Three-Dimensional Mapping of Environmental and Prehistoric Cultural Features
in the Beidha Region
with Spaceborne and Airborne Imagery

---

Introduction

A cultural landscape analysis was begun by the author in 1999 of the region surrounding Beidha, an archaeological site best known for its Pre-Pottery Neolithic B occupation, but which also contains a Natufian component (Kirkbride, 1966; Byrd, 1987). The analysis utilized radar, multispectral, and photographic satellite and aerial remote sensing data and imagery; computerized enhancement and analysis of this imagery; field verification and mapping by use of global positioning system (GPS) and laser transit equipment of observed cultural and natural features; interview of local inhabitants; and a review of pertinent archaeological and environmental literature. Much of the remote sensing data had been collected and analyzed since 1997 in collaboration with Dr. Talal Akasheh of The Hashemite University of Jordan as a part of an ongoing effort to establish and enrich a geographical information system (GIS) database for the Petra World Heritage Site. Al Qantara, a nonprofit research organization affiliated with the University of Palermo, conducted geological and environmental studies and mapping with use of GPS in support of this research. Dr. Paolo Madonia supervised the activities of Al Qantara. The landscape analysis particular to the Beidha region was conducted with the support of an American Center of Oriental Research (ACOR) Near and Middle East Research and Training Program (NMERTP) Post-Doctoral Research Fellowship.

In this article, some of the findings of the landscape analysis of the Biedha area are considered in reference to the chronology of occupation there. The analysis, to begin with, identified factors that attracted human groups repeatedly to the area during the Natufian, and again during the Pre-Pottery Neolithic B (PPNB). There were several, but the one necessary to a combination of factors that produced a threshold of environemental suitability for human occupation was level of precipitation. A drop in precipitation is suggested in this article to be the prime cause for the depopulation of the area during the Pre-Pottery Neolithic A (PPNA). Other findings suggest reasons for the human reoccupation of Beidha during the PPNB as a year round settlement displaying much more complex technologies and social infrastructures than did the previous occupations there. Among the factors that produced this complexity was the capacity for increased agricultural yield associated with the advent of animal husbandry, and the products of animal husbandry itself. It is suggested here that animal husbandry at Beidha enhanced crop yield by providing nitrogen to agricultural soils at crucial times in the agricultural cycle. Animal husbandry was made possible by an elaborated socio-cultural organization, with roots in the PPNA. It was used to control the activities of herd animals, both to provide fertilization and to prevent animals from destroying crops. The cultural landscape of the PPNB at Beidha, as it is described and interpreted in this paper, enhanced the synergy of plant and animal domestication. The cultural system that harnessed this synergy flourished and rapidly became more complex. It utilized the plants and animals that were adapted to the conditions at Beidha during the post Younger Dryas Climatic Optimum. While this analysis focuses on Beidha, some of the factors identified by this analysis regarding cycles of depopulation and repopulation may have general relevance to site distribution patterns as they have been observed to change from the Natufian through the PPNB in the Petra region and elsewhere in the Levant. Before proceeding with this analysis and interpretations of it, a description of the method and theory employed here will be given.

Method

Cultural landscape analysis is a method for identifying natural and cultural features that are relevant to human use of an area, and to increase understanding of the specific ways in which natural and cultural features are codetermined. It considers (Comer, 1999):

· Environmental parameters
· Information retrieved through archaeological excavations and the study of historic sites (and associated documents)
· Traditional use areas
· Current infrastructure (e.g., roads and utilities) and condition of the landscape

This sort of analysis is greatly facilitated by recent innovations in remote sensing and geographical information system (GIS) technology. The analysis depends upon data collected not only through remote sensing, however, but also by archaeological survey and excavation, by examination of current infrastructure to see how this might reflect landscape conditions and be acting to alter landscape use today, and by investigative techniques taken from cultural anthropology. The last of these includes the observation and interview of indigenous populations. Indigenous populations sometimes engage in cultural practices that derive from or are at least similar to those of interest to the archaeologist. This occurs when aspects of the environment and of the cultural practices of interest co-vary in some of the same ways they did during the past.

At Beidha, aerial and satellite remote sensing of several sorts was used to discover a variety of natural and cultural features and zones. These types of remote sensing included:

· SIR-C/X-SAR (space imaging synthetic aperture radar utilizing bands C, L, and X) apparatus carried by the space shuttle Endeavour during a 1994 flight over Petra;

· LANDSAT and SPOT multi-spectral imaging satellites

· Elevation data collected from a digital elevation model (DEM) produced from black and white aerial stereo pairs by Dr. Talal Akasheh, Dean of Research and Graduate Studies at the Hashemite University

· Black and white aerial stereo pair photographs, digitized and incorporated into a GIS for Petra

· Declassified black and white photographs obtained by Russian and American satellites

· Color photographs taken from a tethered balloon.

