Abstract
Sunken features backfilled with domestic refuse represent the prevailing depositional context-type in later prehistory worldwide. Despite being so, this evidence remains poorly understood and has only received sporadic attention, chiefly within Anglophone archaeologies. This paper focuses on ceramics from a suite of such intricate contexts (cut features, burials, settlements, barrows) from Iberia in a diachronic and comparative perspective, from the Early Neolithic to the Late Bronze Age (5500–1100 BC). A total of 10,800 potsherds were examined with a taphonomic and refitting protocol attentive to formation dynamics and tracking intentionality. Results suggest that most of the studied assemblages are unplanned by-products of social life. From the earliest pottery-using communities, habitual actions conditioned the eventual preservation of the extant archaeological record. Fragmentation and deposition were key social practices, ultimately representing enduring trans-cultural phenomena. This research challenges uncontested interpretive premises, namely the ‘reflectionist’ standpoint, and disproves consensual and undue concepts frequently used in mainstream accounts of later prehistory.
Keywords
Introduction
Archaeological research is based on established scholarly principles. Some of them, once criticised (e.g. Brück, 1999), are sensible and helpful, whereas others remain uncontested and pervaded by ethnocentric and positivist ideas. Among the latter prejudices, the ‘reflectionist’ standpoint (Chapman and Gaydarska, 2007: 71) is especially widespread and resilient. This misleading perspective is encapsulated in the ‘Pompeii premise’ (Schiffer, 1985), i.e. a view of the archaeological record as a faithful, unproblematic reflection of past actions and values passively fossilised at discrete slices of time (Lucas, 2012: 102–103). A series of fallacious expectations follow from this ‘reflectionist’ attitude: (a) the existing evidence supposedly mirrors lifestyles and original equipment while in use; (b) breakage and incompleteness are envisaged as unavoidable side effects of sampling bias, or natural features irrespective of depositional context; and (c) beyond the funerary/ritual realm, the contribution of past intentionality to the formation of the material record is considered negligible and can be downplayed. Recent research on depositional practices and the formation dynamics of the archaeological evidence (e.g. Beadsmoore et al., 2010; Garrow, 2006; Mills and Walker, 2008; Rídký et al., 2014) has shed important new light to start reappraising such long-held assumptions. Besides this, in recent years an increasing array of archaeological contributions addressing material culture in social terms have emerged, with novel, ‘bottom-up’ approaches gaining ground. Depositional practices, materiality, the social lives or itineraries of things, and the mutualistic relationships between humans and things encapsulate this new agenda (e.g. Gosden and Marshall, 1999; Hodder, 2012; Jennings, 2014; Knappett, 2005; Meskell, 2004; Mills and Walker, 2008; Thomas, 1999).
Enquiry on later prehistory in the Iberian Peninsula, from the Neolithic through the Bronze Age (5500–850 cal BC), has been very uneven. Generalist ‘top-down’ accounts (Gamble, 2001: 78–81) are the mainstream and have given rise to large-scale narratives in terms of gradual processes of politico-economic development (Cruz Berrocal et al., 2013). In the better-known south-east, well-preserved tombs with rich furnishings and deeply stratified settlements are available. Sophisticated assessments have been gained and scholars are now aware that ‘contextual, depositional and taphonomic factors have not been appropriately considered’ (Aranda et al., 2015: 75). By contrast, archaeology in the northern half of Iberia yields an alternate predominance of either funerary or domestic evidence, chiefly associated with humble pits and gullies (Colomer et al., 1998). Such sunken features represent the prevailing later prehistoric evidence worldwide, but they have been too often overlooked. For the Iberian inner plateau or Meseta (Figure 1), the ambiguity and evasiveness of its empirical basis have ultimately reinforced a scholarly empiricist and ‘reflectionist’ background. Some authors (e.g. Díaz-del-Río, 2001: 118–125; Martínez Navarrete, 1988: 886–888) stressed the prospects of studying formation processes as a crucial way for understanding how this archaeological record was configured and what sort of social practices might have been behind it. Interest on these issues has lately escalated and such research questions are now in the Iberian research agenda (Liesau et al., 2014b; Márquez-Romero and Jiménez-Jáimez, 2013; Vale, 2010; Weiss-Krejci, 2011).
Central Iberia and sites mentioned in the text. 1. Molino Sanchón (Zamora); 2. San Román de Hornija (Valladolid); 3. Pico Castro (Palencia); 4. El Cerro (Burgos); 5. La Lámpara (Soria); 6. El Alto III (Soria); 7. Camino de las Yeseras (Madrid); 8. El Ventorro (Madrid); 9. Humanejos (Madrid); 10. Valdeprados (Avila); 11. Los Tiesos I (Avila); 12. Fuente Lirio (Avila); 13. El Morcuero (Avila).
