Late Holocene human occupation of the Quequén Grande River valley bottom: settlement systems and an example of a built environment in the Argentine

Gustavo A Martínez Consejo Nacional de Investigaciones Cientificas y Tecnológicas (CONICET), Investigaciones Arqueológicas y Paleontológicas del Cuaternario Pampeano (INCUAPA) Departamento de Arqueología, Facultad de Ciencias Sociales de la Universidad Nacional del Centro de la Provincia de , (B7400JWI), Olavarría, [email protected]

Quentin Mackie Department of Anthropology, University of Victoria, British Columbia, Canada [email protected]

Keywords Argentina, Pampas archaeology, landscape archaeology, organisation of technology, built environment

Abstract

This paper presents a preliminary model of the occupational history of the valley bottoms at the edges of the bed of the Quequén Grande River (Argentina) during the late Holocene. The ultimate goal of the research is to situate some aspects of technology, mobility, land-use patterns and settlement systems as a proximal consequence of a long-term process of ‘lithification’, that is, the positioning of lithic raw material across otherwise lithic-free areas of the landscape.

In order to address this issue, distributions of lithic artefacts are used to discuss features of the regional techno- logical organisation and settlement systems and the relationships between people and the landscape. In that sense, lithification, a variant of a ‘provisioning places’ strategy, has implications for other aspects of a human adap- tive system. The lithification process has influenced the organisation of technology, in particular the degree of plan- ning and anticipation necessary, which in turn affects the degree to which technological strategies (eg, curation and expediency) were employed. Lithification also has implications for the organisation of logistical and residen- tial mobility strategies by encouraging reoccupations, changing periodicity of reoccupation, altering landscape use patterns, and making for longer seasonal or task-specific stays. One end result is an artificial conflation of resources, and a lessening of resource heterogeneity. For example, there will be more places where critical resources, such as water, fauna, and flora, co-occur with the lithic resources needed to exploit them. The lithic raw material distribution is only partially dependant on natural occurrence because the environment has been reor- ganised and (intentionally or otherwise) built by human activity. We propose that in the Pampas the late Holocene witnesses a process of ‘building a landscape’ which had implications for social organisation and hence played an important role in regional human adaptation and cultural evolution.

1 Introduction legs, or be placed in bags. Rock can be quarried and Undoubtedly, the spatial arrangement of critical envi- turned into portable tools, or small nodules can ronmental variables such as water, food and lithic raw become part of a person’s gear. But each resource material are important determinants of behavioural has differences in its underlying distribution, inherent organisation and archaeological site location. Yet portability, durability and ability to be cached. Some none of these resources has a completely fixed loca- resources, such as lithics, are essentially indestructi- tion on the landscape. Water flows and can be carried ble and if moved from one place to another may per- short distances by people. Food can move on its own sist in the new location, and be available there for a

Before Farming 2003/4 (1) 1 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie very long time. The environment is altered by a com- 2 Environmental background for the south- bination of these re-positioning activities, which may eastern portion of the humid Pampas have a very transient or an almost permanent effect The Pampas are a plain located in eastern on the world to which humans must adapt. And while Argentina between 31° and 39° south latitude. The many agricultural societies are understood to alter boundaries of the Pampas, which comprise approxi- their landscape, the same is also true of those who mately 600,000 km2, are the Parana river to the subsist by hunting and gathering. north-east, the South Atlantic Ocean to the east, and However, many studies of technological organi- the Sierras Centrales and foothills of the Andes sation take the view that the environment is some- Cordillera to the west. Two principal sub-regions have thing separate from humans, something neutral to been proposed within the Pampas region: the humid be entered into and adapted to (Ingold 1993). This Pampas (or eastern or proper Pampas), and the dry Cartesian view should be questioned. Rather, we (or western) Pampas. These sub-regions are divided believe that humans and their environment make an by the 600 mm isohyet, which runs parallel between inseparable, mutualistic pair, with the human niche the steppe or pseudo-steppe of the Provincia created in a two-way, reflexive interaction with the Pampeana to the north-east, and the xerophitic environment. Post-processualist archaeologies forested area called Provincia del Espinal to the have urged the revision of overly-rigid formulations south-west (Cabrera 1976; Soriano 1992; see figure of human ecology. However, we feel that productive 1). The study area is centred on the middle stream of alternatives to the Cartesian status quo are uncom- the Quequén Grande River (figure 1), which is locat- mon. To exemplify this theory of the mutualistic envi- ed within the Area Interserrana Bonaerense ronment, we use the relatively easily-assessed (Necochea District, , proximal process of lithification (Webb 1993), name- Argentina), in the humid Pampas sub-region. ly the positioning of lithic raw materials across an The humid Pampas are characterised by a tem- otherwise lithic-free landscape. This strategy of pro- perate climate, with a warm period between visioning places (Kuhn 1995) with critical, yet other- November and March. July and August are the cold- wise unevenly distributed resources, could encour- est months. Rainfall decreases from the north-east, age a substantial re-organisation of both the use where there is approximately a 1000 mm annual and perception of the landscape. In effect, the lithic average, to the south-west, where the average is residue of patterned human activity acts to change approximately half that. The sub-region is dominated the very settlement system which gave rise to the by graminaceous steppe or pseudo-steppe vegeta- activity pattern in the first place. Yet, because the tion, characterised by short shrubs and grasses redistributed resources are then subject to depletion (Cabrera 1976; Soriano 1992). The zoographic clas- through use, there may be a long term cyclical or sification of the area falls within the Dominio reflexive relationship between people and the envi- Pampásico of the Guayano Brasileña sub-region ronment they are modifying. It is proposed that lithi- (Ringuelet 1955), in which mammals and birds are fication is a human behavioural component that adapted to open environments. The humid Pampas is results in a long-term built environment (Ingold a grassland, one of whose defining characteristics is 1993), a tangible, enduring record of a people’s own a nearly complete absence of native trees. history, which may have implications for both short Within the humid Pampas, seven biogeographic and long-term mobility, settlement strategies, and areas can be identified, as shown in figure 1. These for the organisation of technology. Taking into are the Northern, Southern, Western, the Salado account the changes in the landscape caused by River depression, the Interserrana, the Ventania and human action, we will discuss some possible ways Tandilia hills. Of these, only the last three areas, in which these anthropogenic changes subsequent- which cover an estimated landscape of ca 90,000 ly influence the occupational history of this region. In km2, are relevant to this paper. The Interserrana area the following sections we introduce a relevant is a very flat grassland plain located between the archaeological case study from the Argentine Tandilia and Ventania hills, with the Atlantic coast to Pampas, followed by discussion in light of these the south-east and the Caruhé depression to the introductory comments. west. The Tandilia hills lie in a discontinuous group,

2 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

Figure 1 The Pampean region, subregions, areas, main physiographic divisions and the location of the middle basin of the Quequén Grande River (modified from Berón & Politis 1997)

300 km long by 60 km wide, the highest point of which et al 1995; Prieto 1996; Tonni et al 1999; Zárate et al is 525 m above sea level. The Ventania area is a con- 2000; amongst others). tinuous range of hills 180 km long and 60 km wide, 3 Late Holocene archaeological background with their highest point reaching 1247 m. These two of the south-eastern portion of Buenos hill ranges were the main prehistoric sources of lithic Aires Province raw material, such as quartzite, chalcedony, granites, Few late Holocene subsurface archaeological sites cherts and others (see discussion below) across the from this area had been studied before 1970, with humid Pampas (Bayón et al 1999; Flegenheimer & most archaeological information derived from surface Bayón 2002; Politis 1984) sites and museum collections (Politis & Salemme Palaeonvironmental research suggests that con- 1990). Since the early 1980s, systematic and extend- ditions throughout the Holocene times were basically ed excavations in stratified late Holocene sites have arid/semi-arid, although during the middle Holocene been more common (see Berón & Politis 1997; Politis (ca 6500–3500 years BP) the hypsithermal brought & Madrid 2001). warmer and, probably, more humid climatic condi- Late Holocene sites in our sample fall within the tions. Around 4500 years BP the ecosystem started to Zanjón Seco Phase of the Interserrana Bonaerense undergo a return to more semi-arid/arid conditions Tradition as proposed by Politis (1984). The main which lasted until 1000 years BP, when current envi- characteristics of the tradition are a dominance of ronmental characteristics were reached (Bonadonna quartzite as raw material, and the use of flakes as

Before Farming 2003/4 (1) 3 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie blanks, with subsequent unifacial (unimarginal) economic changes (see discussion in Lourandos reduction. Traits specific to the phase include small 1985; 1997; Price & Brown 1985; Bender 1981; triangular bifacial projectile points, pottery, bone tools, Williams 1987; Mackie 2001; among others). Thus, and the exploitation of animals such as Lama guani- based on a socially oriented approach, intensification coe or guanaco (large wild camelid), Ozotoceros is understood here as a process where not only eco- bezoarticus (small deer), Rhea americana or Rhea, nomic and resource production are promoting (large flightless bird), Lagostomus maximus (large changes but also more intense and competitive social rodent), and Chaetophractus villosus, Zaedyus relationships would play an important role. In conse- pichiy, and Dasypus hybridus (small armadillos), quence, intensification refers to social as well as eco- among others (Martínez & Gutiérrez in press; Miotti & nomic variables, which may themselves bear directly Salemme 1999; Mazzanti & Quintana 2001; Politis & or indirectly on the economy (see discussion in Salemme 1990). Increased numbers of food-process- Martínez 2003). ing implements such as large grinding stones suggest 3.1 Archaeological overview of the study area a greater dependence on plant resources than in ear- lier times. In the Interserrana and Serrana areas the following On the basis of recent research (Barrientos 1997; sites have been investigated: Arroyo Seco 2, (upper Martínez 1999, 2003; Politis et al 2001; Politis & component), La Toma (later component), Tres Reyes Madrid 2001) it has been proposed that the 1 (upper component), Zanjón Seco 2 and 3, Interserrana and Serrana areas have been more Cortaderas, Fortín Necochea, Campo Borchetto, intensively occupied during the last 3000–2000 years Laguna de los Chilenos, Laguna de Puán, Laguna BP than formerly. Places in the landscape associated del Trompa, San Martín 1, Ea. La Mascota outcrops, with riverine and lagoon environments became the Laguna de Puan, La Liebre outcrops, Arroyo location of residential, multi-purpose and long-lasting Diamante outcrops, El Guanaco, Túmulo de la settlements, frequently used and re-used through Malacara, Paso Otero 1 (upper component), Cerro La time. These changes were part of an increasing China 1 and 2 (upper component), Cueva Tixi (upper process of cultural complexity that included intensifi- and middle archaeological levels), Lobería 1 and 2, cation (Martínez 1999). The concept of intensification among others (see references in Politis & Madrid has been applied to hunter-gatherers societies in dif- 2001). These include both surface and subsurface ferent ways but is generally attached to the notion of sites. complexity (see Ames 1994). It is important to The size of the middle stream of the Quequén 2 remember that no one single factor or prime mover is Grande river is ca 1000 km (figure 2), although the usually responsible for the rise of cultural complexity archaeological sites examined are located in the val- and/or intensification. These processes deserve indi- ley bottoms of the Quequén Grande River and its 2 vidual attention or explanation for each particular tributaries, covering an area of only ca 160 km . In case because there are so many divergent points of general, within the Pampas , settlements view stemming from different theoretical orientations. located along the river valleys are either at localised Lourandos (1985), for example, stressed that trans- flat areas along the riverbanks or on gently undulat- formations related to resource intensification have ing low relief. All of the sites within the study area are commonly been considered as a by-product of exter- located at the edges of the valley bottoms or on low- nal forces such as climatic or demographic changes. lying linear rises immediately adjacent to the river While economic intensification (Crown & Wills 1995) (see figure 2). Subsurface and surface archaeological and resource intensification and production have materials are usually found in the same features been the most commonly addressed in hunter-gath- (aeolian settings) of these landforms. Only about 40 erers societies (see Basgall 1987; Broughton 1994; cm separates the younger part of the La Postrera Janetski 1997; Wohlgemuth 1996), other spheres geological formation (where Zanjon Seco 2 and 3 are such as ritual, exchange, status, social networks, located, see discussion below) from the current soil alliances, and so on can be also intensified, fuelling surface. The archaeological materials from both sur- changes in the technological, subsistence and settle- face and subsurface sites correspond to the same ment base, and accelerating the process of socio- archaeological component, but have been disturbed

