Across Atlantic Ice

by Dennis J. Stanford

  If level 7 does represent a continuous occupation that lasted for a year, there must have been an increase in fuel availability. Although the environment at this time appears to have been relatively treeless, there is an increase of roe deer in the faunal record. Since these are generally considered woodland animals, their presence may imply an increase in the amount of forest near the cave. Enhanced availability of local firewood or other fuels would have enabled a small band to occupy the cave more continuously than had been possible in earlier times.

  Table 8.1 shows that the amount of time spent living in and hunting from the cave varies considerably by level. It is noteworthy that no animals were incorporated into the cave deposits, with the possible exception of levels 8–9, between mid-winter and the end of spring. This means there was no hunting here during this period and that people either hunkered down at the site or abandoned it while the winter storms raged. Frozen or dried meat from fall and early winter kills could have supported a band for a short period, masking the time actually spent in residence. We are impressed, however, by the correlation of the length-of-stay calculation and the seasonality data and therefore accept that the consumption figures tend to indicate the length of time the site was occupied. If the site’s occupiers abandoned it between mid-winter and the end of spring, they must have moved to a location where resources were more abundant, such as sea mammals on the coastal ice.

  The upper Solutrean levels appear to represent two residential patterns, both of which ultimately focus on red deer as the main food source. These patterns differ in use intensity and length of occupation. If the occupations were continuous, the people living in levels 7–9 may have spent up to a year at a time in the cave, whereas in levels 10–13 the stays were much shorter, averaging around three months. In those levels where it is possible to assess the seasons of occupation, they extend from spring to early summer, and the evidence from level 16 indicates that some people also stayed for a short while in early winter.

  SOLUTREAN SETTLEMENT PATTERN MODELS

  The geographer Karl Butzer hypothesized a multistage Paleolithic settlement for this part of Spain.10 His model begins with the Middle Paleolithic Mousterian and traces a slow geographic change in settlement patterns that correlates with a steady increase in human populations and technological advancements. He proposes that the primary residences were initially located at a central place established in the north Spanish piedmont belt, near or along a major river and the piedmont-mountain ecotone, where people could have easy access to a variety of resources in multiple ecological zones: a group living in a camp at this location could exploit both the coast and the mountains. Each major river valley would have defined a core territory, but minor valleys might have been shared with groups on either side. When these areas became depleted, the band would simply move to another river drainage.

  By Solutrean times the population had increased and both the marine and montane zones were being exploited. Multiple groups still had overlapping operational areas, but during the subsequent Magdalenian period there was a shift to both diversification and specialization, with more effective integration of marine and montane resources. According to Butzer this situation developed into two distinct systems, one on the coastal plain and the other along the piedmont-mountain ecotone. The splitting of valleys into these two resource areas would have been particularly adaptive as the number of competing groups increased. The coastal population would have become more dependent on marine resources, requiring higher labor input but producing fairly predictable yields, while the inland groups would have relied on game trails, especially during periods of reindeer migration. The importance of this last phase of Butzer’s model is that it would have accommodated a doubling of the human population in the area.

  Straus, in a model similar to Butzer’s, views the Solutrean settlement pattern as a central-base foraging system with increasing emphasis on aquatic resources, and from La Riera he estimates an average two-hour walk to and from the coast.11

  From our examination of the cave evidence, we generally concur with the original investigators that the occupations can be divided into three groups based on length of residence, types of activities, and animals targeted, but in light of these models we offer the following observations for consideration.

  Levels 2–3 are the earliest Solutrean occupations of La Riera, and because of the way the cave was used they stand out as different from the levels that follow. In fact, despite the presence of several key Solutrean artifact types and evidence of a shift in the selection of raw material sources, the hunting strategy associated with these levels appears to be more like that of the earlier Aurignacian people than of the later Solutreans. We suggest that these earliest hunters foraged around and from a central base residential camp, in this case La Riera, and killed animals as they encountered them, without targeting specific species or ages. When the local resources were depleted the band moved on to occupy another area.

  This settlement behavior is compatible with Straus’s interpretation and is the overlapping operational stage of Butzer’s model, in which the local band shares resource areas with neighboring groups. Our conjecture is that these undated occupations occurred early enough in the LGM that its climatic changes and the increased population had not yet impacted resources in the Vasco-Cantabrian area, where it was business as usual.

  After the onset of the cold, dry period, population pressure began to reduce the size of territories and bands started establishing semi-permanent base camps in strategic locations near more reliable resource areas. Working out of their home base, these foragers used short-term occupation sites in task-specific localities and returned home with whatever goods they acquired during the trip. Levels 4–6, the specialized ibex and red deer hunting camps recognized by Clark and Straus, are clear examples of the use of La Riera as a temporary shelter for staging task-specific foraging trips. The mix of shouldered projectile points and indented base projectile points may indicate that at least two different contemporary social groups used the cave.