Each of these technologies produced graphical images of the landscape surrounding the village site of Beidha. Each type of image highlighted or in some cases revealed for the first time certain cultural and landscape features. This work was much facilitated by employing imagery analysis software (ERDAS IMAGINE). In some cases this required digitizing images not already in digital form. Landscape characteristics of particular interest included cultural features, natural features that might have altered by human occupancy of the area, and natural features, some of which were key to certain cultural practices. Images were rectified and coregistered so that they could be superimposed accurately. Each image thereafter could be used to interpret all other images. Placing the images into a geographical information system (GIS) facilitated this interpretation. Each image (coregistered to all others) constituted a theme in the GIS, and could therefore be turned on and off to facilitate interpretation (Fig.1 at right). Further, features identified in the interpretation of images could be digitized as a separate theme, and selectively turned on and off.

Numerous possible cultural features were noted during this procedure. Groundtruthing established that many were, in fact, cultural (e.g., walls, foundations, cisterns). None of these, however, appear to date to Natufian or Pre-Pottery Neolithic (PPN) times (with the exception of those already discovered, as with the excavated site of Beidha), and so fall outside the scope of this paper. Future analysis of higher resolution radar imagery may reveal cultural features associated with the Natufian and PPN periods, such as previously undiscovered below ground habitation areas.

The procedure, nonetheless, highlighted aspects of the landscape that were largely formed by natural process, but were important to the human occupation of the landscape. Such landscape features were both utilized and assigned meanings by human population in ways that were implicated in the establishment of aggregated human populations, the development of domesticated plants and animals, and the development of a permanent, sophisticated architecture associated with a settled way of life.

Environmental Diversity

Environmental diversity has long been recognized as a factor in the development of a settled way of life. MacNiesh, for example, identifies ecological diversity as the first necessary condition for both the development of a settled way of life and of agriculture (1992:12). The ecological diversity of the Beidha region is remarkable. It results primarily from the varied geological zones of the region, which greatly influenced the development of topography, soils, and plant and animal communities there. Fortunately, geological diversity proved to be very clearly differentiated in radar images of the region (see Fig. 2 for a radar image of the region of Petra). Within these basic zones, other data and imagery was used to produce characterizations of relevant environmental and cultural variables. The essential geological zones, which determine and therefore coincide with gross environmental zones, which can be seen in Fig. 2 (at left), are:

· The Cretaceous limestone highlands to the east of the site (seen as the darker, smoother zone occupying approximately the right third of the image),
· The Cambrian to Lower Ordovician sandstone in which Beidha, and the Nabataean ruins of Petra, are located (the broad, extremely rough textured vertical band in the middle of the image), and
· The desert environment of Wadi Araba to the west, where soils are largely alluvium and wadi sediments, fluvatile and lacustrine gravels deposited in Early Pleistocene through Early Holocene times (the band of variegated dark and light, evidently lower in elevation, that borders the image on the left).

Of particular note is the geographical shelf between the limestone highlands and the sandstone formation. This shelf is a geological unconformity, as it is the juncture between two formations that are very different in age. The Ordovician sandstone is about 450 million years old, while the Cretaceous limestone is approximately 100 million years old. The shelf can be seen in the radar image as the bright band between the two. It is key to many of the resources and landscape attributes most important to the nature of the human occupation of the area at various periods. Associated attributes especially important to humans include including springs, extremely high quality lithic materials, a considerable range of elevations, and the canyons in the sandstone formation that greatly enhance ecosystem diversity. These canyons were formed by seismic activity and water flow associated with the unconformity.

A smaller research area is defined by the water catchment area for the Beidha village site (Fig. 3 at right). The catchment area was defined by constructing a hydrological model. This model made use of a digital elevation model (DEM) produced by Dr. Talal Akasheh, Dean of Research and Graduate Studies at Hashemite University. The hydrology of the Beidha region has been of special interest to archaeologists. Because at present the closest source of water is three kilometers distant, the location of the springs has seemed problematic to the placement of the village on the landscape. Archaeological site distributions models constructed for any area of the world would find this situation highly anomalous, as often 80% or 90% or archaeological sites are located within 500 or 600 meters from water sources.

Natufian Attractors

By the end of the Pleistocene, with a warmer and wetter climate, there was much in the immediate vicinity of Beidha to attract human occupation and to sustain it in high densities. Water is, of course, essential to human populations, and the hydrological analysis provides evidence that it could have been present near Beidha under certain conditions. No springs are now active near Beidha, and evidence that springs were ever nearby is inconclusive. The hydrological flow model, however, suggests that water was present in the canyon during times of high precipitation. Precipitation and water flowing from the springs can be seen in the model to be concentrated near the site of Beidha. Precipitation would produce greater run off, and would also make uphill springs more productive.

In the region of Beidha, we can see a dramatic change in elevation to the east of the settlement, a rise of from about 1,000 meters at the site of the settlement to a peak of almost 1,700 meters over a distance of less than five kilometers (Fig. 4 at left). The sandstone canyons in which Beidha was founded, moreover, provided further environmental diversity. The aspect of canyons, their precise configuration, and soils change as one moves away from the limestone highlands deeper into the canyons produce different microenvironments. This unusually great environmental variation increased the chances that one of the econiches in the area would be productive of necessary resources in the face of overall climatic fluctuations. Environmental variation also lengthened the harvest season for cereals, which would ripen at different times in different microenvironments.