Framed by this intellectual backdrop, a two-year post-doctoral research project 1 was designed to tackle the later prehistoric Iberian evidence with a fresh approach. This investigation is distinctive inasmuch as it adopts a self-critical stance to address formation processes, relies on a systematic methodology to assess taphonomy in a comparative and diachronic approach, and interprets the outcomes from a non-positivist background. Enquiry focused on hand-made ceramics, the most omnipresent archaeological material, whose informative potential has been hitherto underexploited. This paper draws on the results of that research. Its aims are twofold: (a) to contribute to the discussion of fragmentation and intentionality in the archaeological record; and (b) to test the value of careful observation of the minutiae of material culture to support complementary microscale or ‘bottom-up’ social narratives.
Fragmentation and intentionality
Over the last decade, a number of contributions have reinvigorated the original disciplinary interest in archaeological objects, and social insights are being gained from a suite of innovative perspectives. In a fairly simplistic but operational way, these can be grouped into: (a) multidisciplinary-driven approaches revolving around materiality and human–thing relationships (e.g. Hodder, 2012; Jones, 2012; Knappett, 2005; Meskell, 2004; Mills and Walker, 2008), and (b) an upturn in interest in the archaeological physicality of items (particularly ceramics) to illuminate the circumstances surrounding their entering the material record (e.g. Brudenell and Cooper, 2008; Edwards, 2012; Garrow, 2006, 2012; Jennings, 2014; Lamdin-Whymark, 2008). Thus, studies of archaeological materials may either adhere to sophisticated philosophically grounded frameworks, with the risk of overlooking how and why these items survived, or resort to geoscientific and taphonomic strategies which might disregard contingent cultural conditions. Thus, when tackling the archaeological record, there is a growing risk of disconnection between our interpretive and analytical considerations (cf. Lucas, 2012: 87–91).
At the interface of this riddling split reside fragmentation theory (Chapman, 2000, 2012; Chapman and Gaydarska, 2007) and intentionality in the archaeological evidence (David, 2004; Davis, 1992; Dornan-Fish, 2012). Fragmentation is frequently referred to in literature as a mere trope, a taken-for-granted attribute of ancient materials. However, this overwhelming condition of the subject matter of archaeology deserves proper analysis. If adequately problematized and observed, such a variable may offer valuable information on the ways past remains entered the archaeological record. From the early 2000s, the ‘fragmenterist’ strand (Chapman, 2000; Chapman and Gaydarska, 2007: 113–171) has argued for the deliberate breakage of things as a genuine social practice, traceable in diverse prehistoric societies. This theory has attempted to integrate high level social theory with detailed hands-on analysis of items (Bailey, 2001: 1181; Jones, 2012: 19; Matthews, 2007: 186). Besides that, criticism on some early assumptions (Bailey, 2001: 1182) prompted further discussion on formation dynamics (Chapman and Gaydarska, 2007: 71–79). However, a defining aspect of the deliberate fragmentation theory’s formulation remains controversial – ascertaining whether broken objects may be considered planned end products or fortuitous by-products (Bailey, 2001: 1182; Lucas, 2012: 98; Milisauskas, 2002: 889). This latter point connects with the second mentioned theme: intentionality and its archaeological diagnosis. Linking subjective mental deliberation to its physical correlates represents an unresolved challenge (Lucas, 2012: 23–29). Current standpoints (David, 2004: 67; Davis, 1992: 336–337; Russell, 2004: 64) conclude that archaeological remnants can only be partially and indirectly surmised as mediated results of past will.
In the light of this state of affairs, two important points need to be made. First, any decision with material consequences follows cultural pathways of doing, so that outcomes are somehow constrained by space–time fields of possibilities (David, 2004: 67–68; Jones, 2012: 125). Second, the original testing field for the deliberate fragmentation premise was the Balkan Neolithic, with a focus on ‘common but special objects’ (Chapman, 2012: 130) such as fired clay figurines and body shell ornaments (Chapman and Gaydarska, 2007: 113–171). The circulation of these things in daily life was far from secluded (contra Bailey, 2001: 1182), and their informative potential has been realized (Chapman and Gaydarska, 2007: 7). Nonetheless, such artifacts might have mobilized exclusive cultural precepts dealing with their formalized social termination. The heuristic value of more mundane portable objects to illuminate prehistoric sociality deserves further attention (Chapman and Gaydarska, 2007: 204). In short, the present appearance of archaeological materials might have been the result of both preconceived and casual past actions. To tell apart both possibilities, we are to integrate the more ubiquitous and accessible items in our accounts.
