods such as Thermoluminescence (TL) or Electron Spin Resonance (ESR) are not sufficiently precise to date the recorded archaeological MP-UP transition in the Levant. Here we report a consistent set of stratified radiocarbon ages on freshly excavated charcoal from Kebara Cave, Mt. Carmel (Israel), that span the late Middle Paleolithic (MP) and Early Upper Paleolithic (EUP) This study applied novel strategies to improve sample preparation techniques and data analysis to obtain high-resolution radiocarbon models. From this study it is proposed that the MP-UP transition for this site can be placed immediately after 45,200 ± 700 ¹⁴C years BP and before 43,600 ± 600 ¹⁴C years BP or from 49/48 to 47/46 radiocarbon calibrated years before present (years Cal BP).     

Radiocarbon dating of marine shell samples. The marine radiocarbon reservoir effect of coastal waters off Atlantic Iberia during Late Neolithic and Chalcolithic Periods

Radiocarbon dates on marine shells have not been used as extensively as charcoal or bone dates for the setting up of absolute chronologies because interpreting these dates is complicated by the marine radiocarbon reservoir effect. Nevertheless marine shellfish were used widely at least during the Holocene and their shells are abundant and usually well preserved in archaeological deposits located near shorelines. Consequently prior research concerning the oceanographic conditions and the marine radiocarbon reservoir effect of a particular coastal area is needed in order to set up reliable chronologies for that region. Values of regional marine radiocarbon reservoir effect – ΔR – of coastal waters off Atlantic Iberia, some of its variability along the Holocene and its correlation with the upwelling phenomenon were determined during previous research. More data related to the ΔR values for western Portuguese coastal waters during the Late Neolithic and the Chalcolithic, a time interval badly sampled in previous research, were recently obtained. These values can be compared with ΔR values already determined for the Gulf of Cadiz. Besides the importance of these values for a better knowledge of the palaeoceanography and palaeoclimatology of Atlantic Iberia, from now on archaeologists can set up reliable chronologies for Late Neolithic and Chalcolithic contexts using marine shell samples of that origin for radiocarbon dating.     

Radiocarbon Marine Reservoir Effect in Human Remains from the Kitakogane Site,Hokkaido, Japan

Accelerator mass spectrometry (AMS) was used to measure radiocarbon (¹⁴C) in bones excavated from the Late Jomon shell midden at the Kitakogane site, Hokkaido, Japan. Comparison between ¹⁴C ages of terrestrial and marine mammals from the same site showed systematic age differences which could be attributed to the ¹⁴C marine reservoir effect in the western North Pacific Ocean. Furthermore, this effect was clearly observed in human remains from this site, indicating a heavy dependence on seafood. Dietary habit was estimated from δ¹³C and δ¹⁵N measured in the same human remains. Apparent ¹⁴C ages indicated 79% of protein originated from marine sources.     

Testing the human factor: radiocarbon dating the first peoples of the South Pacific

Archaeologists have long debated the origins and mode of dispersal of the immediate predecessors of all Polynesians and many populations in Island Melanesia. Such debates are inextricably linked to a chronological framework provided, in part, by radiocarbon dates. Human remains have the greatest potential for providing answers to many questions pertinent to these debates. Unfortunately, bone is one of the most complicated materials to date reliably because of bone degradation, sample pre-treatment and diet. This is of particular concern in the Pacific where humidity contributes to the rapid decay of bone protein, and a combination of marine, reef, C₄, C₃ and freshwater foods complicate the interpretation of ¹⁴C determinations. Independent advances in bone pre-treatment, isotope multivariate modelling and radiocarbon calibration techniques provide us, for the first time, with the tools to obtain reliable calibrated ages for Pacific burials. Here we present research that combines these techniques, enabling us to re-evaluate the age of burials from key archaeological sites in the Pacific.     

New Insights into Mesolithic Human Diet in the Mediterranean from Stable Isotope Analysis: The Sites of Campu Stefanu and Torre d'Aquila,Corsica