  We suggest, on the basis of the thinness of the occupation lenses, that these levels were produced by small subsets of extended family groupings, perhaps consisting of only male hunters on two-to-three-day trips. The fish remains also suggest that many of these trips were late summer or fall excursions, when the salmon and sea trout were spawning. At the cave the hunters either consumed or filleted the fish whose remains were recovered from the occupation levels, while they perhaps dried or smoked other fish and took them back to a base camp along with selected high-yield food packages.

  Because of the dramatic appearance of aquatic resources such as shellfish, fish, and seal, we propose that semi-permanent base camps for the La Riera hunters were downstream and near the beach or estuaries (figure 8.2). Campsites in sheltered localities near these environments would have had many advantages over upland campsites, including relatively continuous opportunities to obtain high-yield resources such as fish, waterfowl, and some sea mammals. Shellfish would be a great resource when found near a campsite, whereas their advantage would ebb exponentially when transported many kilometers. Another direct advantage would be the possibility of collecting driftwood that ends its long journey from southeastern North America tossed by storms onto the beaches of the Bay of Biscay.

  The adjacent continental shelf was a vast plain and an excellent habitat for herd mammals that, along with the abundant coastal and marine resources, would have provided food for people living on the coast. In fact, for bison and horses this cold dry, steppe environment would have been much superior to the unstable, steep slopes and deeply incised valleys of the Cantabrian piedmont. Bones from dredging operations in the North Sea and English Channel indicate that mammoth herds and other extinct animals also grazed on the plain.12

  FIGURE 8.2.

  Ecological zones where the highest-return resources were available throughout the year.

  If semi-permanent camps were in prominent coastal locations, the
y could have enhanced inter-band social contacts between areas of southern France and Iberia. Here we see the effects of two important terrain-dependent aspects, visibility and accessibility. First, people would have been much easier to find camped at prominent sites along the beach than hidden away in the broken topography and deeply incised countryside of the piedmont and mountains. Second, the continental plain would have been a virtual inter-area autobahn connecting groups from Asturias to the Aquitaine and beyond. Trails on the plain would have been far superior for travel than the seesaw trails up and down the steep and unstable slopes of the high ground and over-mountain passes that were seasonally blocked by snow and ice—although except when the plain’s rivers were frozen over, some sort of watercraft would have been desirable to manage their crossings.

  In terms of Butzer’s model this period represented a transition from the early overlapping operational central place pattern to the beginnings of a regional specialization phase. Although there was probably still overlapping use of neighboring territories, extended kin networks or other alliances would have socially formalized the use of these areas. In levels 4–6 the settlement system would have included entire drainage areas, from the shoreline to the snow line. Once a base camp was established, upland hunting would continue, but on an ad hoc basis and probably highly related to seasonally specific tasks such as fall hunts, when red deer and ibex were in prime condition.

  We agree that the last series of occupation levels (7–17) shows that the cave was again used primarily as a home base, but here the pattern of use was more complicated and variable than in levels 2–3. The refuse midden of level 7 is large enough to suggest a long-term encampment. It appears to be related to the previous three levels in that it maintains a strong component of both estuarine and riverine resources, as well as shared artifact types. It differs in the development of intensive red deer hunting, suggesting another change in the social, settlement, and procurement systems of the valley brought on by a climatic shift, loss of some maritime species, economic specialization, or perhaps continuing population increase.

  By level 7, we think, the population had increased to the point of beginning the “fission” stage of Butzer’s model, when the semi-related group of people (a band) separates into distinct inland and coastal groups. It is clear that bands at this level were beginning to use the piedmont and lower foothills for more extended periods of time or even on a semi-permanent basis. From level 7 up, with the exception of a period between levels 10 and 13, the duration of camp activities seems to range from eight or nine months to perhaps a year.

  There is, however, a period from mid-winter through spring when the cave seems to have been abandoned by all later Solutrean people—oddly enough, corresponding to prime sealing months. In 81 percent of the Solutrean occupation levels for which there are seasonal data, the caves appear to be abandoned during the time when seal hunting would have been optimum on the coast. Figure 8.3 compares the calculation of seasons of the upper Solutrean occupation of La Riera Cave with periods of seal vulnerability (the birthing schedule of seals that live on the ice edge is regulated by seasons, so that pups are weaned by the time the ice breaks up in the spring). This correlation is highly suggestive that inland groups of Solutrean hunters moved down to the coast for several months during late winter to participate in sealing and probably provided the coastal people with mountain products such as ibex, chamois, and roe deer meat and durable goods. For the coastal group, most seal hunting could have taken place anytime there was sea ice in the neighborhood, either landfast or floating. Plus, the harbor seal was a year-round coastal resident and could be hunted during the summer, when sea ice was absent.

  FIGURE 8.3.

  Seasonal chart of seal and walrus presence on the northern Spanish coasts and La Riera Cave use in Solutrean times.

  Perhaps we are seeing evidence of a settlement pattern similar to that of the Inuit of northwest Alaska. There the coastal and inland populations generally operate independently, but during certain seasons of the year the inland groups go to the beach for trading, hunting, and other social activities with their coastal relatives (a periodic regional aggregation group, in Binford’s terminology). We contend that this would have been part of a natural progression leading to Butzer’s adaptive divergence stage.