Estimated archaeological patterns according to three archetypical depositional scenarios.
Taphonomic results.
Neolithic engagements with pottery (c. 5500–3300 BC)
Megalithic monuments and funerary customs dated to the late fifth and fourth millennia BC (Middle Neolithic) feature highly in traditional accounts of this period, at the expense of less visible settlement archaeology. Fortunately, this panorama is improving (Díaz-del-Río, 2001; Jorge, 2014; Weiss-Krejci, 2011); narratives now cover an extended timescale and a richer variety of Neolithic depositional contexts, predominantly of early date (c. 5500–4400 BC) (Balsera et al., 2015: 141–142). These include settlements, mainly open-air and cave occupations, and mortuary expressions, such as pit burials and the deposition of dismembered human remains (Garrido-Pena et al., 2012; Weiss-Krejci, 2011).
The pit site of La Lámpara (Figure 1, no. 5) is one such lowland habitat, and testifies to activities that took place at a temporary camp repeatedly occupied throughout the second half of the sixth millennium BC (Rojo et al., 2008). Excavations affected 1% of the estimated site and unearthed 18 dug-out features (Table 2), some probably used as ground storage silos, but all eventually filled with settlement debris, including one of the earliest ceramic assemblages in Iberia. Excavators (Rojo et al., 2008) posited the key role of deliberate fragmentation and selective deposition as characteristic cultural traits in the unearthed evidence (Figure 2). Re-examination of the ceramic assemblage confirmed and pinpointed such claims. Six of the sunken features are linked by refitting sherds which indicate their coeval infilling with waste discarded shortly before (83% of the total ceramics exhibits good preservation) (Table 2). The high proportion of small- to medium-sized sherds (78%) may reflect prolonged trampling over the course of reoccupations. The burial of an old woman in Pit 1 (Figure 2a) involved certain formalities (Rojo et al., 2008: 81–86, 379–393): the body was accompanied by a jug exhibiting a sharp fracture and a missing neck. This jug features incised decorations depicting a bearded human face or facial tattoo, enhanced by its fracture (Rojo et al., 2008: 377–379) (Figure 2b). Its breakage pattern, with a neat rupture affecting the whole upper part and lacking any cracks or signs of impact in the heavier bottom, suggests deliberate fragmentation. This vessel was likely regarded as a meaning-laden thing, purposefully divided and then carefully placed at the woman’s feet (Figure 2a). This was not the only intentional action recognized there. The layers covering the corpse yielded 246 potsherds showing a ‘freshly broken’ appearance. The refitting operation indicates 126 secure or probable matching sherds belonging to at least 21 partial vessels, probably broken on occasion of the mortuary ceremony. No irrefutable proofs for this possibility were identified.
Early Neolithic site of La Lámpara. (a) Funerary ritual in Pit 1, note participants breaking vessels and deposition of the jug (Rojo et al., 2008: 84). (b) Jug featuring facial human traits with missing neck from Pit 1 (after Rojo et al., 2008: figs. 186 and 187).
Petrographic analyses (Blanco-González et al., 2014) led to a more accurate diagnosis of several probable matching pairs, which provide information on their temporality and allow raising new interpretations. In Pit 13, fragments exhibiting contrasting post-breakage alterations (reuse, weathering, burning) belonged to a common vessel (Figure 3a). Thus, they underwent diverse degradation processes for a prolonged period before they entered this feature. Another two large slabs bearing applied decoration (Figure 3b) and sensibly published as parts of the same coarse jar (Rojo et al., 2008: 140, Figure 115, no. 1) actually derived from two parent jars. Such portions, sharing eye-catching features (unique in the excavated sector), hardly passed unnoticed to people disposing of them. Their joint occurrence within Pit 13 may be haphazard, but the possibility of their entering the same well because of their formal resemblance is also worth considering; perhaps such large chunks were piled and handled together, or maybe they had some meaning and were deemed to finish together.
Early Neolithic sherds from Pit 13 at La Lámpara. (a) Fragments M & N from the same grooved bowl showing divergent appearance. (b) Similar looking potsherds O & P with applied ropes from two different parent jars.
In short, beyond the mortuary realm, there is scope for considering patterning and human agency among the most frequent physical correlates of social life in the Early Neolithic. It has been suggested (Rojo et al., 2008) that the wandering semi-mobile settlement patterns characterized in these temporary camps may match a rationale akin to that held by subsequent Neolithic communities in other European areas (Thomas, 1999). In this light, the formal closure of pits and the showcased peculiarities in the treatment of ceramics might be envisaged as mnemonic resources aimed at marking the landscape at important social episodes.