4 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

Figure 2 Distribution of surface and subsurface sites on the middle stream of the Quequén Grande river (numbers indicate archaeological sites). Modified from Martinez 1999 by modern agricultural activities (Politis 1984; ferent kinds of settlements, linked by multi-purpose Martínez 1999). Thus, even though both kinds of sites base camps and special purpose killing and initial present different analytical possibilities due to their processing activities (Martínez 1999). Although the distinct natural and cultural formation processes, the Zanjón Seco sites (see figure 2) are located close to origins of these sites are related to the same cultural the ploughzone, degradable materials like bones and dynamics and prehistoric behaviours during the last pottery have not been uniformly destroyed by agricul- 3000 years BP (Politis et al 2001). ture. Both the integrity of individual bone specimens and the preservation of the oval garbage structure 3.2 Subsurface sites filled with bones in Zanjón Seco 3 (Politis 1984) sug- At least two different landforms – fluvial plains and gest that these sites are in primary context. This is aeolian deposits – were simultaneously used for dif- possible because these sites are located near the steep banks of the river where ploughing activi- ties are not typically carried out. The portion of workable land beside the river depends on the topographic features and a boundary which separates an area with undisturbed sealed subsurface sites from another with disturbed surficial sites is created, based largely on mod- ern agricultural practices (figure 3). Analysis of the lithic technol- ogy and faunal remains from

Figure 3 View of the Quequén Grande river. Arrows show boundaries between ploughzone and sites such as those at Zanjón undisturbed areas where sealed sites have been excavated Seco indicates that they are the

Before Farming 2003/4 (1) 5 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie result of multiple activities, leading to their interpreta- 3.4 Lithic raw material outcrops in the humid tion as base camps (Politis 1984; Politis et al 2001). Pampas: distribution, quality and abundance Activities carried out at Zanjón Seco 2 include tool Given the main issue addressed in this paper, knowl- reduction, and the secondary butchering and con- edge about the distribution, availability and quality of sumption of prey. Both lithic technology and faunal lithic raw materials is a key topic. As noted, quartzite remains are well represented. Zanjón Seco 3 repre- is the most common raw material in both surface and sents a rubbish deposit or garbage structure, located subsurface sites (ca 90%), followed by chalcedonies, in a small natural oval depression that contains main- quartzes, cherts, silicates, and others. Quartzite ly guanaco bones and very little lithic debris (Politis includes a great variety of rock types that should be 1984; Politis et al 2001). In these late subsurface distinguished from each other, and their provenance sites guanaco is typically the most important animal identified (Bayón et al 1999:188; Flegenheimer & resource, although there is also evidence that small Bayón 2002). Rock outcrops of any kind are rare in animals were also consumed (Politis 1984). the Interserrana area, especially ones suitable as lith- ic resources. In general, the rock exploited came from 3.3 Surface sites the surrounding Tandilia and Ventania hills, but there Natural tall grasses and the dynamics of cultivated is an uneven natural distribution of kinds of rocks fields have influenced the archaeological visibility of across that landscape. The Tandilia ranges are char- these sites and the composition of surface artefact acterised by isolated rather than continuous outcrops scatters. Ploughing in the surrounding area of the (see figure 4) and the rivers do not transport lithic river valleys has had a twofold effect. On the one material more than four km from a watersource. In hand, artefacts become permanently relocated both contrast, the Ventania Hills are formed by continuous horizontally and vertically. On the other hand, archae- ranges, and rivers transport and move quartzites ca ological visibility changes through time due to the 100 km along the main streams that flow into the cycle of tilling activities (ploughing, seeding, harvest- Atlantic ocean. This creates secondary natural ing, etc). Given the conditions of site formation deposits of quartzites from the hills to the sea in the processes previously described (differential erosion boundaries between the south-eastern portion of the and aggradation), combined with the timing of field Interserrana and Ventania areas (Bayón et al 1999; seasons, the number of artefacts present at sites can- Bayón & Zavala 1994; Flegenheimer & Bayón 2002; not be established by any single survey. This means Flegenheimer et al 1996; Flegenheimer et al 1999). that gaining a larger sample of materials will depend Thus, the Tandilia and Ventania hills and the sur- on performing a series of surveys in different seasons rounding areas show different spatial patterns of in the same tilling areas. availability of raw material. Surface artefact scatters have been used from a Two main groups of quartzites can be identified by regional perspective for such different purposes as microscopic thin sectioning, and their differences in providing functional interpretations (Orcutt 1986; distribution across the landscape plotted: ortho- Sullivan 1995), for the reconstruction of patterns of quartzites are present in Tandilia hills and formation processes (Camilli 1989), for understand- metaquartzites are found in the Ventania hills (for ing differential patterns in subsistence (Sullivan more details regarding the continuity, visibility, sec- 1995), as indicators of where subsurface deposits ondary deposition and distribution of these kind of should be investigated (see discussion in Dunnell & rocks see Table1 in Bayón et al 1999:195; Simek 1995) among others. In this study, the ultimate Flegenheimer & Bayón 2002). It is possible to distin- goal of analysing surface material will be to under- guish the two kinds of quartzite by macroscopic stand the distribution of lithic raw material throughout analyses alone, using properties such as grain size the landscape, taking into account its source, (eg, and composition, brightness, the presence of outcrops), its place of deposition, and the conse- siliceous cement, cortex thickness, degree of weath- quences for settlement systems, the organisation of ering, fracture type and properties of surfaces, colour, technology, mobility, and the construction of a land- among others (Bayón et al 1999:201). The most scape through a strategy of provisioning places important quartzites represented in the archaeologi- (sensu Kuhn 1995; see below). cal record belong to the upper orthoquartzites of the

6 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

Figure 4 Map showing the main quartzite outcrops and their spatial relationship to the study area and sites. (1cm = 16.5km)

Sierras Bayas formation in the Tandilia Hills1. The as chalcedony (microcrystalline silicas or the so- well known, fine orthoquartzite outcrops of Del called ftanitas, see discussion in Barros & Messineo Diamante (see figure 4) are in the Tandilia range, 2003), rhyolite, silicates, and so on follow the specifically in the surrounding area of the localities of quartzites in declining abundance. In the case of San Manuel and Barker. With much less frequency, chalcedonies there are still some doubts about silicified dolomites, microcrystalline silicas and cherts whether or not this rock is present in both ranges have also been recognised in this area (see La Liebre (Franco 1991, 1994; Lozano 1991). Outcrops of rhy- outcrops in figure 4; Barna & Kain 1994; olite have been identified in the west sector of Flegenheimer 1991; Flegenheimer et al 1996; Ventania hills and the adjacent plains (Oliva & Flegenheimer et al 1999; Pupio 1996). Moirano 1997). Limited outcrops are also identified in In the northwest part of the Tandilia hills, Barros the Interserrana Area itself (see Llambías & Prozzi and Messineo (2003) have used macroscopic analy- 1975), such as coarse grained quartzites in the Lumb sis and X-ray diffraction to source outcrops of chert locality (Politis 1984; Ormazabal 1994), silicified tuff and silicified dolomites. Geological maps show the (Madrid & Salemme 1991), and quartz (Politis 1984). orthoquartzites of the Sierras Bayas formation are These are low grade materials in an otherwise almost probably present in this area, although actual out- rock-free area. crops or quarries have not yet been recognised in To an even lesser degree, basalt pebbles from the surveys (Barros & Messineo 2003). Atlantic coast – where large quantities of this raw mate- Another point of interest is the presence and rial can be found – were utilised, as well as artefacts and absence of quartzites different than the ortho- blanks mainly featuring bipolar reduction. The frequency quartzites recorded at Tandilia hills, and the other of both basaltic raw material and artefacts decreases rock types in the mountain ranges or in the adjacent toward the interior of the Pampas (Politis et al 2003). plain. In the archaeological sites, raw materials such In addition to the study area data mentioned

Before Farming 2003/4 (1) 7 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie above, Bayón & Flegenheimer (2003) have analysed Quequén Grande River and its tributaries. These a late Holocene surface collection from the El places have been repeatedly revisited, and the col- Guanaco site. These authors discuss the lithic raw lection was assembled under the criteria of a single material database at a regional level and the different collector. All items recovered have been mapped (see modalities of rock transport, noting that ortho- figure 2), all classes of artefacts were identified by quartzites of the Sierras Bayas formation are the best site, and typological and technological studies have quality raw material in the Pampean region. They pro- been carried out (Politis 1984; Martínez 1999). Even pose that artefacts made of these orthoquartzites though no formal distributional approaches (sensu have been moved more than 275 km in all directions. Dunnell & Dancey 1983; Dunnell & Simek 1995; In the particular case of the El Guanaco site, 146 kg Ebert 1992) have been performed, complementary of rocks were transported distances ranging from 13 analyses based on systematic surveys from other km (eg, pebbles from the Atlantic coast) up to almost surface sites (eg, El Puente, La Horqueta, Campo 300 km considering that quartzites are coming from Piacenza sites, etc) have been done by the first both Tandilia and Ventania mountain ranges. author (see Martínez 1999). Although much of this raw material was employed in To date, ca 100 surface archaeological sites have grinding tools, 46 kg were flaked artefacts or cores been recorded in the middle basin of the Quequén made mainly on the Sierras Bayas formation ortho- Grande River. Figure 2 shows the location of the main quartzites. Pyramidal and discoidal cores up to 500 g sites (n=70) which without exception are located are commonly found together with exhausted ones. along the banks of the Quequén Grande River and its Bayón and Flegenheimer (2003) interpreted this as tributaries (an area of ca 160 km2 ). Thousands of evidence of the stockpiling of raw material, and they artefacts have been recovered from this area, of also proposed a site provisioning strategy. Grinding which a sample has been studied in order to record artifacts (eg, mortars and pestles) found at this site artefact variability and attributes like size, weight, and have also been interpreted as site furniture, further volume (Martínez 1999). There are 917 lithic artefacts reinforcing the idea of extensive anticipation and in this sample, comprising different categories such planning. as mortars, pestles, bola-stones, hammerstones and cores. Table 1 shows classes of artefacts, and their 3.5 The lithic artefact sample quantity, the frequency of sites where artefacts were The data presented here are derived from the archae- found, weight of artefacts, number of artefacts per ological materials housed at the Museo de Ciencias km2, and the weight of artefact classes per km2. Naturales del Club de Pesca de Lobería (see In the case of mortars, 299 items have been Martínez 1999). They have been collected by Mr recovered from 52 sites (74 per cent of the sites), and Gesué Noseda, who for 40 years has conducted they are the heaviest items recorded. Considering the amateur surveys in the surrounding area of the area of the valley bottoms (160 km2) an estimation of