Environmental diversity is expanded even more as one descends from the limestone highlands to Wadi Araba. Hans Gebal (1988) has noted four macro ecological zones in a 20 kilometer transect from the highlands to the extremely low elevations in Wadi Araba, which include:

· Swampy environments in the gorges with hydrophytic vegetation
· Dry desert conditions on the "playa" fillings with xerophytic vegetation
· Shady gorges (siqs) with more temperate climate
· Water-storing clefts filled with fine sediments
· Intermittent wadis

The essential framework that produces much of this variation is that observed in Figs. 2 and 3. While many aspects of the landscape have most certainly changed, some perhaps dramatically, we can safely assume that the basic geology has not. Combined with an examination of the geology facilitated by the use of remote sensing, we can infer with a good deal of confidence certain parameters of the paleolandscape, such as soils and topography, when these have been supplemented by on-the-ground observation.

The network of sandstone canyons visible in the radar imagery has provided an attractive habitation area to humans over the millennia. In addition to the environmental diversity mentioned above, the attractions of the sandstone canyons included shelter from the extremes of summer sun and winter wind. Inhabitants have included not only the residents of Beidha, but also numerous other groups. Perhaps the best known of these have been the Nabataeans, who built the core portion of the city of Petra about seven kilometers to the southwest of Beidha, and an annex of tombs in the canyon just north of the one in which Beidha is located, about 750 meters distant. This canyon is called Siq al-Barid, in reference to the cool breezes that are mentioned today by indigenous Bedouin today as occurring in both canyons. Most likely these are created by cooler air that slides down the relatively steep slope along the juxtaposition of the limestone highlands with the sandstone formation (Fig. 5, above right, is a full, three-dimensional view of the landscape). As the air is restricted when it enters east-west canyons, wind speed increases by virtue of the Venturi Effect, which amplifies the cooling effect of the breeze.

To supplement the macro view of the sandstone formation, a more detailed geological examination of the canyon in which Beidha is located was carried out by Al Qantara in November of 1999 and April of 2000. Fig. 6 at left is a map of the canyon. A Bedouin tent can be seen to be located just outside the fence surrounding the excavated site of Beidha (seen also in the photograph at lower right, Fig. 7). Both Beidha and the Bedouin tent are located on a rise, formed in part because the location is one of mechanically sound sandstone. Al Qantara conducted a visual survey of nearby canyons, and noted that the distribution of Bedouin tents coincides to those areas of the greatest mechanical soundness of the sandstone (Madonia, personal communication, 1999). Placing tents in this manner reduces the likelihood of injury to people or damage to property from falling rock, as might occur if tents were placed in areas of unsound sandstone. It also increases the probability that habitations will be out of the path of flash floods, since elevations of the canyon floor are higher in areas where sandstone is sounder.

Interviews with contemporary Bedouin inhabitants of Beidha canyon have revealed several other factors that influence the positions of modern day Bedouin tents. These include, as might be expected, shelter from the sun in the summer and from the wind in the winter. Also, the Bedouin today find it convenient to place tents very near, or within, fields of barley and wheat they are growing, and among flocks of goats they own. Such proximity makes the task of managing both resources much easier. Typically, children are assigned the duty of keeping goats out of fields of growing grain. This takes a good bit of vigilance, but the children enthusiasm to this work, shooing goats away from crops by running at them or throwing pebbles. Failure to keep goats out of a neighbor's field will occasion a social visit by the head of the neighbor's family to the head of the family owning the goats. During the visit, invariably accompanied by tea drinking, the neighbor will mention that while walking in his fields, he has noticed the footprints of goats owned by his host. This observation, made in the course of a friendly conversation, is usually sufficient to correct the problem. If it does not, the owner of the field can take legal recourse, for the owner of the goats can be held liable for damages (Ed-Badbul, personal communication, 1999). Thus, agriculture and animal husbandry today benefits from the close proximity of habitation to fields and flocks and ready access to resources in the canyon, and social convention has become enmeshed with this organization of cultural and natural features on the landscape. A similar arrangement in the past, if not certain, was evidently possible.

It is of special note that goat herding in the canyon would not be feasible, according to Bedouin engaged in this today, if it were not for water that is trucked in to supplement water trapped in a few functional Nabataean water cisterns in or near the canyon. Periodic deliveries of water are made to these cisterns. The water is also used for cooking and washing. Wheat and barley agriculture is at the mercy of precipitation patterns. If a drought period occurs after grains have begun to sprout, goats are allowed to consume the dead plants. Goats are also encouraged to consume plants after the harvest has been gathered.

Herds of wild goats (Capra aegagrus) and the Nubian ibex (Capra ibex) native to the canyon system comprised a resource crucial to the Nabataean population of Beidha. Byrd (1989:81) reported that caprines were the two best represented species in the faunal collections retrieved during archaeological excavations of the Nabataean component of the site. The exact percentages of each species are difficult to determine, because it is very difficult to distinguish between the bones of these species. (Despite these anatomical similarities, Capra aegagrus was eventually domesticated while Capra ibex was not.) The other main large animal species hunted, the gazelle, was from the steppes and desert of Wadi Araba (Byrd 1989: 81).