Depositional stories in the Chalcolithic (c. 3300–2200 BC)
After a millennium (c. 4400–3400 BC) of scant activity (Balsera et al., 2015: 141–142), the third millennium BC in Iberia left a much more visible, and hence investigated, archaeology. The evidence is now far more copious and apparently more explicit in utilitarian terms, and this has facilitated an overindulgent and barely critical ‘reflectionist’ attitude. A varied array of testimonies is known: funerary contexts (e.g. megaliths, tumuli, tholoi, charnel-houses, inhumations in natural and artificial caves, burial pits, etc.), abundant open-air settlements, rock art, and interspersed ditched enclosures (Bueno et al., 2008; Díaz-del-Río, 2001, 2004; Weiss-Krejci, 2011). Some of these contexts are examined here to better understand their archaeological nature.
Fuente Lirio (Figure 1, no. 12) is a Chalcolithic homestead (Fabián, 2006: 128–155). Over 2800 potsherds found within a round hut 6 m in diameter were characterised with the taphonomic protocol (Table 2). At least 140 serving bowls, cooking pots and coarse jars were identified, generally very well preserved (fresh category: 92%), but also featuring eroded sherds (8%), numerous small–medium pieces (59%) and 52% are ‘orphan’ fragments (Table 2). These values do not fit those expected in an abruptly sealed dwelling akin to a ‘Pompeii-like’ context filled with de facto refuse, i.e. still usable items abandoned in situ (Porčić, 2012: 25--37; Schiffer, 1985; cf. Table 1). Ceramics yielded by this hut were more likely everyday waste of a household disposing of its turnover for a relatively short period (<10 years) in the vicinity (hence their low abrasion degree), subsequently partly retrieved (high number of ‘orphan’ sherds) and dumped into the structure during the abandonment stage (cf. LaMotta and Schiffer, 1999: 22–24).
This domestic image vividly contrasts with the so-called ‘Pithouse 013’ at El Ventorro (Figure 1, no. 8). This extraordinarily rich sunken feature contained 39,700 items in 44 m2, including ceramics, flints, faunal debris, querns, copper crucibles and slag, and a few human remains (Díaz-del-Río, 2006: 73). A subset of nearly 4000 potsherds was assessed, including 111 fragments from at least 48 Bell Beakers (Table 2). ‘Pithouse 013’ was initially interpreted as a slow build-up of occupation floors regularly cleaned; if this were the case, only scarce and tiny residues were to be expected (LaMotta and Schiffer, 1999: 21). Our reassessment (Blanco-González and Chapman, 2014b) does not support such a reading. Taphonomic values obtained from this assemblage (Table 2) are comparable to those from the fills of Neolithic monumental ditches (Rídký et al., 2014: 595–596) and point to a ditch segment filled with a bulky and heterogeneous mixture of items with a prolonged temporality, reaching back up to several decades. This detritus included recently broken remains (60% are fresh sherds) and secondary small (55%) and erratic residues (39% are slightly eroded) probably interred during social gatherings. This contrasts with Fuente Lirio, where 7% are small potsherds and only 5% are slightly eroded (Table 2). Importantly, comparisons were made between coarse and supposedly highly valuable pottery. Beaker drinking bowls and beakers, and large carinated bowls for solid foodstuff were deployed in those commensality ceremonies (Garrido-Pena et al., 2011), broken and immediately discarded (large sizes, negligible erosion). However, a quarter of the Beaker sherds, including those featuring schematic designs regarded as highly symbolic (Figure 4a), underwent the same degradation processes as the plain coarse wares. Thus, at this terminal stage, such things did not receive any distinctive treatment (Blanco-González and Chapman, 2014b: 95).
(a) Highly abraded Beaker potsherds from El Ventorro showing a schematic sun and a deer's antlers. (b) Large Bell Beaker smashed on deposition at Molino Sanchón (after Abarquero et al., 2012: lam. 164 and Guerra-Doce et al., 2011: fig. 9). Scales in cm.