Artefact class Number of Number & percentage Weight Number of artefacts Weight by km2 (n:70 sites) artefacts of sites per artefact class (kg) per km2 (kg) (total:160 km2 ) (total:160 km2 )

Mortars 299 52 sites (74.28 per cent) 1701.5 1.86 10.63 Bola Stones 227 45 sites (64.28 per cent) 63.5 1.41 0.39 Pestles 88 27 sites (38.57 per cent ) 28.2 0.55 0.17 Cores 235 26 sites (37.14 per cent ) 112.6 1.46 0.7 Hammerstones 68 16 sites (22.85 per cent ) 18.9 0.42 0.11

Total 917 1924.7 5.7 12.02

Table 1 Summary of artefact frequency, size and distribution in the study area (after Martinez 2002)

8 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

10.6 kg/km2 can be made. In a sample of mortars sites). Weights vary from a few grams up to ca 10 kg coming from the Zanjón Seco locality, the weight of (figure 6), with 70 per cent of the specimens weighing these artefacts varies between 1 and 17 kg, although less than one kg. Core types include amorphous, ca 70 per cent fall between 1 and 9 kg (Ormazabal pyramidal, bipolar, discoidal, prismatic, and globular. 1994). A figure of almost two items per km2 can be Most of the cores have multi-directional flake scars, estimated. although those with only bi- and uni-directional flake The acquisition of raw materials thus becomes a scars are also present. Small chalcedony cores, key element for understanding this pattern. In the which for the most part are exhausted, are less com- middle stream of the Quequén Grande River, out- monly found at these sites (Martínez 1999). crops of a brown-dark coarse grained quartzite have Bola-stones are also well represented and they been recorded at the Lumb locality (see figures 2 and are mainly manufactured on granite. Pestles, ham- 4), 10 and 35 km distant from Zanjón Seco and La merstones and pottery (N=1344) are also found in Horqueta archaeological sites, respectively (see also these contexts. Besides these, a significant amount figures 1 and 2). For Zanjón Seco, lithic raw material of formal-informal tools as well as debris have also from Lumb can be considered as a local resource, but been observed. For example, ca 1000 formal tools this situation is likely different regarding La Horqueta, (sidescrapers, scrapers, knives, etc) have been due to the distance (approximately 35 km from the recovered from the surface site of El Puente (see outcrops) and also the nature of rock and the purpos- Martínez 1999). In general terms, the content of the es for which has been used. The quality of the Lumb assemblages tends to have a pattern: ie, co-presence quartzite outcrops is poor for production of flaking of categories such as projectile points, preforms, side artefacts and they have been almost exclusively used scrapers (the so-called raederas), scrapers, knives, for ground stone artefacts. For instance, in the case notches, short retouched edges, and multi-purpose of the Zanjón Seco sample, ca 85 per cent of grind- tools as well as pottery. A variety of debris from ing artefacts have been made on the coarse grained renewed cores, such as side and platform flakes and Lumb quartzite. Nevertheless, 12 per cent of the mor- microflakes produced by retouching activities, are tars recovered from the Zanjón Seco area were made present, leading to great variability in the debris from on igneous rocks (eg, probably granites) from the blank retouch. Most of this debris does not have Tandilia hill ranges ca 50 km away (Ormazabal, 1994; remaining cortex (see Martínez 1999; Politis 1984). Politis 1984, see figure 1). Due to the formation processes described above, Both cores and flaked formal artefacts have been and the fact that only a few collections have been mainly manufactured on a fine-grained ortho- studied, the quantity of artefacts and lithic raw materi- quartzite. As mentioned before, this quartzite comes al previously described is only a very minimal estima- from the surrounding area of Canteras del Diamante tion. In other words, considering the problems of visi- outcrops (Flegenheimer et al 1996; see figures 1 and bility and natural-cultural formation processes, these 3), located between 50–70 km from the study area in figures do not exhaust the potential information which the Tandilia ranges (see discussion below). Other could be gained from future surveys or provided by exotic raw materials from the Tandilia hills, such as granite and, probably, chalcedony have also been used, although their frequencies are low compared to quartzite. Considering the cores (figure 5), quartzite is the best represented raw material: 235 items have been recovered from 26 sites (37.1 per cent of the

Figure 5 Variety of cores collected at the edges of the Río Quequén Figure 6 Detailed view of one the largest quartzite cores recovered Grande (1 cm = approx 15 cm) from the middle stream of the Quequén Grande river

Before Farming 2003/4 (1) 9 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie other unknown private collections. Nevertheless, the Cores and artefacts found on surface sites at Del available data are sufficient to instigate a discussion of Diamante Quarry resemble those of the middle stream different aspects of the distribution of lithic raw materi- of the Quequén Grande River (figure 7). This quarry is al and the organisation of technology. suggested to have been exploited at least since ca 4500 years BP (Flegenheimer et al 1999). Less good 3.6 Summary of study area archaeology quality quartzites, such as those coming from the The majority of the stone tools in the Interserrana area Lumb locality and isolated outcrops of the Interserrana are made of quartzites obtained from the surrounding area, were also exploited, but with less frequency, as Ventania and Tandilia hills. For the purposes of this were chalcedonies. Due to their differing qualities, paper the most important outcrops are those of good- some quartzites show variable patterns of exploitation quality orthoquartzite located in the upper fluvial valley and distribution. For instance, artefacts made on the of the Arroyo Diamante (Flegenheimer et al 1996; brown-dark coarse grained quartzite coming from Flegenheimer et al 1999), the nearest source to the Lumb locality have a lower frequency. This material study area. The macroscopic characteristics of the was usually used only for ground stone technology. majority of quartzite cores and formal tools found at Conversely, the fine grained orthoquartzites from the the sites of the middle stream of the Quequén Grande Tandilia hills have much higher frequency of associa- River indicates that the raw material belongs to the tion with artefacts and debitage such as cores, and upper orthoquartzites of the Sierras Bayas Formation formal and informal tools with differing degrees of located in the Tandilia hills, more precisely those locat- reduction. Different kinds of artefacts have been man- ed in the surrounding area of the Cantera del ufactured with this raw material, differential uses have Diamente outcrops (see figure 4). These are nearly been inferred, and variable patterns of management, 50–70 km from the sites studied here. As noted, the El exploitation and distribution of these lithic raw materi- Guanaco site (Bayón & Flegenheimer 2003) contains als between the geomorphic zones are clear. At this orthoquartzite from the Sierras Bayas Formation that point it can be suggested that a mixed pattern of was transported 125 km from its source (see figure 4). quartzite exploitation was practiced. The Lumb

Figure 7 Cores observed at the Cantera del Diamante outcrops (photograph by Gustavo Martinez)

10 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie quartzite can be considered as a local raw material in processes and conditions not explained by taking into the case of Zanjón Seco locality (the eastern portion account only one or the other. The concept of lithifi- of the study area, see figure 2), and as an exotic in the cation seems to have properties for merging both case of La Horqueta locality (the western portion of approaches together. the study area). Orthoquartzites and granites from the In this, we follow Ingold’s (1993:169) definition of a Tandilia hills are in all cases considered exotic raw building as ‘any durable structure in the landscape materials, due to their distance. The lithic assem- whose form arises and is sustained within the current blages from both the stratified and the surface sites of human activity.’ Over time, the built environment have common technological characteristics (Martínez loses its distinction from the natural environment, and 1999; Politis 1984). both become part of the human niche. This process of It is important to note that the assemblages reflexive niche construction via long-term historical include a significant number of heavy items such as accumulation has profound implications for how we grinding stones, mainly metates and manos, and think of the archaeological record. We outline some cores which are large in size and number, and which examples of this process before returning to a general are widely spread throughout the area. That is also discussion of its influence in the Pampean case study. the case of hammerstones and bola stones. The Ethnoarchaeological research carried out in tropi- presence of large artefacts and cores a long way from cal rain forest has yielded important information relat- the source of the raw material raises questions of ed to the construction of landscape. Study of the mobility and procurement patterns which we address Nukak, a hunter-gatherer group from the Amazonian below. The Inteserrana area is largely devoid of rock tropical rain forest of Colombia, shows that they of any type, let alone suitable material for making manipulate or manage gathered wild plant resources stone tools. Data from both sealed and surface sites in a way that is not domestication (Politis 1996 a & b). will be used together in a preliminary model which When they move from one camp to another, viable describes the main characteristics of technological seeds of exploited wild plant species remain in the organisation, settlement and mobility strategies, and abandoned camps. In order to build the camp, the also the relationship between humans and the physi- Nukak cut some small trees and branches off, and as cal and social environment as exemplified through a result the sunlight is easily able to reach the soil the process of lithification. We think that the patterned under the canopy. The discarded seeds in these association of these materials at the edges of the val- camps are thus in a better competitive position rela- ley bottom sites shows a strongly structured archae- tive to the other wild plants. Nukak construction, con- ological record, and the analysis of this structure will sumption, and mobility behaviour thus has a twofold permit the recognition of at least some aspects of the effect on their local environment. On the one hand, through exploitation they create a plant resource organisation of the cultural system which produced it. patch which will be further exploitable in future cir- 4 Hunter-gatherers and the built environ- cumstances and, on the other they alter the natural ment distribution of plant resources by changing the diver- The structure of the regional archaeological record sity of wild plants in small spots through the land- depends on the long term behavioural patterns of scape. The Nukak have a high residential mobility hunter-gatherers operating in conjunction with non- and the paths they use to move around are always in cultural factors related to both resource distribution a close spatial relationship with the old and aban- and landscape dynamics (Binford 1990) and to cul- doned campsites which, as a consequence of tural factors such as the material residue of previous Nukak’s behaviour, contain enhanced populations of human activity (Ingold 1993; Gosden 1994; Zedeño those wild plants collected during normal daily forag- 2000). Analyses drawing on the former are common- ing activities (Politis 1996 a & b). place in processual approaches, while those inspired Similar behaviours have been observed in other by the latter are much less common, particularly in hunter-gatherer groups. Laden’s (in Bailey & the archaeology of hunter-gatherers. Nevertheless, it Headland 1991) work among the Efe groups, in the is difficult to establish the boundaries between them Ituri forest has shown that edible plants have a more as complex natural and social factors can include dense concentration in paths and old camps sur-