Cereal grains could be found at the high elevations at which the sandstone and limestone formations surrounding Beidha were located during the Natufian. Donald O. Henry has noted (1989: 232-234), citing Wright's previous work (1977), that wild cereals of the sort which supported the Natufian lifestyle were during the earlier Pleistocene probably to be found only in refuges at lower elevations. At the end of the Ice Age, with warmer temperatures, these cereals could be found at higher elevations like those in which Beidha was located. For reasons that are not entirely clear, it seems that grain size increased during this period. Warmer and drier conditions of the Holocene might have produced a selective advantage for annuals over perennials. He notes also that "annual grasses and legumes did not appear as significant components of the vegetation of Asia until the Neothermal, some 11,000 B.P. [uncalibrated](1977:209)" The larger grain sizes would, of course, have had the additional advantage of rendering seeds more attractive for human collection, and subsequent accidental distribution. For whatever combination of reasons, as Henry notes, there was only limited exploitation of wild cereals in warm environments such as the Nile and Rift valleys during the Late Pleistocene. It was not until roughly 13,000 years ago that evidence has been found of intensive use of wild cereals, and these, notably, were in intermountain and piedmont settings. Worldwide, it is remarkable that the first examples of intensive collection of wheat and barley, and of rice and maize outside the Levant, occurred in upland settings. The importance of the upland settings, Henry suggests (1989:232), might be the fact that seeds matured over a longer period of time, depending upon the exact elevation that seed bearing plants occupied. This is to say that environmental diversity of the sort found in the region of Beidha produced a longer harvesting season.

Brian Byrd, based on the findings of several excavations of the Natufian occupation at Beidha, considers hunting activities to have far outweighed grain-collecting activities. He notes, in fact, very little evidence of processing activities associated with cereal collecting, much less growing. He reports only two grinding artifacts: a broken sandstone pestle fragment and a unifacial discoidal handstone. Other evidence that processing activities were less common than hunting related activities include low frequency of nongeometric microliths and a rarity of sickle blades (1989: 76). The only types of features documented during the excavation of Natufian Beidha were hearths and roasting areas (the latter associated with large animal bones). No features other than these were found: no walls, structures, storage facilities, burials, or stone paving.

In summary, Byrd (1989: 81) considered the Natufian occupation to be:

…best characterized as a short-term or seasonal camp site that was occupied repeatedly over a considerable period of time…It was neither a sedentary settlement nor a highly specialized field camp or station (Binford 1980), but rather witnessed occupation by a diverse group of individuals either once or a number of times for a portion of each year.

He sees this conclusion to fit well within the generally accepted model of the Natufian annual cycle, which is speculated to have centered on a sedentary base camp (Byrd 1989:80). Transitory camps, in the model, are situated around it in a radiating manner (Henry 1985, Bar-Yosef 1983).

The formation of the sandstone canyons began with seismic activity that produced massive cracks that were later widened by the action of flowing water. Seismic activity is consistent with the juxtaposition of the sandstone and limestone geological areas. The sandstone formation is Ordovician (ca. 450 million years old). The faults, and seismic activity, began when the limestone was formed, much later, during the Cretaceous Period (ca. 100 million years ago). The water that widened the canyons came not only from precipitation, but also from springs that occur near the fault line. The springs occur here because water can migrate quite well through limestone, but not as well through sandstone. Water encountering the sandstone would then weep, change direction, pond, or perch.

In addition to creating springs, this hydrological dynamic produced extremely high quality flint formations. Water passing through the sandstone picked up silica, which was deposited where water flow speed changed, that is, at the interface between the limestone and the sandstone. Flint was deposited in large amounts. As the flint formed, it, too, provided an impervious surface, which also promoted the formation of springs where it even more efficiently blocked the path of water percolating through limestone (Paradise, personal communication, 2000).

Field observation and remote sensing imagery provide data that support this model of flint formation. Although not found as a continuous stratum, flint is plentiful at the juncture between the sandstone and limestone formations. As seen in Fig. 8 at left, tabular formations have been noted near the springs above Beidha (Kirkbride, 1989:121). Flint cobbles are common in the dry wadi beds in the canyon in which Beidha is located, having washed down from these formations. Large quantities of flint detritus can be found in ephemeral stream sediment near Umm Seyhoun (where Bdul Bedouin whoa one inhabited Petra have been relocated). An enormous area of flint has been exposed atop a hill in the highlands to the south of Petra, just above the town of Taybet. This area can be seen quite clearly in the radar image of the Petra region. The sources are all associated with the unconformity between the sandstone and the limestone. The flint produced extremely effective hunting and animal processing tools, as well as efficient tools for cutting and shaping a variety of materials.