Beyond the expected placement of entire Bell-Beaker vessels in funerary contexts (e.g. Bueno et al. 2008), an array of ‘oddities’ is being recognised during the third millennium BC, such as the smashing-on-deposition (Jones, 2012: 132). A case in point is a large Beaker carinated bowl (Figure 4b) thrown into a shaft at Molino Sanchón (Figure 1, no. 1). It has been related to the ritualised closing of this salt processing site, as a way to give thanks for the release of valuable substances (Abarquero et al., 2012: 331–334; Guerra-Doce et al., 2011: 814, 816). The discovery of incomplete Beaker vessels in contexts sealed during this period is also widespread (e.g. Liesau et al., 2014b). At the causewayed enclosure of Camino de las Yeseras (Figure 1, no. 7), this occurrence has been understood within extended cycles of retrieval, manipulation, and exhibition of relics from ancestors – involving human remains (especially skulls and large bones), pottery, and metal items (Liesau et al., 2014a: 141–147). Partial Beaker wares fractured elsewhere and subsequently handled and selectively deposited are also well-known practices (Garrido-Pena et al., 2011: 122, 125). Examination of the ceramic sample from Valdeprados (Figure 1, no. 10) confirmed such a sequence. There, a heap of dismembered human bones from a male was accompanied by a gold sheet, a copper dagger, and three Palmela spearheads. The fill yielded 254 sherds, 69% of them without conjoins (Table 2) including 30 decorated Beaker sherds from three beakers and two complete plain vessels. A plain beaker was certainly smashed inside the pit; all its fragments were found in there, they exhibit homogeneous erosion, and the impact point is very apparent (Table 1). However, the incised beakers were not destroyed in situ, as previously thought (Garrido-Pena et al., 2011: 125). Since these sherds exhibit intense attrition marks, they were most likely gathered from elsewhere and brought there. This pattern goes hand-in-hand with the hypothesis of the manipulation of relics (Liesau et al., 2014b; Woodward, 2002) and is clearly recognised in several Beaker-associated tumuli, such as the El Alto III barrow (Figure 1, no. 6) (Rojo et al., 2014: 33–34) or Los Tiesos I and El Morcuero (Figure 1, nos. 11 and 13) (Blanco-González, 2014a: 4–6). These two monuments show over 75% of ‘orphan’ sherds, and a third of their assemblages are highly abraded (Table 2), indicating off-site breakage and prolonged degradation (Table 1). Finally, partial Bell-Beaker goods also indicate intentional destruction for further purposes. At Humanejos (Figure 1, no. 9), a matchless hypogeum tomb with a ladder and a large burial pit had been looted and their contents intentionally broken and dispersed during the Beaker period. Such desecration acts were likely aimed at obliterating the memory of those buried there, in a context of factional dispute (Flores-Fernández and Garrido-Pena, 2014).
The social lives of Bronze Age people and things (1800–1100 BC)
In the northern half of Iberia, most Bronze Age remains are recovered from deliberately filled sunken features, the almost unique depositional type-context (e.g. Colomer et al., 1998: 58). This is especially true in the Central Meseta. In traditional accounts (e.g. Fernández Castro, 1995), the second millennium BC in this macro-region was led by wandering herders who were technologically underdeveloped (ephemeral occupations in pit sites), politically regressive, and isolated from large-distance exchange networks (lacking elite burials or prestige items). This clichéd vision was reinforced by their perpetuation of long-lasting traditions: pit-digging on lowland spots occupied by their predecessors; reuse of old caves and megaliths; and striking decorative resemblances with earlier potteries. Such narratives do not adequately match the elusive remnants left by those peoples, whose materiality seems at odds with conventional ways of archaeological inference (Blanco-González, 2015). In short, the ‘reflectionist’ stance has played a particularly ominous and misleading role here. New lines of enquiry are proving useful in discrediting such stereotypical but pervasive and durable readings. It is increasingly clear that dwellings were systematically dismantled and the customary funerary ritual was the exposure of corpses, normative procedures leading to no traceable vestiges, save exceptions under abnormal circumstances (Blanco-González, 2015; Esparza et al., 2012). As a consequence, reconstructions of subsistence lifestyles and social organisation are especially intricate (Delibes and Romero, 2011; Díaz-del-Río, 2001). Even the most seminal arguments, such as the duality of low-lying farmsteads and hilltop steadily inhabited sites (Arnáiz and Montero, 2011: 559–561) need a more in-depth examination. The taphonomic and refitting examination of several ceramic assemblages (Table 2) was aimed at tackling such questions.