Before Farming 2003/4 (1) 11 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie rounding areas than in other randomly sampled areas cal forest of Popayán (southwestern Colombia). of the forest. In this regard, it is suggested that ‘ . . . Criticising the ecological-reductionism approaches the Efe’s use patterns of the forest and its plants has Gnecco (2000:123–126; 2003) suggests that the resulted in a non-random distribution of edible plants hunter-gatherer groups were already altering the nat- and has inadvertently enhanced the abundance and ural structure and distribution of resources in these exploitability of those plants for human foragers’ ecosystems with a high sense of territoriality as early (Laden, personal communication in Bailey & as in the Pleistocene-Holocene transition. Headland 1991:266). Deur (2000) has documented While discussing the rise of complex hunter-gath- the enhancement of naturally occurring rhizomes in erer societies in the Western District of Victoria, small ‘gardens’ in Kwakwa’kawakw (Kwakiutl) territo- Australia, Williams (1987) describes the construction ry on northern Vancouver Island, Canada. Elsewhere of artificial features such as mounds, which do not on the Northwest Coast of the Americas, non-agricul- appear in the archaeological record of this region until tural people nonetheless managed many resources, after 2500 years BP. Except for sandy deposits, all including the weeding, tilling, and burning of mead- the poorly drained soils of the region become water- ows to promote production of camas bulbs, a variety logged, but by building mounds a more intensive set- of edible lily (Beckwith 2002). Although disused for a tlement of the region becomes possible as the settle- century, these meadows still show an unnaturally ments need no longer be restricted to well-drained high concentration of the edible varieties of camas. In lunettes. Once the mound is built, it can be used as a the same vein, Hutterer (1982:175) has suggested campsite during periods of prolonged wet weather or that ‘certain aspects in the patchy distribution on flooding. Mounds can be resurfaced for long term use plants in tropical forests may be an effect of and the largest clusters are found in areas where long-range and continuous human presence.’ sedentism was made possible by a high biomass and Another example of hunter-gatherers manipulating diversity of foods resources. Thus, the construction of the resource structure of the environment is the wide- mounds was a necessary condition which allowed spread practice of deliberate fire-setting in order to people to live in these flooded areas on a semi- enhance young green growth with the aim of attracting sedentary to sedentary basis. Williams suggests that herbivores (eg, Johnson 1994; Turner 1991; for a the stockpiling of discarded sediment into mounds review see Lewis 1982). Once again, it is possible that represents a significant change in behaviour leading short term intentionally-motivated behaviour had long to an increased investment in durable facilities and, term unintentional consequences for the overall because of this, people increasingly returned to the patchiness of the environment, and this may have same camp-site. The building of mounds is linked to influenced, or even created, the long-term fire regimes all of the changes in social networks and not only rep- which are important parts of many forest ecosystems. resents a substantial change in settlement patterns The manipulation of wild plant resources by and sedentism, but they also can be considered as a hunter-gatherer populations, which is a different territorial symbol (Williams 1987:318–319, see also process from domestication (at least in the classical Lourandos 1997). sense of this term), has had some influences on the A similar case has been identified in north-west reproduction, distribution and, abundance of these and the lowlands of southern (South plants. This manipulation or management (intentional America) with the so-called cerritos (small mounds). or otherwise) of wild resources by everyday hunter- Research carried out by Andrade Lima & Lopez Mazz gatherer foraging, mobility, technology and settlement (in press), Lopez Mazz (2001), Bracco et al (2000), systems strategies shows that in some tropical forest Cabrera (1999), among others, has shed new light on (eg, Amazonia) and temperate forest (eg, British the knowledge of these societies in localities such as Columbia) environments, the very concept of primary Merin lagoon, San Luis, etc. The cultural develop- forest should be reviewed (Politis 1996 a & b; Deur & ment of hunter-gatherer-fisher societies in an envi- Turner in press; Boyd 1999; cf Denevan 1992). ronment of rich and biodiverse wetlands linked with The ethnoarchaeological cases mentioned above the Atlantic can be tracked back to the last 5000 can be reinforced with archaeological observations years BP. As was the case in Australia, the mounds made by Gnecco (2000, 2003) in the mountain tropi- were built as a strategy to make the settlement of fre-

12 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie quently flooded areas feasible. An important change scape in anticipation of future needs (Close in settlement patterns occurred ca 2500 years BP, 1996:551), is relevant. In effect, this caching allowed after the hypsithermal, when a specialised economy an expedient lithic strategy to be implemented of hunting and gathering was established along with despite the natural lack of raw material (Close 1996: the management (without implying domestication) of 551). As described above, a third example of the lithi- palms like butiá (Butiá capitata). All these features fication of a rock-free environment is found in the suggest a process of intensification related to a broad Pampas. All these examples can be compared to spectrum economy (López Mazz 2001:243).There Parry and Kelly’s (1987) paradigmatic example of the was an increment of population, a specialised econo- provisioning places strategy: in some semi-sedentary my in a highly productive environment, more seden- contexts of North America in which activities were tism and reoccupied places, a reduction in residential concentrated in long term-residential locations, peo- mobility, etc, along with technological innovations ple simply moved raw materials to these places, such as pottery, microlithics, etc. Traits of complexity henceforth exploiting them casually or expediently, in are indicated by an incipient social differentiation, set- predictably occupied sites. In the case of the tlement system hierarchies, monumental activities, Pampas, places with other resources such as plants, variety of funerary patterns, cooperative participation water and guanaco were provisioned with large in the construction of cerritos, etc. Mounds with high- cores, hammerstones, anvil stones and mortars er numbers of structures, variety of forms, dimen- (Martinez 1999). The implications of this built envi- sions and settlements along the main natural land- ronment are discussed below. forms of the landscape show the construction of land- 4.1 Provisioning places, building environments scape. Realising that foragers can manipulate their envi- Hunter-gatherer settlement systems can include dif- ronment is clearly essential if one is to understand ferent kinds of mobility (residential and logistical) in how they adapt with that environment. We perceive a different degrees, producing mixed patterns which do general trend towards acknowledging that the not simply accord with the ends of the collector-for- human-environment relationship can be mutualistic at ager continuum (Bamforth 1990; Binford 1980, 1982; all scales of societies. Lithification is a further exam- Kelly 1983, 1992, 1995, among others). For this rea- ple of this environmental manipulation. In western son, we are not trying to define the organisation of the New South Wales, Australia, a systematically sam- late Holocene cultural system of the study area into pled surface survey of a sandy environment mapped types, ie, as collectors or foragers. Rather, we will try almost 200 cobbles of lithic raw material (quartzite, to identify whether the mobility strategies included, to silcrete and sandstone), most of which must have some degree, both components. The same reasoning been imported from at least 40 km distant (Webb is applied to such other dichotomies as the two basic 1993). Many of these cobbles were worked in one strategies commonly recognised in the organisation manner or another, including use as grinding stones, of technology, curation and expediency (Binford anvils, and cores. Importantly, at least some were 1979; Nelson 1991), and in such technological strate- imported whole, their flaking qualities untested. Webb gies as provisioning places and provisioning individu- (1993:108) interprets this pattern as the deliberate als (Kuhn 1995). Thus, from an organisational point of positioning of cobbles across the landscape in places view it is crucial that curated and expedient technolo- where there was a potential requirement for stone, a gies, logistical and residential mobility systems, col- process she terms the ‘lithification of a sandy envi- lector and forager settlement systems, and provision- ronment’. Another example of this process is given by ing strategies must not be considered, through typo- Close (1996). In the eastern Sahara there are many logical thinking, as mutually exclusive systems sandy stretches lacking in lithic raw materials. Close (Binford 1987:45; Kuhn 1995:25; Nelson 1991:65). documents that untested cores weighing up to 40 kg The study of technological organisation is one of were transported distances of 10 to 15 km. While the the most sensitive ways of examining how these largest cores were probably carried by pack animals, strategies interrelate. Raw material, tool design and this was not always the case and the underlying settlement system analyses can be used to elucidate motive, to stockpile a missing resource on the land- how technological changes reflect broad scale behav-

Before Farming 2003/4 (1) 13 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie ioural organisation in past societies (Binford 1987;Carr sion themselves it is usually with light, versatile, gen- 1994; Nelson 1991; Torrence 1989,1993; Odell 1996; eral purpose items. On the other hand, provisioning Shott 1986). Thus, the exploitation of a critical resource places refers to a strategy for coping with anticipated such as lithic raw material should allow one to under- requirements by supplying places in the landscape stand some aspects of the past cultural systems. with raw materials, cores and implements having fur- Raw material availability plays an important role in ther toolmaking potential, which are likely to be need- the organisation of technology (eg, Andrefsky 1991, ed in future circumstances. This strategy requires 1994; Bamforth 1990; Nelson 1991; cf Carr 1994 and accurate knowledge about the location where differ- Odell 1996). The principal lithic raw material outcrops ent activities might occur. Materials provisioned to a exploited by indigenous people in the humid Pampas place can be of any size, limited only by limits on are located in the hilly areas. Most formal artefacts human effort and desire. Beyond these differences, such as cores, sidescrapers, scrapers, and projectile both strategies share the same basic principle of points are made primarily of quartzite and secondari- planning or anticipatory organisation. ly of chalcedony, both of which require lengthy trans- It is proposed that one of the central effects of the port into the region (see Politis 1984; Flegenheimer et lithification process in the study area was to make a al 1999; Franco 1994). There is an uneven natural heterogeneous lithic resource environment more homogeneous: a system of artificial quarries was distribution of lithic raw material and an absolute established by the placement of large numbers of absence of rocks in the lower portion of the tributary large cores across the landscape. This process would streams that flow into the Quequén Grande River, the help produce conflation of resources in some geo- only known exception being the poor quality morphic zones, in this case, the valley bottom of the quartzites from Lumb. The most frequently used river. It is likely that people were bringing lithic raw quartzites are naturally absent alongside the materials to these places along the river in order to Quequén Grande River and evidently the late combine them with other important resources such as Holocene people were moving rocks throughout dif- guanaco2, water, and small animal and plant commu- ferent geomorphic zones (ie, the Interserrana, nities along this riverine environment. The durable Tandilia and Ventania areas as well as the Atlantic archaeological result are sites which contain unusual- coast), in a spatial pattern reflecting the availability of ly large cores and other lithic site furniture. raw material. The lithification process has implications for the Nevertheless, the presence and distribution of extension and periodicity of settlement. The availabil- lithic raw materials does not exclusively depend on ity of stone in these areas would have encouraged their natural distribution. An anthropogenic stockpile longer stays3 and the reoccupation of these places of lithic raw material forms, in effect, an artificial out- (sensu Camilli 1989 and Camilli & Ebert 1992). crop or quarry, and can therefore become a source Artefacts and debris generated through a provision- for subsequent exploitation. This stockpiling process ing places strategy would predominate in assem- is known as lithification (Webb 1993:108). Given the blages representing more prolonged occupational amount and the characteristics of the rocks and arte- events or in a pattern of more regular and predictable facts previously described for the middle stream of reuse of places. The presence of stockpiled metates, the Quequén Grande River, a strategy of deliberate manos, hammerstones, cores, among other materi- positioning of lithics across the landscape in places als, is suggestive of site furniture (Binford 1979), where there was a potential requirement for stone which encourages multiple reoccupations. One possi- could have taken place. This process would have bility is that cores, blanks and tools derived from them implications for depth of planning in artefact produc- acted as insurance gear, which is: ‘cached through tion, transport, and maintenance, and the strategies the region, not in terms of specifically anticipated sea- by which potential raw material needs are met (Kuhn sonal needs, but in terms of what might generally be 1995:22). Within this framework, two basic strategies needed at the location at some time in the future’ can be considered: provisioning people and provi- (Binford 1979:271). Nevertheless, the possibility that sioning places. The first refers to strategies in which they have been left in these places anticipating sea- individuals always have at least a limited toolkit at sonal needs (eg, ‘seasonal gear’, Binford 1983) can- hand. Because individuals are mobile, if they provi- not be excluded.