The Neolithic Landscape

Thus, the juxtaposition of the two geological formations generated two of the resources most necessary to the cultural developments at Petra: water and flint. In the transition to agriculture, the flint was shaped into the microliths that were hafted onto sickles used for grain collection. Production of fine tools of this sort required especially high quality lithic material. Suitable material was surely highly prized, and the flint at the discontinuity would have been a powerful attraction to populations who had already developed the agriculture/animal husbandry package.

It is interesting to note that collecting grains by harvesting with sickles may have hastened domestication of cereals by selecting for grains that were attached to stalks by less brittle rachises. Grain that stayed attached to the stalk during harvesting by sickle or by pulling the plant up by the roots was more likely to be planted, grain that fell off during collection was more likely to be consumed by birds or rodents. Bruce D. Smith summarizes arguments for this in his recent publication about the emergence of agriculture (1998:72-74). He reports on research by Gordan Hillman and M. Stuart Davis, who estimate that only three centuries would have been sufficient to produce the morphological changes in emmer and einkorn wheat and in barley that are associated with domestication. He also notes studies by Mordechai Kislev, who estimates that 20 to 200 years would be enough to make seeds attached with durable rachises dominant among plants in fields harvested with sickles or by pulling up lands by the roots.

The high quality flint at Beidha may, then, be implicated in the domestication of the barley that was native to the area in the Pre-Pottery Neolithic (Kirkbride, 1989: 120). The flint, because it could be effectively and efficiently worked into microliths, encouraged the use of sickles for harvesting. Using sickles for harvesting produced a selective advantage for barley plant characteristics that made agriculture an increasingly attractive and viable way of life. Kirkbride wrote in 1989 that most of the botanical material she recovered was from the earliest level with permanent buildings (Level VI) was barley. Carbonized plant remains and imprints in clay roofs and walls established that Hordeum spontaneum, the wild, hulled 2-row barley, was being cultivated. She could say at that time that, "This cultivation of wild cereals is a stage in the process of domestication that was known of theoretically, but had never yet been demonstrated in the field" (1989: 120).

To the east, the limestone provided a habitat for animals and plants of different sorts than found in the sandstone canyons. It also provided nutrients that were important to the health of grains growing wild, or later cultivated. These nutrients were potassium and phosphorus. (Phosphorus is also produced by human and animal occupation of an area.) The minerals were carried into the sandstone canyon area by flowing water. The presence of ample quantities of potassium and phosphorus throughout the interface are of the sandstone and limestone formations is demonstrated in Figs. 9 and 10 (below), which shows the results of soil tests for these minerals on samples taken at six points within and outside the canyon system.

Potassium and phosphorus supply two of the three nutrients most essential to the health of plants. The third is nitrogen, which is not supplied by geological formations, but is extracted from the atmosphere by organisms, and so placed in the soil by organic materials. It is likely that droppings from goats living in the canyon supplied enriched the soil there with nitrogen. It is reasonable to speculate that higher goat populations and densities associated with animal husbandry would have resulted in even greater nitrogen enrichment. The importance of nitrogen is illustrated by Fig.11 at right, which shows how not only yield, but also protein content in wheat increases as nitrogen is provided to the wheat plants.

The presence of the goats, however, was something of a double-edged sword in regard to grain crops. While goats do fertilize soils in ways that are beneficial to plants, they also devour the plants, a problem that becomes greater as populations and densities increase. Thus to maximize the benefit of the potentially beneficent synergistic relationship between plants and goats, human intervention was needed. Sustained human intervention required that cultural mechanisms be put into place.

Enrichment of Soils by Goat Herds During the PPNB

Byrd notes that analyses of soils conducted at Beidha demonstrated that concentrations of Potassium (P), Nitrogen (N) and organic matter (OM) were much lower in the Natufian than in the PPN strata. He interpreted this, in part, as a reflection of the fact that soil aggradation was occurring during the Natufian, which would have diluted mineral concentrations. Aggradation had ceased prior to the PPN, so that materials associated with human occupation were more concentrated. He also thought it to be another indicator that habitation was more intense during the PPN than the Natufian (Byrd 1989:88). It seems possible, though, that it reflects the high density of the goat population in the area during the PPNB, as well. A high concentration of goats could have increased concentration of these minerals at the habitation sites if herds were kept nearby, as they are today. This would be particularly likely if the habitations were not just nearby, but positioned within some of the fields, as is the case with the current Bedouin occupation of the canyon. The current resulting distribution of nitrogen is demonstrated in Fig. 12, above left .

Explaining the Discontinuity of Occupation at Beidha

The findings presented above are relevant to an explanation of the chronology of occupation at Beidha. The most striking feature of that chronology is the ample evidence of both a Natufian and a PPNB occupation, but no indication that the area was the site of a PPNA settlement.

It is, of course, the case that in the Levant, PPNA sites are relatively few in number, and remarkable for their large size. Both Natufian and Pre-Pottery Neolithic B sites are much more numerous, and occupy a much wider range of environments than do PPNA sites. It is to date generally held by archaeologists that this pattern is associated with climatic changes following the end of Pleistocene glaciation, some of which have been mentioned above. In particular, the colder and drier Younger Dryas, from about 13,000 to 11,500 years ago, is thought by many to have rendered most sites unsuitable for the production of grains in quantities sufficient to support populations that had grown up during the Natufian. Therefore, populations practicing agriculture could occupy only the small percentage of land in the Levant still suitable for cereal crops during the Younger Dryas. As the Younger Dryas abated, according to this argument, technologies sharpened by necessity during the PPNA were put to use in a more widespread distribution of environments suitable for agriculture. The specific environmental situation in which Beidha is located and the chronology of occupation there support the above scenario. They also suggest additional dimensions to it.