The Middle Bronze Age (1800–1450 BC) lowland pit site of El Cerro (Figure 1, no. 4) yielded unusual findings from undisturbed contexts, such as a semi-sunken dwelling, a triple burial pit with children and Neolithic and Beaker ceramics. The characterization of a selected ceramic collection from 14 features instigated a rethinking of their formation dynamics, stressing the terminal stages (Sánchez-Polo and Blanco-González, 2014). Most pits were casually filled with random aggregates of refuse made up of recently broken (fresh sherds: 72%) and very fragmented ceramics (47% are small sherds) (Table 2). However, several deposits exhibit a greater deal of intentional arrangement: the inhumation of three children in a gully, the placement of a cow’s leg in a nearby shaft, and the cache of Early Neolithic ceramics on top of a subsoil feature of Bronze Age date (Figure 5a). The radiocarbon-dated children were interred simultaneously, and, according to aDNA, were siblings (Sánchez-Polo and Blanco-González, 2014: 7–9). It seems plausible that the concurrent deaths of three young members of a family might have altered its social and cosmological order. The archaeological evidence at El Cerro might be accounted for as an extraordinary response by the living group. It likely prompted the abandonment of the polluted hut and its ancillary pits, the deviant inhumation of the corpses and the mobilization and final deposition of potent ancestral relics (Lillios, 1999) or otherworldly tokens – the anachronistic potsherds.
(a) Cache of well-preserved Early Neolithic sherds from Bronze Age feature at El Cerro (Photos: A. L. Palomino and A. Sánchez). (b) Vessels with outstanding social biographies from Pit 23 at Pico Castro. Scales in cm.
The Late Bronze Age (1450–1150 BC) hilltop site of Pico Castro (Figure 1, no. 3) is one of these prominent steep crags regarded as central places. The taphonomic and refitting operation with ceramics from 23 pits from Sectors IV and V at its summit (representing 0.2% of the total site) was conducted to evaluate the temporal dynamics of its occupations (Table 2). The very low representativeness of vessels within the studied features (81% are orphan sherds) and contrasting values in their preservation condition (66% are freshly broken and 29% lowly eroded) suggest diverse temporalities and formation trajectories between Sectors IV and V, 230 m apart. Thus, rather than being a permanent settlement, Pico Castro was probably used for centuries as a seasonal gathering place (Blanco-González, 2014b: 452). Again, most cut features were filled with detrital re-deposited residues, but there are also remarkable patterned deposits involving selective deposition of things. Pit 23 stands out as an almost unique assemblage of complete or almost entire ceramic vessels placed together: three bowls exhibiting geometric decorations (some are one-off astral motifs) and several coarse pots and jars (Figure 5b). Their thorough examination demonstrated prolonged social lives after breakage, and protracted abandonment gestures (Blanco-González, 2014b). These involved the detachment of sherds, their undergoing an array of degradation processes (calcite accretion, attrition, exposure over fire), and eventual gathering and gluing prior to their inverted placement (Figure 5b).
To sum up, there is room to advocate the contribution of culturally idiosyncratic practices in the ways of managing material culture during the second millennium in Central Iberia. Average ways of disposing of waste (Garrow, 2012: 109–113) aimed at erasing all material remains from the earth’s surface. Fortunately, especially valued things or unforeseeable developments prompted deviant procedures of disposal, which contributed decisively to the formation of this misunderstood archaeological record (Blanco-González, 2015). Bronze Age people encapsulated the involvement with partial items, despite the fact that planned breakage of vessels has not been demonstrated in the chosen case-studies. Endurance of such atavistic habits is to be regarded as a cultural revival: second-millennium people chose to draw on previous cultural manifestations, rather than being passively conducted by historical inertia.
Concluding remarks
This paper has stressed the very fragmentary and unbalanced nature of the extant material evidence throughout Iberian later prehistory. Entrenched empiricist premises have led to the analysis of this problematic archaeological record in an unduly confident way, whereby the domestic sphere supposedly faithfully reflects the material conditions of life, and the mortuary/ritual occurrences (of undisputable intentional character) mirrors shared beliefs and past social organisation. In order to challenge these contentious yet consensual assumptions, the taphonomic and depositional research presented here has examined a selection of especially ambiguous contexts. This attempt has proven effective in supporting a ‘bottom-up’ reappraisal (Gamble, 2001: 195–196).
This diachronic perspective has tracked the extended currency of several ever-changing social practices, as linking threads running through space–time. The breakage, dispersal, retrieval, handling, and final deposition of ceramics and further substances (biological remains, stones, metals, ashy sediments, etc.) were crucial to those Iberian societies. They responded to contingent circumstances, but their everlasting contribution to the formation of the archaeological evidence is to be underscored.