14 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

Available raw material derived from tools already technology. Nelson (1991) points out that an expedi- discarded in these places could have been easily ency strategy requires the accurate anticipation of utilised in order to repair or recycle tools. Variation in both sufficient materials and sufficient time to make the extent of artefact reduction could also reflect tools. The former condition seems to be satisfied by scavenging and the reuse of previously abandoned the bulk procurement and stockpiling of raw material materials (Kuhn 1995). The availability of lithic raw along the river. The latter condition, sufficient time, is material would also have conditioned decisions about satisfied if these sites are indeed base camps whether to replace or maintain tools through further (Martínez 1999; Politis 1984; Politis et al 2001), which reduction, and it would have been linked to the cost implies longer occupations and regular reuse. Taken of stockpiling raw materials (Kuhn 1994). Thus, recy- together, these characteristics could encourage the cling and re-use of tools would have been carried out deployment of expedient technology. Both expedient in these sites leading to the production of a variety of and curated strategies therefore require a high artefacts and debris. Particularly, formal tools like degree of anticipation, and understanding precisely sidescrapers (raederas) and circular scrapers have what this entails could be an important key to under- undergone a systematic series of transformations standing mobility systems and the reoccupation of from the time of their manufacture until they were dis- places (Binford 1983; Gamble 1995). carded or lost (see Dibble 1987). In this sense, such Lithification also has implications for mobility tools would have been flexible (Nelson 1991), and strategies. Both logistical and residential mobility utilised within a maintenance strategy (Bleed 1986). could be inferred from certain technological features The modular design of the sidescrapers is compatible of the lithic assemblages of the study area. The diag- with this idea. nostic technological feature to identify a logistical The presence and variability of formal tools and component of this system should be the bulk pro- debris as well as the several stages of manufacture curement of lithic raw material, which in the study and broken implements represented in these sites area includes both grinding implements and cores. suggest that tool production, maintenance, reworking The bulk of lithic resource extraction must have been and repair (a complex strategy of lithic production carried out in other geomorphic zones (eg, the sensu Ebert 1992) would have been carried out in Tandilia hills) and the material then transported back these places where residential multipurpose camps to base camp locations in the river valley bottoms (cf have been established. The relatively stable geomor- Stafford & Hajic 1992; Binford & O’Connell 1989). The phology of the late Holocene aeolian settings where fact that the bulkier items like raw stone, large cores, these sites are located would have led to long term unworked tool blanks and so on were moved around multiple reoccupations (Camilli 1989; Ebert 1992), the landscape reflects the planned provisioning of resulting in persistent places (Schlanger 1992), used places. This is one of the major characteristics of sys- repeatedly during the long-term occupation of the tems organised logistically: ‘a wide-ranging ‘caching’ region (Martinez 1999; Politis 1984; Politis et al 2001). strategy which insures the dispersion in the habitat of Another implication of this lithification process is goods and materials which may be needed later’ for the technological strategies of curation and expe- (Binford 1983:285). However, as part of this process, diency. Through lithification, both strategies could artefacts and raw materials can have different life- have been simultaneously employed to some degree. cycles and frequencies of use and replenishment. On Curated technology should mean the anticipation of one hand, even if metates and large cores are tool use requirements is accounted for by the classed as site furniture they have different implica- advance procurement of raw material and manufac- tions for the future availability of raw material. ture of implements. Likewise, expedient technology Grinding implements can be understood as stable depends on the availability of raw material at the time site furniture without any further transformation and place of need through happenstance or, where beyond abrasion by using. Cores, however, are the material is not abundant or evenly distributed, unstable site furniture because they have a reduc- through caching and stockpiling (Nelson 1991:80; tionistic use-life history over which they are trans- Andrefsky 1994). The treatment of lithic raw materials formed into other instruments (some of which will (bulk procurement, stockpiling strategy, and so forth) leave the site or be exhausted), and into detritus. in the study area provides insights into expedient Thus, cores are implicated in a lithification-depletion

Before Farming 2003/4 (1) 15 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie cycle whereas metates and manos are not. On the caching and anticipation, usually employed in the other hand, in order to identify the provisioning places archaeology of hunter-gatherers, is also interesting in strategy it is not necessary that the archaeological its own right. record shows a great number of these kinds of arte- Some authors (eg, Franco 1994) have reasoned facts. As site furniture elements, they need not show that lithic raw material in the Interserrana area is a high frequency because, while the frequency of scarce and its distribution differs from, for example, other kinds of artefacts tends to increase through time other resources such as guanaco, which are abundant as a site is used and reused, the presence of site fur- and wide ranging. Franco (1994:86) suggests that the niture elements in sites tends to be relatively constant establishment of caches of cores, flakes and artefacts through time (Binford 1979; Camilli 1989; Nelson in the Interserrana area was performed to save ener- 1991). gy through making lithic procurement more efficient. However, the settlement system could also Indeed, provisioning the landscape with insurance include, to some degree, a component of residential gear is the probable proximate mechanism for the lithi- mobility. The previously mentioned process of fication of the landscape. Nevertheless, while it is true homogenisation in resource distribution (eg, places that natural outcrops in the Interserrana area are with an accumulation of important resources such as scarce and unevenly distributed, this area should be lithics, water, fauna, etc) could encourage the further thought of together with Ventania and Tandilia when employment of a residential mobility strategy, with considering problems of acquisition of lithic raw mate- groups moving between these places marked by rial. The main outcrops at Tandilia and Ventania are agglomerated resources. scarce only in a spatial sense: these outcrops actual- Therefore, in the Pampas we argue that both ly contain abundant material (eg, Flegenheimer et al places and people were provisioned. In the former, 1999). Hence, the notion of availability and acquisition the lithification-depletion process was potentially of raw material should be challenged, because lithic unstable, with the implication that, through building raw material is not a scarce resource, only an incon- the landscape via lithification the mobility strategy veniently distributed one. Moreover, if lithic raw mate- could have fluctuated between the two ends of the rial is readily available (or not) it is partly because residential-logistical continuum. While logistic mobili- humans have made social, economic, and technolog- ty was employed in the provisioning places strategy ical decisions that create this condition (Nelson (resources moved relative to consumers; Binford 1991:77; see discussion above). The availability of 1980, Kelly 19834), residential movements (con- raw material during the late Holocene in the south- sumers moved relative to resources) could have been eastern portion of the Pampean region is not only a possible between these persistent places (sensu reflection of the geological setting, but was produced Schlanger 1992), which contain a (partially anthro- by humans modifying and building the resource struc- pogenic) conjunction of resources and materials. ture of their environment. Understanding the lithification process involves 5 Discussion working at different temporal and spatial scales, and From the ethnoarchaeological and archaeological hence the dynamics of the longer term process examples a common conclusion can be drawn: regarding the timing of the lithification, the shape and through different activities such as the management frequency of the renewal and depletion cycle should of wild plant resources, the movement of lithic raw be understood. Nevertheless, in discussing to what materials, the building of facilities, deliberate burning extent this process was unintentional or intentional we practices, etc, hunter-gatherers manipulate the envi- would argue that in most aspects, such as technology, ronment in ways that alter the natural distribution of mobility, and settlement strategy, they share the oper- resources, changing their availability and abundance. ational need of accurate anticipation. Although almost All of these activities result in a built environment, and any act of knapping stone could be considered as lithi- the lithification of the late Holocene Pampean land- fication via the production and deposition of waste scape is another example of this general process. We (see discussion below), the difference lies in a delib- feel that the process, while shedding light on some of erate action of provisioning places and anticipation. In the commonplace conceptual frameworks, such as this light, the process of lithification should be seen, at

16 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie least to some extent, as an intentional one, in which and in social relations (Martínez 1999, 2003; Politis et people monitored the degree and kind of raw material al 2001). These changes would have generated new resources in order that they would be able to confi- environmental and social conditions during the late dently predict their future needs. This sort of planning Holocene and the management of lithic raw material is adequate for proximate contingencies, and would can potentially shed new light on our understanding include the anticipated presence of site furniture and of this process. In this regard, there is something that other elements of the intentionally-built environment. cannot be explained under the microeconomic In other words, one interpretation of this process umbrella of evolutionary ecology, or its derivative opti- would be simply as long term caching behaviour in mal foraging theory: as part of the lithification process response to the otherwise largely lithic-free environ- a great amount of lithic raw material remained under- ment (ie, Kuhn’s provisionment of places) but at exploited, as shown by the cases of quartzite cores another scale with a different implication for landscape up to ca 10 kg (figure 6)5. The cycle of lithification- approaches, a complementary interpretation is that depletion was permanently activated beyond immedi- lithification produced a built environment. The conse- ate needs. Following Rosenberg (1994:329) it can be quences of each interpretation are different. suggested that ‘…beyond obvious proximate utility, Anticipation entails a conscious behaviour but the the key determinant is familiarity and perceived com- results go beyond the interests of a single generation patibility with key utilitarian elements of the emergent because the outcomes in the long term are a change- system’. Regarding lithic raw material procurement able/changed landscape, and not a pristine nature for and exploitation, the structure of the archaeological future generations to inhabit. This notion has been record of the eastern portion of the Pampas is proba- invoked by one of the authors (Martínez 2002, 2003) bly part of a wasteful behaviour (Neiman & Monticello in order to understand mechanisms linked with learn- 1997), and thus does not necessarily fit the expecta- ing, cultural variation and the introduction of behav- tions of optimality (see discussion below). ioural and artefact innovations during the late Any strategy of lithic reduction will inevitably pro- Holocene, particularly in the study area and in the duce detritus which could be considered as a trivial eastern portion of the Pampean region. From this, the lithification of the activity area. What separates this existence of learning mechanisms such as guided process, inherent since the first stone tools were variation, direct bias, indirect bias and frequency made (Potts 1991), from lithification in the sense we dependent bias under the theoretical foundations pro- are using the term here are the questions of scale vided by the Dual Inheritance models (Cavalli-Sforza and intentionality. First, lithification is intrinsically a & Felman 1981; Boyd & Richerson 1985; Durham regional phenomenon, in which heterogeneities of 1991, Richerson & Boyd 1992; Rosenberg 1994) was resource distribution are evened out through human explored. As was mentioned, during the last agency. Lithification does not always imply intensifi- 3000–2000 years BP the Interserrana and Serrana cation, which in the Pampas we believe is signalled areas have been more intensively occupied, suggest- by a suite of adaptive changes that integrate into a ing an increase in population density, changing new organisational background. Second, while the demography, a greater dependence on plants (sup- creation of a debitage scatter is largely unintentional ported by the indirect evidence of larger amounts of and even undesired, lithification always includes an facilities such as processing grinding stones), tech- element of intentionality and anticipation. The inten- nological innovations and artefact acquisitions (eg, tion might not initially have been to lithify a resource- the bow-arrow and pottery), reoccupation of specific poor environment, but rather merely to stockpile raw places in the landscape associated with riverine and materials for an expedient reduction strategy. Over lagoon environments represented by residential, time, the implications of repeated stockpilings might multi-purpose and long lasting settlements frequently have become clear, and an intentional strategy of used through time, etc. These changes were part of environmental alteration could have arisen, which an increasing process of cultural complexity which itself then acted to transform social organisation. In would have included a process of intensification that this model, lithification can be seen as an exaptation, mainly affects three spheres: intensification in pro- (Gould & Vrba 1982), in which a structure which duction, in the use of specific places of landscape, evolved for one purpose then is coopted for another