To summarize the Natufian occupation of Beidha, it was, by all evidence, one that was seasonal and based in hunting. As noted, the archaeological excavation of the Natufian occupation recovered no botanical remains and almost no artifacts definitely associated with cereal harvesting or processing. In contrast, a great deal of material associated with hunting, especially of caprines, and, of these, especially, Capra aegagrus, the wild ancestor of domesticated goats (Capra hircus aegagrus), was recovered. About this period at Beidha, Bruce D. Smith has written (1998:81):

The spread of cereal cultivation between 10,000 and 9,000 B.P. [uncalibrated] was by no means universal or everywhere alike. Well-watered settlements adopted the combinations of cereal crops that best matched their distinctive hunting-and-gathering economies, while settlements in less favored settings, such as Beidha in the southern Levant, continued to rely exclusively on wild plants and animals.

The PPNB reoccupation of Beidha occurred 2,000 years after the first. According to the chronology reconstructed in Fig. 13 at right, this was just after the 500-year interval of marked and rapid warming and increased precipitation that followed the close of Younger Dryas. Thus it would seem that the site was reoccupied precisely when the climate again became very hospitable to the plant and animal species exploited by the Natufian occupants. The second occupation, however, was much more complex. As recently as 1998, E.B. Banning said of Kirkbride's excavations at Beidha that they "…provide the best glimpse we have to date of a whole PPNB village" (1998:194). Among other characteristics, Beidha, although small, displayed an array of sophisticated architecture that bespeaks of social and cultural complexity.

As shown in Fig. 13, the correlation between climatic episodes, cultural periods, and the dates at which Beidha was occupied seems very clear (the chart is a composite of dates taken from a variety of sources: Bar-Yoseph and Belfer-Cohen, 1992: 21-24; Birks and Ammann, 2000:1391-1393; Steig, 1999:1485-1887; Bar-Yoseph and Meadow, 1995: 44-45; Gebel, 1992: 86; Simmons, 1997:314). The current landscape analysis suggests the principal reason for this. Beidha depended upon most of its water from two sources. These were the few, and currently somewhat meager, springs located upslope, along the geologic unconformity between the sandstone and limestone, and precipitation. While minor weeps in the sandstone (caused by the interface of different types of sandstone) might have occurred at various times in the past, as they do now (Paradise, 2000: personal communication), they would probably have contributed only a small percentage of the total water available to the various forms of life inhabiting the canyon in which Beidha is located.

Water from precipitation would benefit cereals and other vegetation that was exploited by humans. Water from both precipitation and the springs (which would have flowed more freely during periods of heavier precipitation) would be channeled past Beidha, according to the hydrological model seen in Fig. 3. Abundant precipitation, then, would have brought a dependable source of water for humans occupying the canyon and the animal and plants species there that were being exploited by the humans.

The presence of a dependable source of water at certain time periods might explain how it happens that the inhabitants of Beidha during both the Natufian and the PPNB so heavily exploited the Bezoar goat (Capra aegagrus or Capra hircus aegagrus). Bruce D. Smith regards this as anomalous. Smith places the native range of Capra aegagrus somewhat to the north of Beidha (1998:59). He goes so far to suggest that hunters at Beidha "…may well have been killing ibex rather than goats," noting the difficulties in telling the differences between bones of the ibex (Capra ibex) and Capra aegagrus. Smith makes this statement despite the fact that Kirkbride was quite sure that most of the caprine bones she recovered were Capra aegagrus. As Smith also notes, however, "…it is not yet clear how far the range of the wild goat extended south into the Levant. The southern, more arid portion of this region was inhabited by the wild ibex (1998:58)" Thus the explanation for this anomaly might be that goats ranged further south at times of heavier precipitation. The combined attraction of the goats and flowing water, especially when put alongside the other attractors mentioned above, were evidently enough to act as a catalyst for the Nabataean occupation at Beidha. When precipitation decreased with the advent of the Younger Dryas, the Nabataean occupation of Beidha ended. Quite likely this was because the water and the goats in the canyon had disappeared.

Package and Punctuated Equilibrium

Human habitation was reestablished at Beidha when precipitation regained high levels, about 500 years after the start of the Climatic Optimum. There is strong evidence that some of the cultural complexity evident in the PPNB occupation of Beidha developed there. In the PPNB component of the site, cereals were found that were being cultivated, including barley in a state of transition from its wild form, and emmer wheat. The hulled, two row Hordeum spontaneum was recovered from the earliest PPNB levels. This would strongly indicate that native barley was being harvested at Beidha, not domesticated cereal that had been brought in from elsewhere. Similarly, Dexter Perkins, who conducted the analysis of the faunal material recovered by Kirkbride, considered that "…the Madamagh-Beidha faunas strongly suggest that goats were domesticated at Beidha during the Neolithic," (1966:66). He identified the species domesticated as Capra hircus aegagrus. Another highly visible change was in habitation type. The round and relatively irregularly shaped structures of the lowest levels at Beidha (Levels VII-V) change by steps into the more regular, strictly rectilinear structures of Level III.

