Did Neolithic farming fail? The case for a Bronze Age agricultural revolution in the British Isles.

Stevens, Chris J. ; Fuller, Dorian Q.

[ILLUSTRATION OMITTED]

Introduction

Gordon Childe classically defined the Neolithic as the transition to agriculture, linking all aspects of Neolithic life from cereals to ceramics into a single cultural package (Childe 1952). For the British Isles, the effect of this 'Neolithic package' on subsistence has been subject to recurrent debate, concerning both the rate of change from hunter-gather to farming, and the nature of Neolithic subsistence, particularly the importance of cereals. The various stances can be broadly divided into two opposing models. The first, the Gradual Indigenous Transition model (Figure 1A, after Thomas 1999), sees a slow change from a hunter-gather to an increasingly agricultural economy, initially associated with the acculturation of farming by hunter-gather groups. In contrast the Rapid Introduced Neolithic (Figure 1B, see Rowley-Conwy 2004, 2011), proposes the swift emergence of a fully agricultural subsistence base around 4000 BC, fuelled by migrating farmers from the continent who quickly displace or acculturate existing hunter-gatherers.

[FIGURE 1 OMITTED]

Encapsulated within these models is considerable controversy over the degree of dependence on cereals. In keeping with the Rapid Introduced Neolithic, are those that see the swift emergence of fully-fledged farmers, with cereals and domesticated animals quickly becoming the main dietary staples (e.g. Rowley-Conway 2004, 2011). Compliant with the Gradual Indigenous Transition, are those who emphasise the role of pastoralism, alongside the continued hunting of wild animals and the collection of wild plant foods, with less reliance on cereals (e.g. Entwhistle & Grant 1989; Thomas 1999). Our purpose in this paper is to argue for a punctuated model for the introduction of farming to Britain when viewed over the long term.

Assessing the evidence for plant use

Several lines of evidence can be used to examine these models, but each has its limitations. Pollen studies show a variable mosaic of clearance during the Neolithic, but many parts of the British Isles remain relatively wooded (Richmond 1999). Further, several lines of environmental evidence suggest cyclical phases of clearance and woodland regeneration continuing throughout the Neolithic in many regions (Thomas 1999: 32; Hey & Robinson 2011). Such patterns could be in keeping with small, intensive plots of cultivation inferred for Britain, similar to those of Early Neolithic mainland Europe (Bogaard & Jones 2007).

Other types of evidence for cereal cultivation, including querns and ard marks, suggest less reliance on cereals than that seen in later agricultural periods. For example, granaries, storage pits and, for many regions, field systems are unknown before the Middle Bronze Age (Bradley 2007: 181-93). Other studies indicate that the incidences of dental caries--associated with the consumption of processed cereals--are very low in British Neolithic populations compared with later periods (McKinley 2008). Similarly, isotope evidence indicates that Neolithic diets were higher in animal protein than later periods, where cereals are accepted as the main dietary staple (Richards 2000). All of this points to cereal cultivation and consumption in the Neolithic, but at a scale that differed in extent to the Middle Bronze Age and later periods.

Before the advent of routine flotation, one factor influencing the view of a fully cereal-based Neolithic was the high number of cereal impressions seen within Neolithic pottery (Jones 1980). However, a major shift in perception regarding subsistence can be traced to a seminal paper by Moffett, Robinson & Straker (1989), in which the evidence for carbonised cereals was shown to be slight in the British Neolithic compared with that for wild foods. By the late 1990s systematic sampling for charred remains from Neolithic features, irrespective of the nature of the deposit, was common practice within British archaeology (Robinson 2000: 85). Further, even if less visible to excavators, cereal grains often occur in samples with hazelnut shell. While occasional large deposits of grain continued to be recovered, the situation remains largely unaltered for many sites, with wild foods frequently better represented than grain (Robinson 2000; Stevens 2007). Much of the charred evidence for subsistence within the British Neolithic and Early Bronze Age comes from isolated pits or clusters of pits, which yield hazelnut shell waste. These can be argued to contain refuse relating to occupation and past settlement (e.g. Hey & Barclay 2007: 400; Hey & Robinson 2011: 256; cf. Thomas 1999: 32). Charred plant food remains of any sort are generally absent or sparse on ceremonial sites (Robinson 2000: 86).

The relative reliance on cereal cultivation and wild resources remains contentious. Some researchers argue that Neolithic sites are frequently under-sampled; that sampling may target 'black' deposits containing more visible charred hazelnut shells and this, together with the prevalence of ceremonial/ritual sites over settlement sites, may account for the low numbers of charred cereal grains recovered. Further, they argue that cereal grains will be underrepresented compared with hazelnut shells, and when these factors are taken into account the archaeobotanical data support cereals as a major dietary staple, only supplemented by wild plant foods (Jones & Rowley-Conwy 2007; Jones & Legge 2008).