Before Farming 2003/4 (1) 17 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie purpose. It is a powerful fact that evolution does not within the short term of cumulative single-event inten- act on a blank slate, but modifies that which has gone tionality. Instead, theoretical bridges between individ- before, occasionally producing structures which are ual action and the long term broad scale occupation- adaptive, but not necessarily optimal. By analogy, al history of the landscape are needed. lithification can be seen as exaptation built upon short Gosden’s (1994) usefully ambiguous concept of term stockpiling behaviour. Some behaviours which the landscape of habit removes us from discussions of arise under one set of socio-cultural variables can the intentionality of the lithification over the long term, become coopted for new purposes when the need and potentially leads to a greater understanding of the arises, and persist because of peoples’ familiarity landscape than could any view solely based on sys- with them as part of the habitus. It follows that lithifi- temics, feedback, and proximate anticipation/fulfil- cation need not be optimal in a strict sense. A cost- ment of need. The landscape becomes understood as benefit approach to optimality is a useful heuristic the sum of the constituent acts of dwelling, meaning it device in many cases. When we note that the case of is a circular reasoning to think of long term adaptation long term stockpiling of surplus lithic material is not to something that is, in effect, a record of past and optimal, we do not imply that optimality models are present activity in the world (see the concept of not useful tools for interpreting past social behav- ‘taskscape’ in Ingold 1993). It is a powerful fact that life iours. Indeed, deviations from optimality expectations must be lived amongst that which went before, and are often of most interest. this must be as true of hunters and gatherers as any This suggests that explanations might have been other group of people. While hunters and gatherers linked with other dimensions that go beyond the are not commonly thought of as creating their own merely utilitarian and adaptive patterns. Taking into worlds, their actions in the environment may produce account the socio-environmental conditions previous- material culture and environmental change, which in ly described for the last 2000 years BP that were part itself is productive of further actions. Material things of an increasing process of cultural complexity, the are not ‘passive brute objects, but are engaged in learning and transmission processes which may have reflexive relationships with the people who create or been most influential are indirect bias and frequency use them’ (Gosden 1994:17), a relationship which dependent bias (see definitions in Boyd & Richerson over the longer term creates a built environment, a 1985; Richerson & Boyd 1992). Under these socio- niche constructed through action and imagination environmental circumstances it is expected that con- (Lewontin 1982:159). In this way, the material world formist versions of behaviours based mainly on imita- acts as a prompt, with the structured residue of past tion have taken place. Mechanisms such as guided actions (the archaeological site in the past) helping variation would have played a minor role (if at all) due structure future action. This is a kind of indirect com- to the risk that strategies such as trial and error entail munication mediated through material culture (Wilson in more complex societies where the weight of tradi- 1975:186; cf Mackie 2001:62–63). The raw material tion and punishment are strongest and cultural mech- left behind at sites acts as a signal which encourages anisms encourage imitation rather than innovation. reoccupation, even, hypothetically, when there is con- This model is in agreement with the lithification-deple- siderable time lag between occupation events, or no tion cycle due to the fact that they lead to a routinisa- relationship between occupiers and re-occupiers. In tion (see below) in which alternative behaviours are effect, the lithic residue of past life may contribute to not evaluated. Rather, existing strategies were main- how learning by imitation takes place – it forms a land- tained through imitation and copying (Martínez 2002), mark, which structures future action by transforming a prompted by the material residue of previous action. space into a place (Zedeño 2000:108–111). The This line of thought enhances Kuhn’s argument process of imitation is not always one of observing about provisioning places because it accommodates people and doing as they do, but can include observ- both intentional and unintentional action over the very ing the material record, and doing as they must have long term. While it is clear that some degree of inten- done to produce that record. The latter behaviour is, of tion is needed to move the amount of lithic raw mate- course, not dependent on direct contact between peo- rial implied in this processes, the long term effects of ple or generations. The archaeological site in the past provisioning places cannot be adequately examined was not passive, but was a structurer of human action

18 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie and, through its ability to cue subsequent behaviours, produced by a change in the organisational properties could have led to increased routinisation in mobility of behaviour in the late Holocene. The large cores and and settlement patterns. Routine is produced by, and mortars stockpiled along the edges of the Quequen produces, the habitus of individuals (Bourdieu 1977). Grande River may have served as a structuring com- Yet, as noted, cores are unstable site furniture, and ponent of the social landscape by providing a new, the repletion-depletion cycle they impose itself will artificial distribution of this resource. Through the late produce long term variation in mobility patterns – con- Holocene, the strategy of provisioning places had a tinuity and change are embedded in the same recur- long-term recursive influence and was part of a social sive relationship to the landscape. phenomenon which re-organised the complex rela- Our suggestion then, is that the landscape of habit tionships between subsistence, technology, mobility can be glimpsed archaeologically in those aspects of and, settlement systems. Thus, the landscape history material culture or patterns of deposition, such as lithi- itself should be considered a potentially crucial vari- fication, which both result from and structure practical able: shaping, and being shaped by, the long-term action, in the sense of Bourdieu’s (1977) habitus. To human adaptation to the Argentine Pampas. echo Bourdieu, in the Pampas, a structured disposition Acknowledgements to stockpile lithics gave rise to a structured deposition of lithic raw material, which itself acted as a structuring The authors wish to thank Luis Borrero, Nora deposition, whose persistence in the environment influ- Flegenheimer, Cristina Bayón, Clive Gamble and enced the technological organisation and settlement Gustavo Politis for comments and suggestions made patterns of subsequent generations. on a draft version of this paper. We are grateful to INCUAPA (Department of Archaeology, Facultad de 6 Conclusions Ciencias Sociales, UNCPBA) for supporting the field- The concept of landscape has been recently dis- work and research. This paper was made possible cussed and different views of it have been proposed thanks to a programme of cooperation between the (Gosden & Head 1994; Pickering 1994; also see Universities of La Plata and Southampton, which was papers in Ucko & Layton 1999, amongst others). The funded by The British Council and Fundación principle that humans seldom are in a position to occu- Antorchas. The programme, Comparative Study of py truly pristine environments, and, particularly, that Hunter-Gatherers of Palaeolithic Europe and the hunter-gatherer societies are not adapted to a neutral Pampean Region (Argentina), was directed by niche, but actively (if not intentionally) engage in con- Gustavo Politis and Clive Gamble. Many thanks to structing that niche, should become keys for better Mariano Bonomo for helping us with the analysis of understanding hunter-gatherer landscapes. We pro- the lithic collection. We acknowledge the lifetime of pose that the lithification of the valley bottoms of the avocational archaeology performed by Mr Gesué middle stream of the Quequén Grande River illus- Noseda, whose methodical and serious work provid- trates how the environment can be built and changed ed the excellent data for this paper, and for Pampean by hunter-gatherer social practice. Although multiple archaeology. We also thank the Museo de Ciencias factors could have been involved, the most important Naturales del Club de Pesca de Lobería for their sup- point is that the lithic resource distribution over the port in this project. landscape was not a wholly natural pattern, but was

Notes 1 When orthoquartzites are mentioned they refer to Sierras Bayas Geological Formation. 2 By examining the guanaco bone assemblages from Zanjon Seco 2 and 3 Politis (1984) concluded that guanaco were killed nearby. 3 The ‘garbage structure’ of Zanjón Seco 3 shows a pattern of intra-site variability which suggests planning for longer periods of occupation, an idea also reinforced by the previously mentioned distribution and quantity of grindstones. 4 Binford (1980) and Kelly (1983) only discuss the movement of food resources when discussing logistical and residential strategies. In this regard, we consider critical resources on the Pampas to include non-food items, such as lithic raw material and fuel. Although Binford (1979:273) suggests that procurement of raw material is normally incidental to the ‘execution of basic subsistence tasks’, we are not presently concerned with what tasks are embedded in which. Rather, we are interested in the effects of the collection of raw material and its patterns of distribution across the land- scape. 5 Note that the Australian (Webb 1993) and Saharan (Close 1996) lithification examples also went beyond immediate needs and left large quan- tities of ‘wasted’ (surplus) material across the landscape.

Before Farming 2003/4 (1) 19 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

References Ames, K 1994. The Northwest Coast: Complex Hunter-Gatherers, Ecology, and Social Evolution. Annual Review of Anthropology 23:209–229.

Andrade Lima, T. & López Mazz, J in press. La Emergencia de Complejidad entre los Cazadores-Recolectores de la Costa Atlántica Meridional Sudamericana. Boletín de Arqueología Americana.

Andrefsky, W 1991. Inferring Trends in Prehistoric Settlement Behaviour from Lithic Production Technology in the Southern Plains. North American Archaeology 12 (2):129–144.

Andrefsky, W 1994. Raw-Material availability and the Organisation of the technology. American Antiquity 59 (1):21–34.

Bailey, R, & Headland, T 1991.Tropical Rain Forest. Is it a productive environment for human foragers? Human Ecology 19 (2):261–285.

Bamforth, D 1990. Settlement, Raw Material, and Lithic Procurement in the Central Mojave Desert. Journal of Anthropological Archaeology 9:70–104.

Barna, M & Kain, S 1994. Una fuente potencial de aprovisionamiento lítico en el Cerro El Sombrero. Partido de Lobería (Provincia de Buenos Aires). Actas y Memorias del XI Congreso Nacional de Arqueología Argentina. Resúmenes. Revista del Museo de Historia Natural de San Rafael. Tomo XIV (1/4). San Rafael: Mendoza:206–207

Barrientos, G 1997. Nutrición y dieta de las poblaciones aborígenes prehispánicas del sudeste de la Región Pampeana. Unpublished PhD dissertation. Facultad de Ciencias Naturales y Museo de La Plata, Universidad Nacional de La Plata, La Plata.

Barros, P & Messineo, P 2003. Identificación y aprovisionamiento de chert o ftanita en la cuenca superior del arroyo Tapalqué (Pdo de Olavarría, Pcia De Buenos Aires, Argentina). In Seelenfreund, A, Stern, C & Civalero, MT (coords) Identificación, aprovechamiento e intercambio de materias primas: enfoques teóricos y metodológicos y su aplicación a estudios regionales. Paper presented at 51° Congreso Internacional de Americanistas, Santiago de Chile, Chile.

Basgall, ME, 1987. Resource intensification among hunter-gatherers: acorn economies in prehistoric California. Research in Economic Anthropology 9, 198:21–52.