This overview began with the earliest pottery-using communities, c. 5500 BC. In Pit 1 at La Lámpara, some purposeful gestures are apparent in both the regular fracture of a jug, likely reconfigured to enhance its anthropomorphism (Rojo et al., 2008: 377–378) and its placement accompanying a dead woman (Figure 2). Beyond the funerary realm, patterning and willingness are also traceable in the filling of sunken features. The association of two unique sherds belonging to different jars within Pit 13 (Figure 3b) may have responded to their joint pre-abandonment manipulation, or may have come about as a genuine case of selective deposition. In any case, this latter option is strongly backed by the incorporation of particular ceramics (Rojo et al., 2008) and disarticulated human bones (Garrido-Pena et al., 2012; Weiss-Krejci, 2011) to contemporary depositional contexts. The Chalcolithic period exhibits a plentiful empirical basis for the deliberate breakage of pottery: Bell Beakers smashed within pits as votive offerings (Abarquero et al., 2012), or despoiled in tombs to erase the memory of competing factions (Flores-Fernández and Garrido-Pena, 2014). Several strands of evidence bear witness to the circulation and discriminatory abandonment of partial things such as broken Bell-Beaker vessels or isolated bones, likely regarded as relics (Liesau et al., 2014b; Woodward, 2002). Besides this, the eloquent third-millennium BC evidence allows confronting uncontested assumptions. Thus, the hut at Fuente Lirio, with its well-preserved and abundant refitting sherds, but with peculiarities diverting from a true ‘Pompeii-like’ context (Table 2), represents a touchstone against which to assess prospective ‘occupation layers’. Rather than being so, many of them will prove to be accumulations occurred once dwellings were left (even emptied), during the abandonment phase (LaMotta and Schiffer, 1999: 22–24). Moreover, interpretation should neither add functional hierarchy nor rely on taken-for-granted misnomers (‘burial goods’, ‘prestige items’), even when utilitarian roles seem straightforwardly inferable. Ditched enclosures (El Ventorro, Camino de las Yeseras), deviant deposits (Valdeprados), and Beaker-associated tumuli (El Alto III, El Morcuero) were neither residential nor solely funerary/ritual in character. Finally, in contrast to expectations, Bronze Age low-lying (El Cerro) and hilltop (Pico Castro) sites involved complex abandonment dynamics and multiple temporalities. Deliberate breakage has not been confirmed for the second millennium BC, because of the sheer partiality of surviving remains and the absence of suitable contextual information. Yet, the conscious handling of partial things is very manifest in both studied sites (Figure 5) and is increasingly recognised elsewhere. Thus, the rare triple burial pit at San Román de Hornija (Figure 1, no. 2) yielded half of a vessel, the remainder being missing. Following Chapman and Gaydarska (2007), such a behavior has been related to enchainment between the dead and the living (Esparza et al., 2012: 309). Moreover, the manipulation of ancient things, best categorized as relics or heirlooms with prolonged biographies (Lillios, 1999; Jennings, 2014), parallels the practices of circulating and placing radiocarbon-dated human bones from ancestors (Esparza et al., 2012). Therefore, it is possible to argue for a long-term, transcultural phenomenon consisting of the manipulation and eventual discard of incomplete items; it originated in the Early Neolithic and only came to an end over four millennia later, in the Late Bronze Age.
All in all, the studied archaeological cases are the physical correlates of a wide suite of irretrievable motives (David, 2004: 67; Davis, 1992: 336–337). These resulted from devoting efforts to attain hardly surmised utilitarian goals, which not necessarily matched our own rationale of instrumentality and causation (Brück, 1999: 320–322). The depositional episodes glimpsed in this narrative constituted key media for the ongoing reproduction of sociality through their continual reworking (Edwards, 2012: 94–96; Jones, 2012: 22–28, 127). It was their iterative performance in face-to-face social interactions which was likely important, rather than the imperishable results of contingent (sometimes even improvised) procedures aimed at transient purposes. The resulting outcomes are always difficult to categorize (Jones, 2012: 134) and might be described in geoscientific terms as anthropogenic accumulations somewhere in a sliding continuum of depositional practice (Brudenell and Cooper, 2008: 30; Garrow, 2012: 94; Lamdin-Whymark, 2008: 175; Lucas, 2012: 123; Thomas, 2012: 125) between two opposite ends (Table 1): (a) planned products which testify to careful patterned or ‘ritualised’ closures (Lamdin-Whymark, 2008); and (b) casual aggregates or palimpsests (Lucas, 2012: 112–123), unintended effects ‘generated through habitual and unconsidered adherence to cultural conventions’ (Thomas, 2012: 125) which came about by intending something else (Garrow, 2012: 107). Formal, odd, or ‘structured’ deposits are identified by their integrity (complete items, articulated portions) and spatial patterning, features which first attracted archaeological attention. This paper has focused particularly on the much neglected and under-theorised consequences of routine average practice, also known as material culture patterning (Garrow, 2012: 109–113). In this regard, such manifold cases confirm that variability was indeed the standard (Garrow, 2012: 111). Observations also suggest that even unplanned results (e.g. accidental breakage, unconcerned transfer of refuse, casual depositional patterning) have a great potential to inform depositional practice (Thomas, 2012). The backfilling of prehistoric sunken features (Thomas, 1999) is the most representative case in point (Table 1). Taboos and precepts guiding these habitual ways of engaging with substances did not impede (and even sometimes might have facilitated) the expedient and opportunistic retrieval, reuse, or recycle of previously discarded items. It is worth noting that since no Pompeii-like contexts (Table 1) are represented here, most studied ceramic assemblages (Table 2) experienced significant removal from their original utilitarian contexts (if any). Remarkable selection and diminution dynamics mediated the extant material fraction due to assorted combinations of natural and cultural factors (Lucas, 2005: 127–129, 2012: 61–68; Thomas, 2012: 126–127). This realisation entails important interpretive consequences. A great deal of such materials, especially the frequent cases of material culture patterning (Garrow, 2012: 109–113), is to be considered as an indirect source informing of severely distorted prehistoric activities beyond those terminal ones referred to here.