Mesolithic human remains are rare in the archaeological record of the French Mediterranean. Only the island of Corsica has so far produced relatively well‐preserved burials, and recent archaeological excavations have brought to light new Mesolithic human remains. The site of Campu Stefanu , located in Sollacaro in the southeast of the island, contained a collective burial of seven to eight individuals in a previously unobserved funerary context. A re‐evaluation of collections in regional museums yielded the remains from another Mesolithic individual from the site of Torre d 'Aquila , excavated at Pietracorbara, in the northern part of the island, at the beginning of the 1990s. These two discoveries presented the rare opportunity to obtain new radiocarbon dates and paleodietary insights from this crucial time period using stable isotope analysis (δ¹³C, δ¹⁵N) on collagen. From Campu Stefanu, one individual had sufficient collagen preserved for radiocarbon dating, revealing that it is the oldest Mesolithic human known on the island, dated to 10216–9920 cal. BP. At Torre d'Aquila, radiocarbon dates indicate that the individual belonged to a younger Mesolithic phase than Campu Stefanu, dated to 9903–9596 cal. BP. δ¹³C and δ¹⁵N isotope ratios are similar between the Campu Stefanu and Torre d'Aquila individuals and indicate a diet dominated by the consumption of terrestrial animal protein and a lack of marine resources. These findings are in contrast with the previous results from two other Mesolithic individuals from Corsica from the sites of Araguina Sennola and Monte Leone , for which about 25–30% of the consumed proteins came from a marine diet. The dietary variability recorded in Corsica is consistent with results obtained from Mesolithic human remains of Sicily and the Iberian Peninsula. We can hypothesise, that despite the nomadic lifestyle, the distance to the sea played a major role in Mesolithic food choices in Corsica. Copyright © 2016 John Wiley & Sons, Ltd.     

Implications of a freshwater radiocarbon reservoir correction for the timing of late Holocene settlement of the Elk Hills,Kern County,California

Uncertainties regarding the magnitude of freshwater radiocarbon reservoir effects can introduce random errors into dates on archaeological freshwater carbonates. As a result, many archaeologists avoid dating freshwater shells unless no other datable materials are available. The chronology of prehistoric occupation of the former Naval Petroleum Reserve No. 1 (NPR-1) at Elk Hills, Kern County, has been established with 50 radiocarbon dates on freshwater mussels (Gonidea and Anodonta sp.). Characterization of any freshwater radiocarbon reservoir effect is crucial for the accurate interpretation of inferred settlement and subsistence changes on the Elk Hills. Paired charcoal and freshwater mussels sampled from closely associated contexts were dated to identify a freshwater reservoir effect. Paired Anodonta and Gonidea sp. shells were dated to investigate interspecific differences in fractionation. Results indicate that a 340 ± 20 ¹⁴C yr correction should be applied to conventional ¹⁴C dates on freshwater carbonates in the Buena Vista Basin before calendar calibration. Evidence of interspecific differences is inconclusive. Dates recalibrated with the reservoir correction indicate that widespread occupation of the Elk Hills is correlated with increasing precipitation towards the end of the Medieval Climatic Anomaly and during the Little Ice Age, suggesting that slough resource exploitation may have been driven by regional population pressure rather than drought-related declines in aquatic productivity.     

Species specific marine radiocarbon reservoir effect: a comparison of ΔR values between Patella vulgata (limpet) shell carbonate and Gadus morhua (Atlantic cod) bone collagen

A number of recent studies of the UK coastal environment have assessed the ¹⁴C marine radiocarbon reservoir effect (MRE) via quantification of ΔR values for several periods throughout the Holocene using marine mollusc shells. However, none have employed fish bone as the marine sample, and the importance of being able to use this material as a reliable dating tool is evident when considering the boom in the British fish trade during the first millennium AD, the so-called ‘fish event horizon’, and the corresponding volume of fish remains that appear in the archaeological record from this time. This study compares ΔR values derived using the multiple paired sample approach employing burnt cereal grain (Hordeum sp.) as the terrestrial sample and either fish bone [North Sea cod (Gadus morhua)] or marine shell [limpet (Patella vulgata)] as the marine sample. The results show a general trend of increasing ΔR for the fish bone compared to shell, however, the differences are not statistically significant when the standard error for predicted values is used as the measure of variability in the ΔR values.     

A Simple(R) Model to Predict the Source of Dietary Carbon in Individual Consumers

Quantitative individual human diet reconstruction using isotopic data and a Bayesian approach typically requires the inclusion of several model parameters, such as individual isotopic data, isotopic and macronutrient composition of food groups, diet‐to‐tissue isotopic offsets and dietary routing. In an archaeological context, sparse data may hamper a widespread application of such models. However, simpler models may be proposed to address specific archaeological questions. As a consequence of the intake of marine foods, individuals from the first century ad Roman site of Herculaneum showed well‐defined bone collagen radiocarbon age offsets from the expected terrestrial value. Taking as reference these radiocarbon offsets and using as model input stable isotope data (δ¹³C and δ¹⁵N), the performance of two Bayesian mixing model instances (routed and concentration‐dependent model versus non‐routed and concentration‐independent) was compared to predict the carbon contribution of marine foods to bone collagen. Predictions generated by both models were in good agreement with observed values. The model with higher complexity showed only a slightly better performance in terms of accuracy and precision. This demonstrates that under similar circumstances, a simple Bayesian approach can be applied to quantify the carbon contribution of marine foods to human bone collagen.     