Bayón, C & Zavala, C 1994. Coastal sites in southern Buenos Aires. A review of ¨Piedras Quebradas¨. In Rabassa, J & Salemme, M (eds) Quaternary of South America and Antarctic Peninsula 10:229–253.

Bayón, C, Flegenheimer, N, Valente, M, & Pupio, A 1999. Dime cómo eres y te dire de donde vienes: proceden- cia de rocas cuarcíticas en la Región Pampeana. Relaciones de la Sociedad Argentina de Antropología XXIV:187–217.

Bayón, C & Flegenheimer N 2003. Aquí todo viene de afuera. Cambio de planes a través del tiempo para el trasla- do de roca. In Seelenfreund, A, Stern, C & Civalero MT (coords) Identificación, aprovechamiento e intercambio de materias primas: enfoques teóricos y metodológicos y su aplicación a estudios regionales. Paper presented at 51° Congreso Internacional de Americanistas, Santiago de Chile, Chile.

20 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

Beckwith, B 2002. Colonial Eden or Indigenous Cultivated Landscape? Reconstructing nineteenth century camas meadows in southern Vancouver Island. In Garry Oak Ecosystem Restoration: Progress and Prognosis. Victoria, Canada: Victoria Chapter of the Centre for Ecological Restoration:64–72

Bender, B 1981. Gatherer-Hunter Intensification. In Sheridan, R & Bailey, R (eds) Economic Archaeology, British Archaeological Reports, International Series 96:149–157.

Berón, M, & Politis, G 1997. Arqueología Pampeana en la Década de los ‘90. Análisis y perspectivas. In Berón, M & Politis, G. (eds) La Arqueología de la Región Pampeana en la Década de los ’90. San Rafael: Museo de Historia Natural, INCUAPA (Investigaciones Arqueológicas y Paleontológicas del Cuaternario Pampeano),FACSO (Facultad de Ciencias Sociales), UNICEN (Universidad Nacional del Centro de la Provincia de Buenos Aires):7–32.

Binford, L 1979. Organisation and Formation Processes: Looking at Curated Technologies. Journal of Anthropological Research 35(3):255–273.

Binford, L 1980. Willow smoke and dogs tails: Hunter-gatherer settlements systems and archaeological site for- mation. American Antiquity 45 (1):4–20.

Binford, L 1982. The archaeology of place. Journal of Anthropological Archaeology 1 (1):5–31.

Binford, L 1983. In Pursuit of the Past: decoding the Archaeological Record. London:Thames and Hudson.

Binford, L 1987. Researching Ambiguity: Frames of Reference and Site Structure. In Kent, S (ed) Method and Theory for Activity Area Research: An Ethnoarchaeological Approach. New York: Columbia University Press:449–512.

Binford, L 1990. Mobility, Housing and Environment: A Comparative Study. Journal of Anthropological Research 46 (2):119–152

Binford, L & O’Connell, J 1989. An Alyawara day: the stone quarry. Debating Archaeology. Studies in Archaeology. New York: Academic Press:121–146

Bleed, P 1986. The optimal design of hunting weapons: Maintainability or Reliability. American Antiquity 51 (4):737–747.

Bonadonna, F, Leone, G, & Zanchetta G 1995. Composición isotópica de los fósiles de gasterópodos continen- tales de la Provincia de Buenos Aires. Indicaciones paleoclimáticas. In Alberdi, MT, Bonadonna, FP & Tonni EP (eds) Registro Continental de la Evolución Climática y Biológica de los Ultimos 5 Ma por Correlación entre el Hemisferio Norte (SO de Europa) y el Hemisferio Sur (Argentina), Madrid: Monografías Museo Nacional de Ciencias Naturales:75–104.

Bourdieu, P 1977. Outline of a Theory of Practice. Cambridge: Cambridge University Press.

Boyd, R, & Richerson, P 1985. Culture and the Evolutionary Process. University of Chicago Press. Chicago.

Boyd, R (ed) 1999. Indians, Fire and the Land in the Pacific Northwest. Corvallis: Oregon State University Press.

Bracco, R, Cabrera, L & López, JM 2000. La prehistoria de las tierras bajas de la cuenca de la laguna Merín. In

Before Farming 2003/4 (1) 21 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

Duran, A & Bracco R (eds) Simposio Internacional de arqueología de las tierras bajas. Montevideo:Ministerio de Educación:13–38.

Broughton, JM 1994. Declines in Mammalian Foraging Efficiency During the Late Holocene, San Francisco Bay, California. Journal of Anthropological Archaeology 13 (4):371–401

Cabrera, A 1976. Regiones Fitogeográficas . Enciclopedia Argentina de Agricultura y Jardinería 2 (1):85.

Cabrera, L 1999. Funebria y Sociedad entre los constructores de Cerritos del Este uruguayo. In López, JM & Sans, M (eds). Arqueología y bioantropología de las tierras bajas. Montevideo: FHCE/Universidad de la República: 63–78.

Camilli, E 1989. The occupational history of sites and the interpretation of prehistoric technological systems: an example from Cedar Mesa, Utah. In Torrence, R (ed) Time, energy and stone tools. New Directions in Archaeology. Cambridge: Cambridge University Press:17–26

Camilli, E, & Ebert, J 1992. Artifact Reuse and Recycling in Continuous Surface Distributions and Implications For Interpreting Land Use Patterns. In Rossignol, J & Wandsnider, L (eds) Space, Time and Archaeological Landscapes. New York and London: Plenum Press:113–133

Carr, C 1994. Technological Organization and Prehistoric Hunter-Gatherer Mobility: Examination of the Hayes Site. In Carr, C (ed) The Organization of North American Prehistoric Chipped Stone Tool Technologies. Ann Arbor, Michigan:International Monographs in Prehistory. Archaeological Series 7:35–44.

Cavalli-Sforza, L & Felman, K 1981. Cultural Transmission and Evolution. A Quantitative Approach. Princeton: Princeton University Press.

Close, A 1996. Carry that Weight: The use and transportation of stone tools. Current Anthropology 37(3):545–553.

Crown, PL & Wills, WH 1995. Economic Intensification and the Origins of Southwestern Pottery. In Barnett, W & Hoops, J (eds) The Emergence of Pottery. Washington: Smithsonian Institution Press

Denevan, W 1992. The Pristine Myth: The Landscape of the Americas in 1492. Annals of the Association of American Geographers 82: 369–385.

Deur, D. 2000. A Domesticated Landscape: Native American Plant Cultivation on the Northwest Coast of North America. Unpublished PhD Dissertation, Department of Geography and Anthropology, Louisiana State University.

Deur, D & Turner, N (eds) in press. “Keeping it Living”: Indigenous plant management on the Northwest Coast. Seattle: University of Washington Press.

Dibble, H 1987. The interpretation of Middle Paleolithic scraper morphology. American Antiquity 52:109–117.

Dunnell, R & Dancey W 1983.The Siteless Survey: a Regional Scale Data Collection Strategy. Advances in Archaeological Method and Theory 6:267–287.

Dunnell, R & Simek, JF 1995. Artifact Size and Plowzone Processes. Journal of Field Archaeology (22):305–319

22 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

Durham, W 1991. Coevolution: Genes, Culture, and Human Diversity. Stanford, California: Stanford University Press.

Ebert, J 1992. Distributional Archaeology. Albuquerque, USA: University of New Mexico Press.

Flegenheimer, N 1991. La Liebre, un sitio de cantera-taller. Boletín del Centro 2:58–64.

Flegenheimer, N, Kain, S, Zárate, M & Barna, A 1996. Aprovisionamiento de Cuarcitas en Tandilia. Las canteras del Arroyo Diamante. Arqueología 6:117–141.

Flegenheimer, N, Zárate M & Valente, M 1999. El área de canteras Arroyo Diamante, Barker, Sierras de . In Diez Marín, C, (ed) Actas del XII Congreso Nacional de Arqueología Argentina. La Plata: Facultad de Ciencias Naturales y Museo de la Plata, Universidad Nacional de La Plata Tomo III:134–138.

Flegenheimer, N & Bayón, C 2002. Cómo, cuándo y dónde? Estrategias de abastecimiento lítico en la Pampa Bonaerense. In Mazzanti, D, Berón M & Oliva, F (eds) Del Mar a los Salitrales. Diez mil Años de Historia Pampeana en el Umbral del Tercer Milenio. La Plata: Universidad Nacional de Mar del Plata, Facultad de Humanidades:231–241.

Franco, N 1991. Algunas tendencias distribucionales en el material litico recuperado en el Area Interserrana Bonaerense. Boletin del Centro Nro 3:72–79.

Franco, N 1994. Maximización en el aprovechamiento de los recursos líticos. Un caso analizado en el Area Interserrana Bonaerense. In Lanata, JL & Borrero, L (eds) Arqueología de Cazadores-Recolectores. Limites, casos y aperturas. Edición especial, Arqueología Contemporánea 5:75–88.

Gamble, C 1995. Making Tracks: Hominid networks and the evolution of the social landscape. In Steele, J and Shennan, S (eds) The Archaeology of Human Ancestry: Power, sex and tradition. London: Routledge: 253–278

Gnecco, C 2000. Ocupación Temprana de Bosques Tropicales de Montaña. Serie Estudios Sociales, Universidad del Cauca, Colombia.

Gnecco, C 2003. Ecological Reductionism and the Initial Peopling of South America. Paper presented at session N2 First People in the Americas (Politis, G, coord). The Fifth World Archaeological Congress, Washington DC. EE.UU.

Gosden, C 1994. Social Being and Time. Oxford: Blackwell.

Gosden, C & Head, L 1994. Landscape – a usefully ambiguous concept. Archaeology in Oceania 29:113–116.

Gould, SJ & Vrba, ES 1982. Exaptation: A missing term in the science of form. Paleobiology 8(1):4–15

Hutterer, K 1982. Interaction Between Tropical Ecosystems and Human Foragers: Some General Consideration. Working Paper, Environment and Policy Institute, East-West Center, Honolulu, Hawaii.

Ingold, T 1993. The Temporality of the Landscape. World Archaeology 25(2):152–174.

Janetski, JC 1997. Fremont hunting and resource intensification in the eastern Great Basin. Journal of Archaeological Science 24 (12):1075–1088

Before Farming 2003/4 (1) 23 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

Johnson, L 1994. Aboriginal burning for vegetation management in northwest British Columbia. Human Ecology 22(2):171–188.

Kelly, R 1983. Hunter-gatherer mobility strategies. Journal of Anthropological Research 39(3): 277–306.

Kelly, R 1992. Mobility/Sedentism: Concepts, Archaeological Measures, and Effects. Annual Review of Anthropology 21:43–66.

Kelly, R 1995. The Foraging Spectrum: Diversity in Hunter-Gatherer Lifeways. Washington and London: Smithsonian Institution Press.

Kuhn, S 1994. A Formal approach to the design and assembly of transported toolkits. American Antiquity 59:426–442.

Kuhn, S 1995. Mousterian Lithic Technology. An Ecological Perspective. Princeton: Princeton University Press.

Lewontin R 1982. Organism and Environment. In Plotkin, H (ed) Learning, Development, and Culture. Chichester: Wiley:151–170.