The agriculture/animal husbandry package that underlay this rich complexity, however, is probably best thought of not so much as a set of plant and animal species associated with permanent architecture as a cultural system. As I have argued elsewhere (Comer, 1996), cultural systems are based in habitual ways of regarding, maintaining, and manipulating the world. They include social controls held in place by groups in society that have benefited from them in the past and expected to benefit from them in the future. Perhaps most essentially they depend upon repetitions of behavior patterns in which meanings are assigned to people, animals, features on the landscape, and, indeed, virtually all aspects of human experience. Such behaviors tie members of societies to an imagined world order that is predictable and enduring. These patterns are tenaciously maintained so that the world can continue to be ordered in familiar, predictable ways. Typically, they are reformulated only in the face of a perceived imminent collapse of the culture.

The cultural package that provided the basis for the second flowering of human occupation at Beidha had been developed during the PPNA. Bruce D. Smith notes that agriculture began in the Levantine Corridor sites with "…a common characteristic-a dependable source of water for growing crops in the form of a high water table." The secure source of water they used to "ensure good harvests from species they had moved down from their natural habitats on dry slopes at higher elevations." A prerequisite of this kind of transport of crops is at least a prototypical cultural package of the sort described in the paragraph above. He argues that a dependable source of ground water was the common variable at PPNA sites (1998:74).

The other major economic component of the full agriculture/animal husbandry package was the domestication of goats. In a recent publication, Zeder and Hess argue that domestication of the goat first occurred in the Zagros Mountains of western Iran some 10,000 years ago (calibrated) (Zeder and Hess 2000 pp. 2254-2257). Yet is important to note, with Bar-Yoseph and Meadows, that, "In the Near East…where we have our earliest evidence for the domestication of food species, there is no site known in which the herding of domestic animals can be shown to have taken place without the population having had access to cultivated plants" (1995:83).

I would suggest that this observation provides an important clue to both the way in which the agricultural/animal husbandry package was formed, and the environments to which it most easily spread. All cultures integrate the landscape into the sort of imagined world order mentioned above. Landforms are scenes of the mythological exploits of the ancestors and gods. Directions are ideologically significant. This integration becomes more complex and elaborate as the behavior sets associated with agriculture are adopted. The organizing principal for all cultural systems is kinship, real or fictive. Meanings assigned to the landscape and the beings that inhabit it, including animals and plants, are conflated with the kinship system. Thus we have totems, clans, and other social units that are tied in the thinking and in the practical conduct of everyday life to the human relationship with the landscape. These culturally defined units regulate human behaviors in ways that are needed to coordinate the tasks necessary to any economy. In a society supported by agriculture, they regulate, for example, sowing and reaping, land use, processing and storing grain, and ways in which the harvest is shared. These sets of prescribed and proscribed behaviors can, with not much modification, provide a good structure for animal husbandry, as well. They could, for example, organize the movements of domesticated animals through the landscape in a way that would complement agriculture. The rewards would have been enormous. Increased agricultural productivity would be one of them. Reinforcing these behaviors would have been the milk and the materials for making tents, clothing, rope, and other useful items that cannot practically be obtained from herds of wild animals.

PPNB sites have yielded striking artifacts that are associated with the complex ideological structure that developed with the agricultural economy. These include plaster skulls, plaster figurines, and even architecture itself. The last of these clearly reflects and reinforces a more complex social partitioning. It seems likely, as noted, that the basic cultural structure for this developed in the PPNA, although it was quite surely elaborated in the PPNB. A credible scenario is that the PPNB occupation at Beidha is an example of something like a cultural version of what Stephen J. Gould calls "punctuated equilibrium" in biological evolution. According to this theory, environmental change can open econiches in which organisms that are appropriately preadapted will opportunistically and rapidly evolve further. When precipitation once more reached high levels, Beidha canyon became fertile ground for pre-adapted agricultural/animal husbandry occupation, which flourished and evolved there apace, and probably synergistically, with other PPNB sites. The occupation continued until the lush precipitation abated, which, at Beidha, removed an essential element from the PPNB cultural system: water. It is probably not too much to say that a dependable supply of large amounts of water becomes increasingly important as cultural complexity rises. The complex culture that had developed at Beidha in the PPNB could not survive without it.

Bibliography

Banning, E.B.
1998 The Neolithic Period: Triumphs of Architecture, Agriculture, and Art. Near Eastern Archaeology 61,4:188-235.

Bar-Yosef, O. and A. Belfer-Cohen
1992 From Foraging to Farming in the Mediterranean Levant. Pp.21-48 in A.B. Gebauer and T.D. Price (eds), Transitions to Agriculture in Prehistory. Madison, Wisconsin: Prehistory Press.