What makes bones shiny? Investigating trampling as a cause of bone abrasion

Taphonomic modifications on animal bones have the potential to provide a wealth of information on the depositional histories of faunal assemblages. However, certain modifications have received little attention and their interpretation remains complex due to their varied or uncertain aetiology. This has hindered progress in approaches to taphonomic research and it remains relatively rare that a comprehensive suite of modifications is recorded during zooarchaeological analysis. Abrasion, defined as a shine or polish on bone, is one such modification, with a plethora of processes having been cited as a potential cause. Relatively little holistic analysis of archaeological specimens has been carried out and consequently the interpretative potential of the modification is yet to be realised. This paper examines the degree to which the process of trampling causes bone abrasion. Trampling causes multiple, sub-parallel, linear striations on bones and has been suggested by some researchers as a cause of abrasion (see Andrews and Cook, Man 20:675–691, 1985; Behrensmeyer et al., Palaeogeogr Palaeocol 63:183–199, 1986; Fiorillo, Univ Wyoming Contrib Geol 26:57–97, 1989; Myers et al., Am Antiquity 45:483–490, 1980; Nielsen, Am Antiquity 56:483–503, 1991; Olsen and Shipman, J Archaeol Sci 15:535–553, 1988). Research presented here involves statistical analysis of a large and diverse faunal dataset from seven British sites. Results from both correlation and logistic regression analysis demonstrate the very close relationship between the two modifications, although this is not the case at every site. These findings strongly suggest that trampling is a major cause of abrasion in a British context. Once the relationship is established at a specific site, the modification can be more reliably used for reconstructing the taphonomic trajectory of an assemblage.     

A canine surrogacy approach to human paleodietary bone chemistry: past development and future directions

When archaeological human remains are absent or otherwise unavailable for bone chemistry-based paleodietary reconstructions, dog remains may provide an appropriate surrogate material for approximating ancient human diet. This “canine surrogacy approach” (CSA) has developed over the past thirty years and is becoming more common in archaeological science literature. A dearth of continued innovation in CSA applications as well as recent criticisms of its feasibility may reflect the absence of a cohesive overview of the approach’s development, its underlying analogical nature, as well as variation and inconsistency in the ways it has been applied. Considering the CSA’s invaluable potential to partially circumvent the destructive analysis of human remains, thereby addressing the increasingly recognized concerns of indigenous groups, such considerations would be timely and germane. Recent research has characterized the role of analogy in CSA applications and devised a framework for making CSA interpretations (Guiry J Archaeol Method Theory 19(3):351–376, 2012a, b). In contrast to, and complementing that work, this paper provides an outline of the CSA’s inception and evolution with particular emphasis on identifying the impetuses for, and trends in, its development. In addition to clarifying the CSA’s origin as well as where and why it is applied today, this review provides an opportunity to identify future directions for productive methodological innovation.     

Paleodiet in northern Chile through the Holocene: extremely heavy δN values in dental calculus suggest a guano-derived signature?

Dental calculus extracted from the teeth of 28 prehistoric human skeletons from coastal and valley archaeological sites in northern Chile that date from the Archaic period (∼2300 BC) to the Late Intermediate period (AD 1476) was analyzed for stable isotope (δ¹⁵N, δ¹³C) compositions. δ¹⁵N compositions were extremely heavy (+17.8 to +33.1‰), but comparable to some studies using conventional human biomaterials (hair, nail, muscle) at other prehistoric sites in northern Chile. There was a negative correlation between δ¹⁵N and δ¹³C for coastal sites, but a positive correlation for valley sites. Results for the valley sites point to a diet that was influenced by marine resources throughout all time periods. The unusually heavy δ¹⁵N values for the coastal sites require a dietary component with a δ¹⁵N composition significantly heavier than that of marine resources. The hyper-aridity at the study area (mean annual rainfall of 0.5–0.6 mm/year) is a likely contributing factor to the production of heavy δ¹⁵N values, but is unlikely to account for the heaviest δ¹⁵N values (>+30‰). One possible explanation for the heaviest δ¹⁵N values is that dietary components were impacted by isotopically-heavy guano, which is abundant in the region. Guano may have been used as a fertilizer during crop cultivation at the onset of the Formative period, continuing through the Historic period after Spanish contact. The indirect impact of guano from abundant bird habitats in the region may have influenced wild foodstuffs harvested throughout the pre-agricultural period. Results provide support for the utility of dental calculus as a proxy for obtaining stable carbon and nitrogen isotope signatures for use in paleodietary studies.     