However, it is not only cereal grains that are sparse in British Neolithic charred assemblages, but crop-waste in general, including grains, weed seeds and chaff. Chaff is often prolific on later sites, but virtually absent from Neolithic ones (Robinson 2000), while hazelnuts, by contrast, are poorly represented from the Middle Bronze Age onwards. Given that high numbers of charred glume bases can be related to waste from the routine processing of hulled wheat stored as spikelets in later periods, then ipso facto their absence in the British Neolithic could imply that crops were stored fully processed. Under such a scenario, not only chaff, but hulled wheat grains--which are readily lost during dehusking--would be underrepresented (Stevens 2007). An alternative explanation would be to attribute the paucity of cereal remains to the transient nature of much Neolithic settlement, in which little of anything was preserved charred. Transient settlement would equally reduce circumstances for hazelnut preservation in comparison with better preservation contexts of later less ephemeral sites, thus making the contrast in the presence/absence of hazelnut shells all the more striking.

An alternative perspective is that the quantities of hazelnuts represent a real pattern of use. On Mesolithic sites, where wild plant foods would have formed a significant proportion of the diet, charred fragments of hazelnut shells are commonly recovered. Conversely, they are generally sparse or absent from the Middle Bronze Age and Iron Age assemblages. The implication is that they signify a greater importance of wild foods in the Neolithic than in subsequent periods (Robinson 2000).

In summary, Neolithic subsistence economies in the British Isles were clearly different from the undisputed cereal-dominated systems that followed. While the introduction of cereals in the Early Neolithic may have revolutionised British economies, the charred evidence indicates that cereals were less frequently discarded and carbonised in the Neolithic than in later periods. The reverse is seen for wild food remains, especially hazelnut shell. On balance, a significant change in subsistence occurring between the Neolithic and the Middle Bronze Age is the logical deduction from all the available evidence, perhaps more in keeping with a punctuated transition to long-term and persistent cultivation of cereals.

Dating plant use

Taken together, the Neolithic and Early Bronze Age--the latter often combined with the Late Neolithic--span 2500 years (c. 4000-1500 BC), more than one third of the entire time since cultivation was introduced to the British Isles. Should we assume that subsistence remained unchanged throughout this period, or was uniform across the British Isles as a whole?

A recent examination of radiocarbon dates on cereal grains or associated material indicated their introduction around 4000-3900 cal BC (Brown 2007). Similarly, Collard et al. (2010), utilising summed probability distributions on radiocarbon dates to estimate population change, concluded a rapid population increase between 4000 and 3700 cal BC, more in keeping with the migration(s) of farmers with rapid demographic growth, and some acculturation of indigenes (as per Rowley-Conwy 2011; Sheridan 2011), but not with a gradual piecemeal adoption by native Mesolithic populations. However, both studies included radiocarbon dates on wood charcoal, which can yield dates much older than those on short-lived material, e.g. cereals and hazelnut shells (Brown 2007).

The most reliable method we have for confirming the presence of cultivation at any point in time and space is by obtaining direct radiocarbon determinations on domesticated cereals or other cultigens themselves. Additionally, the collection of radiocarbon dates on wild foods as part of this dataset allows changes in subsistence to be examined across time and space through comparative trends in the summed probability distributions for each. Further, and in conjunction, the two sources should allow changes in population levels (as per Collard et al. 2010), alongside those of subsistence, to be explored.

The importance of direct AMS dates on grains has been a key methodological advance in the study of early agriculture. Cautionary tales of allegedly early domesticates in the Epipalaeolithic are well known (Wendorf et al. 1984; Legge 1986). For the British Isles these problems are a concern on later Neolithic sites, where large deposits of charred cereals are generally absent and their density extremely low (see Robinson 2000; Jones & Rowley-Conwy 2007: tab. 23). In light of these problems the inferred persistence of cereal agriculture through the Neolithic, based solely on the presence of potentially intrusive charred cereals in low quantities, needs to be seriously questioned.

Re-examining the radiocarbon chronology for cereals and nuts

In order to investigate these issues the authors compiled a database comprising over 700 direct radiocarbon dates on cultigens and wild food remains from 198 sites, spanning the Mesolithic to the end of the Bronze Age (700 cal BC), across the British Isles (summarised in Table 1; itemised in the online supplement). The list contains 415 direct dates from 115 sites (including intrusive material) on cultigens, mainly cereals, but also flax, pulses and grapes, and 359 dates from 114 sites for wild food remains, mainly hazelnut shell (Corylus avellana), but also sloe (Prunus spinosa) and crab apple (Malus sylvestris). The list adds a large number of new direct dates on cereal and wild foods to the datasets examined by Collard et al. (2010) and Brown (2007), while avoiding charcoal dates, as well as significantly spanning a much greater period of time. The compiled list no doubt contains omissions, but the number of dates is sufficient to be able to begin to make some important observations on the occurrence of cereals at any point in time and space, and in turn the extent and role of cultivation from the Neolithic to the Late Bronze Age.