Pottery has demonstrated its informative value as a socially embedded material, an accessible medium which afforded consecutive human–thing entanglements (Hodder, 2012). Because of its physical properties, both mixing/forming and detaching/dividing were two inseparable aspects acknowledged by pottery-using groups from the very onset. Prehistoric things might have accrued meanings and these were context-specific and far from immanent or fixed. Thus, certain vessels (Figure 5) became redolent of sociality throughout their complete existence (Blanco-González, 2014b: 452; Jennings, 2014: 48–49; Jorge, 2014: 494–450), and this observation speaks volumes on their social biographies or itineraries (Gosden and Marshall, 1999; Hahn and Weiss, 2013; Meskell, 2004). By contrast, several examples, such as the case of Bell-Beaker ‘symbolically’ decorated sherds (Figure 4a), could have lost their social esteem and finished as socially deactivated refuse. Importantly, such things were endowed with ascribed connotations irrespective of their intrinsic physicality. A fetishist attitude (cf. Meskell, 2004: 46–50) has focused excessive attention on ornamented ceramics, thus curtailing the testing of further possibilities at alternative analytical scales. Both plain coarse- and profusely decorated fineware, partial and eroded or complete pots were actually handled in uncanny ways, such as those traced in Pit 23 from Pico Castro (Figure 5). Old fractures and missing parts are too often disregarded as evidence for post-depositional, unproblematic disturbance (e.g. Fitzpatrik, 2011: 37–43). Yet, even after undergoing degradation processes such as breakage or erosion, some vessels kept accruing further meanings and participated in social episodes, as if such alterations represented but a threshold to a new status rather than the end of their social lives. In the face of this realization, western–modern categorisations, such as damaged or dysfunctional, should be challenged (Jones, 2012: 58) in order to not eclipse an adequate understanding of the otherness of the past. 3
In conclusion, the focus on ceramic fragments has illuminated the wide variability behind a series of challenging depositional contexts in later prehistoric Iberia. This paper has spotlighted assorted degrees of intentionality and diverse ways of framing the social practices of fragmentation and deposition in contingent prehistoric contexts. This reappraisal controverts key assumptions stemming from the customary and widespread ‘reflectionist’ standpoint: (a) the surviving evidence is a deformed and faint image of past material conditions of life; (b) fragmentation and incompleteness emerge as highly context-dependent archaeological features instead of natural, only modern, or unavoidable side-effects; and (c) past human intervention was a driving factor decisively channelling all long-lasting physical consequences of social action, both planned end products and random effects. Important insights have been gained on the fragmented nature of the evidence and the kind of narratives we can draw upon these past fragments.
Footnotes
Acknowledgements
This research has been possible due to a Marie Curie fellowship (PAST FRAGMENTS FP7-PEOPLE-2011-IEF-298285) and a post-doctoral contract (FPDI-2013-17394) funded by the Spanish Ministry of Economy. The paper benefited from the advice of Prof. John Chapman (University of Durham), and the improvements by Dr Ben Jennings (University of Bradford) and Prof. Katina Lillios (University of Iowa) who also corrected my English versions. I am indebted to Prof. Lynn Meskell (Stanford University) and three anonymous reviewers for their insightful suggestions and criticisms on an early draft.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The author(s) received no financial support for the research, authorship, and/or publication of this article.