Focus: sulphur isotope variation in archaeological marine fauna from northern Europe

We undertook combined stable carbon, nitrogen and sulphur isotope analysis of marine and terrestrial faunal remains from two prehistoric coastal shell middens in Denmark. The aim of the investigation was to establish whether sulphur isotopes were useful in discriminating terrestrial and marine consumers in this region. We found that sulphur isotopes do separate marine and terrestrial fauna but, contrary to our expectations, we observed a negative correlation between δ³⁴S and δ¹³C values for marine fauna. We conclude that similar studies over a broader geographical range are needed before sulphur isotopes can be reliably used to study coastal resource exploitation.     

Fisherfolk and farmers: Carbon and nitrogen isotope evidence from Middle Horizon Ancón,Peru

This study investigates human dietary patterns and economic trends at the coastal site of Ancón, Peru during the Andean Middle Horizon (550AD –1000AD ) using stable isotopic data from 32 individuals buried at the site. δ¹³C and δ¹⁵N results from human bone collagen and δ¹³C from human tooth enamel and bone carbonate indicate that inhabitants consumed a mixed diet composed primarily of marine protein and C₄ resources, with only marginal reliance on C₃ foods. Over time, Ancóneros appear to have relied more heavily on C₄ resources, particularly maize, despite the fact that the crop could not have been grown locally. These results are notable given that C₃ rather than C₄ or marine foods dominate the site's archaeological record. These data suggest that Ancón's inhabitants either had access to more fertile land up‐valley where maize could be cultivated successfully or that they engaged in trade relationships with their valley neighbours. A third possibility is that increased maize consumption at Ancón during the Middle Horizon resulted from Wari imperial influence and interregional exchange. Comparisons of δ¹³C values in enamel and bone carbonate from Ancón individuals indicate that δ¹³C_{carb_enamel} values are significantly more positive than δ¹³C_{carb_bone} values. This suggests that the diets of young children were systematically enriched in ¹³C compared to that of adults, perhaps as a result of nursing activity and/or differential dietary practices among various age groups at the site. Copyright © 2009 John Wiley & Sons, Ltd.     

Stable Isotopes,Chronology, and Bayesian Models for the Viking Archaeology of North-East Iceland

This paper reviews the results of a long-term research project that used stable isotope analyses (δ¹³C, δ¹⁵N, δ³⁴S) and Bayesian mixing models to better model the chronology for a presumed Viking Age cemetery at Hofstaðir, near Lake Mývatn in north-east Iceland. δ¹³C and radiocarbon dating indicated that many of the individuals consumed a large amount of marine protein, which results in a marine reservoir effect (MRE), making ages older than expected. In addition to the MRE, geological activity in the region has the potential to introduce massive quantities of radioactive ‘dead’ carbon into the freshwater system, resulting in a very large freshwater reservoir effect (FRE) that can offset radiocarbon ages on the order of a few thousand years. The radiocarbon dates of organisms that derive an unknown proportion of their carbon from both marine and freshwater reservoirs are extremely difficult to ‘correct’, or, more appropriately, model. The research not only highlights the complexities of dealing with multiple reservoirs, but also how important it is to develop models that are temporally and geographically relevant to the site under study. Finally, it shows how this data can be used to inform the development of chronological models for refining the dating for archaeological activity.     

Radiocarbon reservoir effects in human bone collagen from northern Iceland

Human bone collagen from a series of Icelandic human pagan graves was radiocarbon (¹⁴C) dated to aid understanding of early settlement (landnám) chronologies in northern Iceland. These individuals potentially consumed marine protein. The ¹⁴C age of samples containing marine carbon requires a correction for the marine ¹⁴C reservoir effect. The proportion of non-terrestrial sample carbon was quantified via measurement of carbon stable isotopes (δ¹³C) using a simple mixing model, based on δ¹³C measurements of archaeofaunal samples. Non-terrestrial carbon was also quantified in six pig bones from the archaeofaunal dataset. Assuming all non-terrestrial carbon in human and pig bone collagen was marine-derived, calibrated age ranges calculated using a mixed IntCal09/Marine09 calibration curve were consistent with an early settlement date close to landnám, but several samples returned pre-landnám age ranges. Measurements of nitrogen stable isotopes (δ¹⁵N) strongly suggest that many of the human bone collagen samples contain freshwater diet-derived carbon. Icelandic freshwater systems frequently display large freshwater ¹⁴C reservoir effects, of the order of 10,000 ¹⁴C years, and we suggest that the presence of freshwater carbon is responsible for the anomalously early ages within our dataset. In pig samples, the majority of non-terrestrial carbon is freshwater in origin, but in human samples the proportion of freshwater carbon is within the error of the marine component (±10%). This presents a major obstacle to assessing temporal patterns in the ages of human remains from sampled graves, although the majority of grave ages are within the same, broad, calibrated range.     