  Both Butzer and Straus make passing mention of a possible Solutrean coastal adaptation, but they do not seriously consider it while model building because there is no proof. Based on the hard data available, it would indeed appear that economic intensification and integration of marine and montane resources began during the Magdalenian, but we think this reflects archaeological invisibility caused by the post-LGM sea level rise rather than a difference in cultural development.

  This point is illustrated by Darrin Lowery’s work on Late Woodland occupation in the watershed of Chesapeake Bay in North America.13 The eastern shore of the bay includes large estuarine sites with dense shellfish refuse and ample evidence of the use of other marine resources, as well as interior hunting-oriented sites with limited debris. The interior sites are generally located near freshwater springs that had been attractive to hunter-gatherers since Paleolithic people settled in the area. If global warming were to cause today’s sea level to rise 2 meters, the Late Woodland sites with marine refuse middens would disappear and there would be no archaeological record of the use of marine resources. The settings of the surviving terrestrial Late Woodland sites would mimic the settings of the European Paleolithic sites and provide fewer clues than exist at La Riera to indicate that the Late Woodland peoples gained most of their resources from the bay and ocean. Similar situations exist in many other places in the world where sea level rise or coastal subsidence has wiped out entire components of prehistoric settlement systems.

  We assume that, like the Late Woodland fishing folks, most prehistoric people operating out of semi-permanent base camps did not carry marine hunting and fishing equipment with them while exploiting upland terrestrial environments unless it was multi-functional, such as spear throwers and darts, which are effective on both open land and water. Bows and arrows and self-barbed lances are also effective in a variety of hunting and fishing environments. Boats and toggling harpoons, on the other hand, would have been left behind.

  We would not normally expect to see marine food items carried any great distance from the coastline. In an excellent summary article of the issues involved with archaeological evidence for aquatic adaptations, Jon Erlandson points out that when modern hunters-gatherers forage beyond their base camp for 5–10 kilometers, they do not usually transport the skeletal remains of shellfish, fish, or sea mammals back to their residential camp.14 Therefore, sites located more than about 5–10 kilometers from an ancient shoreline are not likely to contain substantial evidence of marine resource use, and even in sites as close as 1 or 2 kilometers from shore the density of aquatic remains drops significantly. The exceptions to this are marine mammal ivory and teeth that were used for adornment. These specimens can be transported great distances from the coast and were traded to remote interior peoples. Levels 4–6 at La Riera Cave contained a remarkable amount of estuarine and marine remains considering the cave’s location in an upland setting and its distance from the coast.

  TECHNICAL INNOVATIONS

  Clark and Straus illustrate resource intensification beginning in Solutrean times with the development of technical innovations to increase hunting efficiency.15 Among the new technologies they consider are decoys, nets, movable fences, and weapons such as spear throwers and possibly bows and arrows. All of these hunting aids would have been made of perishable materials (e.g., wood, plant fiber, animal tissue), which are usually not preserved in archaeological contexts. Except for evidence such as spear thrower parts recovered in excavations and their depictions in rock art, the use of these new hunting technologies by Solutrean people remains in the domain of speculation. But since the number of new artifact types found in Solutrean sites is relatively extensive, it
is clear that these people were heavily invested in research and development (see chapter 5). It seems probable, therefore, that they developed and used not only these hunting aids but also myriad other tools and exploitive and protective systems for which there is no direct evidence. This would be especially true of equipment for the exploitation of marine resources, which was likely stored at campsites near beaches long ago destroyed by sea transgression.

  All of the available evidence suggests that European Paleolithic armaments prior to Solutrean times consisted primarily of thrusting and casting weapons. A wide variety of new projectile point types and several new weapon systems first appear during the Solutrean and are suspected by Clark and Straus to be the result of experimental efforts to develop more efficient means of killing game as a response to population increase. Although this may have been true, we think the situation was more complicated and that several issues directly related to climate change should be equally considered as part of the impetus for the development of new hunting gear and techniques. Perhaps the diverse varieties of the projectile point types seen during Solutrean times were experimental prototypes, many of which were abandoned through time.

  The Solutrean people were among the first modern humans to face a glacial period in Europe, and climatic change, especially associated with deforestation, would have had a direct impact on their hunting techniques and available raw materials. The loss of forests and redistribution of animals would have greatly diminished the opportunities for ambush hunting such as seen in levels 2–3 of La Riera Cave. Instead of a multitude of places along game trails in the woods, there would have been fewer locations where hunters could hide. Game stalking would have become a fine art as opportunities to get close to targets were greatly reduced. This would have been especially true on the vast plain of the continental shelf, which became a mecca for ungulates. An additional stress resulting from the loss of the forest environment would have been the dramatic decrease of wood for tools, especially long straight spear shafts, and for fuel. It is probable that driftwood cast up on the ice age shores augmented the supply of wood from the rare and scrubby native trees.