Lewis, H 1982. Fire technology and resource management in aboriginal North America and Australia. In Williams, N & Hunn, E (eds) Resource Managers: North American and Australian Hunter-Gatherers. Boulder, Co: Westview Press:45–67

Llambías, E & Prozzi C 1975. Ventania. Relatorio de Geología de la Provincia de Buenos Aires. Actas del VI Congreso de Geología Argentina:79–101.

López Mazz, JM 2001. Las estructuras tumulares (Cerritos) del litoral atlántico uruguayo. Latin American Antiquity 12 (3):231–255.

Lourandos, H 1985. Intensification and Australian Prehistory. In Price, T & Brown, J (eds) Prehistoric Hunter- Gatherers: The Emergence of Cultural Complexity. New York: Academic Press:385– 423.

Lourandos, H 1997. Continent of Hunter Gatherers. New Perspectives in Australian Prehistory. Cambridge: Cambridge University Press.

Lozano, P 1991. Cerro Aguirre: un sitio de aprovisionamiento de materia prima lítica en la localidad de Sierras Bayas (Pcia de Buenos Aires). Actas del X Congreso Nacional de Arqueología Argentina. Shincal 3(3):145–150.

Mackie, Q 2001 Settlement Archaeology in a Fjordland Archipelago: Network Analysis, Social Practice and the Built Environment of Western Vancouver Island, British Columbia, Canada since 2,000 BP. Oxford: BAR International Series.

Madrid, P & Salemme, M 1991. La ocupación tardía del sitio 1 de la Laguna Tres Reyes, Adolfo González Chaves, Pcia de Buenos Aires, Boletín del Centro 3:165–179.

Martínez, G 1999. Tecnología, subsistencia y asentamiento en el curso medio del Río Quequén Grande: un enfoque arqueológico. Unpublished PhD dissertation. Facultad de Ciencias Naturales y Museo de la Plata, Universidad Nacional de La Plata, La Plata.

Martínez, G 2002. Organización y cambio en las estrategias tecnológicas. Un caso arqueológico e implicaciones

24 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie conductuales para la evolución de las sociedades cazadoras-recolectoras Pampeanas. In Martínez, G & Lanata J (eds) Perspectivas Integradoras entre Arqueología y Evolución. Teoría, Métodos Casos de Aplicación. Serie Teórica del INCUAPA (Investigaciones Arqueológicas y Paleontológicas del Cuaternario Pampeano) , Vol 1 FACSO–UNICEN (Facultad de Ciencias Sociales, Universidad Nacional del Centro de la Provincia de Buenos Aires):121–156.

Martínez, G 2003. Late Holocene Intensification and Construction of Landscape in Hunter-gatherers of the Southern Hemisphere. Paper presented to the Session Case studies and conceptual concerns among southern hemisphere gatherer-hunter-foragers societies. Empowerment and Exploitation: South-south & south-north archaeological encounter. World Archaeological Conference 5 (Washington DC, USA).

Martínez, G & Gutiérrez, M in press. Tendencias en la explotación humana de la fauna durante el Pleistoceno final- Holoceno en la Región Pampeana (Argentina). In Mengoni Goñalons, GL (ed) Zooarchaeology of South America. Oxford: BAR.

Mazzanti, D & Quintana C (eds) 2001. Cueva Tixi: cazadores y recolectores de las sierras de Tandilia oriental. Geología, Paleontología y Zooarqueología. Laboratorio de Arqueología. Universidad Nacional de Mar del Plata. Publicación Especial 1.

Miotti, L & Salemme, M 1999. Biodiversity, taxonomic richness and specialists-generalists during Late Pleistocene/Early Holocene times in Pampa and Patagonia (Argentina, Southern South America) Quaternary International 53/54:53–68.

Neiman, F & Monticello V 1997. Conspicuous Consumption as Wasteful Advertising: a Darwinian Perspective on Spatial Patterns in Classic Maya Terminal Monument Dates. In Barton, C & Clark, C (eds) Rediscovering Darwin: Evolutionary Theory and Archaeological Explanation. Archaeological Papers of the American Anthropological Association No 7:267–290.

Nelson, M 1991. The Study of Technological Organisation. Archaeological Method and Theory 3:57–100.

Odell, G 1996. Economising Behavior and the Concept of “Curation”. In Odell, G (ed) Stone Tools. Theoretical Insights into Human Prehistory. New York and London: Plenum Press:51–79.

Oliva, F & Moirano, J 1997. Primer informe sobre aprovisionamiento primario de riolita en Sierra de la Ventana (Provincia de Buenos Aires, Argentina). In Berón, M & Politis G (eds) La Arqueología de la Región Pampeana en la Década de los ´90. Estado de las Investigaciones y Perspectivas. Investigaciones Arqueológicas y Paleontológicas del Cuaternario Pampeano (INCUAPA), Universidad Nacional del Centro de la Provincia de Buenos Aires. Museo de Historia Natural de San Rafael. Mendoza:137–146.

Orcutt, J 1986. Site Surface Characteristics and Functional Inferences. American Antiquity 51 (2):296–312.

Ormazábal, P 1994. Estrategias Alimentarias en la Región Pampeana: Una Aproximación desde la Arqueología y la Etnohistoria en el Procesamiento y Almacenamiento de Alimentos. Unpublished BA dissertation. Facultad de Ciencias Sociales de Olavarría, Universidad Nacional del Centro de la Provincia de Buenos Aires.

Parry, W & Kelly, R 1987. Expedient Core Technology and sedentism. In Johnson, J & Morrow, C (eds) The organ- ization of Core Technology. Boulder, CO: Westview Press:285–309

Pickering, M 1994. The physical landscape as a social landscape: a Garawa example. Archaeology in Oceanía 29:149–161.

Before Farming 2003/4 (1) 25 Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

Politis, G 1984. Arqueologia del Area Interserrana Bonaerense. Unpublished PhD Dissertation. Facultad de Ciencias Naturales y Museo de la Plata, Universidad Nacional de La Plata, La Plata.

Politis, G 1996a. Moving to Produce: Nukak Mobility and Settlement Patterns in Amazonia. World Archaeology 27 (3):492–511.

Politis, G 1996b. Los Nukak. Bogota, Colombia: Instituto de Investigaciones Amazonicas Sinchi.

Politis, G & Salemme, M 1990. Pre-Hispanic mammal exploitation and hunting strategies in the eastern Pampa subregion of Argentina. In Davis, L & Reeves, B (eds) Hunters of the recent Past. London: Unwin-Hyman.

Politis, G & Madrid, P 2001. Arqueología Pampeana: Estado Actual y Perspectivas. In Berberian, E & Nielsen, A (eds) Historia Argentina Prehispánica. Córdoba: Editorial Brujas, Tomo II:737–814.

Politis G, Martínez, G & Bonomo, M 2001. Alfarería temprana en sitios de cazadores recolectores de la Región Pampeana (Argentina). Latin American Antiquity 12 (2):167–181.

Politis, G, Bonomo, M & Prates, L 2003. Territorio y movilidad entre la costa atlántica y el interior de la Región Pampeana (Argentina). Porto Alegre, Brasil: Revista de Estudios-Iberoamericanos XXIX (1):11–35.

Potts, R 1991 Why the Oldowan? Plio-Pleistocene toolmaking and the transport of resources. Journal of Anthropological Research 47(2):153–76

Price, T & Brown, J 1985. Aspects of Hunter-Gatherer Complexity. In Price, T & Brown J (eds) Prehistoric Hunter- Gatherers: The Emergence of Cultural Complexity. New York: Academic Press:3–20.

Prieto, A 1996. Late Quaternary Vegetational and Climatic Changes in the Pampa Grassland of Argentina. Quaternary Research 45:73–88.

Pupio, A 1996. Resultados preliminares del análisis del material faunístico del sitio cantera-taller La Liebre. Jornadas Chivilcoyanas en Ciencias Sociales y Naturales:191–194.

Richerson, P & Boyd, R 1992. Cultural Inheritance and Evolutionary Ecology. In Smith , E & Winterhalder B (eds) Evolutionary Ecology and Human Behavior. New York: Aldine de Gruyter: 61–92.

Ringuelet, R 1955. Panorama Zoogeográfico de la Provincia de Buenos Aires. Notas del Museo de la Plata. Zool 18 (156):1–45.

Rosenberg, M 1994. Pattern, Process, and Hierarchy in the Evolution of Culture. Journal of Anthropological Archaeology 13:307–340.

Schlanger, S 1992. Recognizing Persistent Places in Anasazi Settlement Systems. In Rossignol, J & Wandsnider. L (eds) Space, time and Archaeological Landscapes. New York and London: Plenum Press:91–112.

Shott, M 1986. Technological Organisation and Settlement Mobility: An Ethnographic Examination. Journal of Anthropological Research 42:15–52.

Soriano, A 1992. Río de La Plata Grasslands. In Coupland, R (ed) Ecosystems of the World 8a, Natural Grasslands. Introduction and Western Hemisphere. Amsterdam: Elsevier:367–413.

26 Before Farming 2003/4 (1) Late Holocene human occupation of the Quequén Grande River valley bottom: Martinez & Mackie

Stafford, C & Hajic, E 1992. Landscape scale. Geoenvironmental Approaches to Prehistoric Settlement Strategies. In Rossignol, J & Wandsnider, L (eds) Space, time and Archaeological Landscapes. New York and London: Plenum Press:137–161.

Sullivan, A 1995. Artifact Scatters and Subsistence Organization. Journal of Field Archaeology (22):49–64.

Tonni, E, Cione, A & Figini, A 1999. Predominance of arid climates indicated by mammals in the Pampas of Argentina during the Late Pleistocene and Holocene. Palaeogeography, Palaeoclimatology, Palaeoecology 147:257–281.

Torrence, R (ed) 1989. Time, Energy and Stone tools. New York: Cambridge University Press.

Torrence, R 1993. Strategies for Moving on in Lithic Studies. In Carr, P (ed) The organisation of North American Prehistoric Chipped Stone Tool Technologies. Ann Arbor, Michigan: International Monographs in Prehistory. Archaeological Series 7:123–136.

Turner, N. 1991. Burning mountain sides for better crops: aboriginal landscape burning in British Columbia. Archaeology in Montana 32(2):57–73.

Ucko, P & Layton, R 1999 (eds). The Archaeology and Anthropology of Landscape. Shaping your landscape. London and New York: Routledge One World Archaeology Series, 30.

Webb, K 1993. The lithification of a sandy environment. Archaeology in Oceania 28:105–111.

Williams, E 1987. Complex Hunter-Gatherers: A View from Australia. Antiquity 61:310–321.

Wilson, E.O. 1975. Sociobiology: The new synthesis Cambridge, MA: Harvard University Press.

Wohlgemuth, E 1996. Resource intensification in prehistoric central California: evidence from archaeobotanical data. Journal of California and Great Basin Anthropology. 18 (1):81–103.

Zárate, M, Kemp, R, Espinosa, M & Ferrero, L 2000. Pedosedimentary and palaeoenvironmental significance of a Holocene alluvial sequence in the Southern Pampas, Argentina. The Holocene 10 (4):481–488.

Zedeño, MN 2000. On what people make of places: a behavioural cartography. In Schiffer, M (ed) Social Theory in Archaeology. Salt Lake City: University of Utah Press:97–111

Before Farming 2003/4 (1) 27