Lower Pecos and Coahuila peyote: new radiocarbon dates

Peyote, a psychoactive cactus native to the Chihuahuan Desert, has been preserved from excavations at only two archaeological sites: Shumla Caves in the Lower Pecos region of southwest Texas and shelter CM-79 near Cuatro Ciénegas in Coahuila, Mexico. We determined three indistinguishable radiocarbon ages of 5160 ± 45, 5200 ± 35, and 5210 ± 35 ¹⁴C years BP, yielding a mean age of 5195 ± 20 ¹⁴C years BP for the three specimens from Shumla Caves. For one of the Cuatro Ciénegas specimens we obtained the first direct radiocarbon date of 835 ± 35 ¹⁴C years BP. This study demonstrates the use of peyote by inhabitants of the Lower Pecos region of the Chihuahuan Desert about 6000 calendar years ago, and confirms its use by inhabitants of the Cuatro Ciénegas region of the Chihuahuan Desert in Late Prehistoric times. The Shumla Caves' specimens are composed of an aggregate of ground peyote mixed with other plant material, i.e., they appear to be manufactured peyote effigies, and are definitely not intact peyote buttons.     

Bromine in teeth and bone as an indicator of marine diet

Nitrogen isotope ratios in tissues are often used to aid in the reconstruction of the marine component of diets in past populations. Elevated δ¹⁵N values normally found with high trophic level marine resource consumption can, however, also be mimicked by physiological conditions (breastfeeding, pathological states affecting nitrogen balance), climate (aridity), and anthropogenic environments (manuring). This paper presents a pilot study testing whether bromine concentrations can provide a tool for teasing apart possible causes of variation in nitrogen isotope ratios. Blind analyses of bromine in archaeological samples correctly identified individuals derived from coastal vs. inland regions and high vs. low marine consumption time periods. Thus, bromine can provide an important supplement to other isotopic analyses of diet.     

Variability in the marine radiocarbon reservoir effect in Muscat (Sultanate of Oman) during the 4th millennium BC: reflection of taphonomy or environment?

Radiocarbon (¹⁴C) dating of archaeological coastal sites in arid areas is challenging because of the relative rarity of datable terrestrial artefacts. Marine artefacts provide an abundant source of carbon, which is technically easy to date but radiocarbon dates need to be corrected for local variations from the global average marine ¹⁴C reservoir age (ΔR). In archaeological sites, this is usually done by the comparison of the ¹⁴C age measured between coeval pairs of terrestrial and marine samples that have no inbuilt age. Here, we present the results of a study that aims to determine the local ¹⁴C marine reservoir effect (MRE) along the Eastern coast of the Arabic Peninsula during the 4th millennium BC, a period of important cultural, demographic and social changes in this region. Previous estimates of the MRE based on associated shells and charcoal vary by about 400 ¹⁴C yr and this is a serious limitation to the establishment of a precise chronology in the area based on marine material. In order to maximize the chances to obtain contemporaneous terrestrial–marine pairs for a new assessment of the MRE in this region we dated charcoal, shells, fish and turtle bones as well as human bones found in three recently and carefully excavated graves from Ra’s al-Hamra RH5. A large variability was found between charcoals (up to 240 ¹⁴C yr) and marine specimens (up to 785 ¹⁴C yr) present in the same grave. This variability was attributed to inbuilt ages, diagenesis and possibly environmental factors although taphonomic processes cannot be excluded. We discarded the oldest charcoals and used the ¹⁴C ages obtained on human apatite as a cut-off value to remove the marine outliers. Two out of the three graves provided terrestrial and marine assemblages which did not differ statistically and were used to calculate an average MRE of 645 ± 40 ¹⁴C yr (ΔR = 255 ± 55 ¹⁴C yr). This value is consistent with previous estimates based on the ¹⁴C dating of marine sediments off Pakistan and suggests stability of the MRE in the Arabian Sea since the 4^th millennium BC.