Bar-Yosef, O. and Richard H. Meadow
1995 The Origins of Agriculture in the Near East. Pp. 39-94 in T.D. Price and A.B. Gebauer (eds), Last Hunters, First Farmers. Santa Fe, New Mexico: School of American Research Press.

Birks, Hilary H. and Brigitta Ammann
2000 Two Terrestrial Records of Rapid Climatic Change During the Glacial-Holocene Transition (14,000-9,000 Calendar Years B.P.) from Europe. Proceedings of the National Association of Science, 97, 4:1390-1394.

Byrd, Brian F.
1987 Beidha and the Natufian: Variability in Levantine Settlement and Subsistence. Ph.D. Dissertation, University of Arizona, University Microfilms, Michigan.
1989 The Natufian Encampment at Beidha: Late Pleistocene Adaptation in the Southern Levant. Vol. 1, Excavations at Beidha. Jutland Archaeological Society Publications XXIII:1. Moesgard, Arhus: Aarhus University Press.

Comer, Douglas C.
1996 Ritual Ground. Berkeley, California: University of California Press.
1999 Cultural Site Analysis as a Tool for Research and Planning, an Example from Cape Coast, Ghana. Paper presented at the World Bank Cultural Site Management Workshop, April 28, 1999.

Ed-Badbul, Ali
1999 Personal communication.

Gebel, Hans Georg et al.
1988 Late Epipaleolithic Aceramic Neolithic Sites in the Petra-Area. Pp 67-100 in A.N. Garrard and H.G. Gebel (eds), The Prehistory of Jordan. Oxford: BAR International Series, No. 396, Part 1.
1992 Territories and Paleoenvironment: Locational Analysis of Neolithic Site Settings in the Greater Petra Area, Southern Jordan. Pp. 85-96 in Susan Kerner (ed), The Near East in Antiquity: German Contributions to the Archaeology of Jordan, Palestine, Syria, Lebanon, and Egypt.

Halbaek, H.
1966 Pre-Pottery Neolithic Farming at Beidha: A Preliminary Report. Palestine Exploration Quarterly 98:61-65.

Henry, Donald O.
1989 From Foraging to Agriculture: The Levant at the End of the Ice Age. Philadelphia: University of Pennsylvania Press.

Kirkbride, Diana
1966 Five Seasons at the Pre-Pottery Neolithic Village of Beidha in Jordan. Palestine Exploration Quarterly 98:5-61.
1968 Beidha: Early Neolithic Life South of the Dead Sea. Antiquity 42:263-74.
1989 The Environment of Petra During the Pre-Pottery Neolithic. Pp. 117-124 in Studies in the History and Archaeology of Jordan, Vol. 2. Amman, Hashemite Kingdom of Jordan: Dept. of Antiquities.

MacNeish, Richard S.
1992 The Origins of Agriculture and a Settled Way of Life. Norman, Oklahoma: University of Oklahoma Press.

Madonia, Paolo
1999 Personal communication.

Paradise, Tom
2000 Personal Communication

Perkins, Jr., Dexter
1966 The Fauna from Madamagh and Beidha. Palestine Exploration Quarterly 98:66-67.

Saskatchewan Agriculture and Food
1995 Nitrogen Fertilization in Crop Production Fact Sheet. Saskatchewan: Government Press.
1998 Managing Wheat for Protein Premium Fact Sheet. Saskatchewan: Government Press.

Smith, Bruce D.
1998 The Emergence of Agriculture. New York: Scientific American Library

Steig, Eric J.
1999 Mid-Holocene Climate Change. Science 286, 5444:1485-1487.

Wright, H. Jr.
1977 Environmental Change and the Origin of Agriculture in the Old and New Worlds. Pp.281-318 in C.A. Reed (ed), Origin of Agriculture. The Hague: Mouton.

Zeder, Melina A. and Brian Hesse
2000 The Initial Domestication of Goats (Capra Hircus) in the Zagros Mountains 10,000 Years Ago. Science 287: 2254-2257.

Figures and Captions

Fig. 1 Coregistered satellite images

Fig. 2 SIR-C/X-SAR (Space Shuttle Synthetic Aperture Radar) image of greater Petra Area, showing main geologic/environmental zones.

Fig. 3 Beidha region water catchment.

Fig. 4 Elevations in Beidha are in meters.

Fig. 5 Three-dimensional elevation map of the Beidha area, showing the location of Beidha and Dibidibah springs.

Fig. 6 Map of the immediate sandstone canyon system in which Beidha is located.

Fig. 7 Bedouin tent occupying high ground next to Beidha.

Fig. 8 Possible sources of flint used by inhabitants of Beidha

Fig. 9 Phosphorus content in soils near Beidha.

Fig. 10 Potassium content in soils near Beidha

Fig. 11 Ideal yield and protein response to nitrogen availability.

Fig. 12 Nitrogen in parts per million. Amounts tend to be highest in areas most frequented by goats (canyon).

Fig. 13 Chronological chart of Beidha occupation.