These dates were selected from the corpus only after a critical review of the context. While a site-by-site study of charred assemblages might provide some insight into temporal and spatial variation for subsistence, a review of data compiled by Jones & Rowley-Conwy (2007: tab. 23.1) demonstrated that such an approach would prove unfruitful without direct dating controls. A detailed examination of the site reports showed many cases in which the phasing for deposits from which cereals were recovered was suspect. These include features with no dating evidence, features containing later 'intrusive' pottery, and deposits cut by later features rich in charred cereals. Of particular concern was that in several cases radiocarbon dating demonstrated that the cereals or other cultigens were themselves intrusive and not of Neolithic date, e.g. Sewerby Cottage Farm, York; Brampton, Cambridgeshire; Flagstones, Dorset; Hazelton North, Gloucestershire; Chigborough Farm and Slough House Farm, Essex; with more recent examples including Drumoig, Leuchars, Fife; Whitehorse Stone, Kent; Kingsmead Quarry, Berkshire; and Harlington, Middlesex (see online supplement for details).

The selected radiocarbon dates were divided into two groups: wild foods (mainly hazelnut shells) and crops (mainly cereals), and were calibrated in OxCal (Version 3.10: Bronk Ramsey 2005), using IntCal09 calibration curve (Reimer et al. 2009). Following the lead of several recent studies on the spread of the Neolithic in Europe (e.g. Gkiasta et al. 2003; Shennan & Edinborough 2007; Collard et al. 2010), summed probability distribution of all calibrated radiocarbon dates in each group were separately compiled from the dataset (Figures 2 & 3). The summed probability distributions for each group were then examined for regions within the British Isles (Figure 4) and finally for the islands and mainland (Figure 5).

[FIGURE 2 OMITTED]

The spread of crops

As seen by Brown (2007), the earliest evidence points to an introduction of cereals around 3950-3850 cal BC. This is rapidly followed by an increase in the summed probability distributions between 3800 and 3700 cal BC for dates both on wild foods and cereals (Figures 2 & 3), supporting the sharp increase in population associated with migrating farmers and subsequent demographic growth (Collard et al. 2010). It is worth noting that the adoption of cereals through cultural diffusion by native hunter-gatherer groups in the strictest sense of the term is highly improbable within such a short time span. Not only did crops have to be brought into the British Isles, but also accompanying technology and knowledge of all aspects of crop husbandry had to be learnt and understood. For non-literate societies such knowledge would only be gained through participation in the practices themselves. If indeed the existing native populace learnt such practices, they must either have participated within cultivating societies or brought cultivators into their own, implying some balance of migration and acculturation.

Looking at the summed probability distributions on a region by region basis, some suggestive patterns emerge (Figure 4). It should be noted that some regions are poorly represented, at least in the radiocarbon dates we have been able to find. The initial appearance of domestic crops appears to be rapid throughout the British Isles, with broadly similar dates appearing within south-east England, and only slightly later in eastern Scotland. There then appears a slight delay before crops spread to the Scottish islands, with the data suggesting that the collection of hazelnuts was far less important there than on the mainland (Figure 5).

[FIGURE 3 OMITTED]

Did Neolithic farming fail?

After the initial population increase on the mainland (peaking around 3700 cal BC), the summed probability distributions indicate a sharp decline in the number of cereal dates, while dates on wild foods continue to form a prominent part of the dataset. This latter event, around 3650-3600 cal BC, broadly coincides with the construction of causewayed enclosures (cf. Cleal 2004), perhaps relating to a shift from cereal-focused farmers inhabiting long-houses to a slightly more mobile and pastoral-based society.

Using various lines of evidence, such as the fewer cereal impressions in pottery, indications for regenerated woodland, lack of settlement evidence and the general paucity of charred cereal remains, many authors over the past half-century have suggested that a shift from an arable to a mobile pastoralist society occurred during the British Late Neolithic (Piggott 1954: 365; Bradley 1978; Jones 1980; Moffet et al. 1989; Robinson 2000). Our study not only supports these theories, but can pinpoint it as a broadly synchronous event occurring around 3350 cal BC (Figure 5), marking the transition from the Early to Middle Neolithic. Significantly, the almost total absence of cereal dates after this date raises a real possibility that in some regions, particularly mainland England, cereal cultivation not only declined, but was potentially abandoned.

[FIGURE 4 OMITTED]

While dates on hazelnut shells are present for many sites after this period, they are less common than seen for the Early Neolithic, suggesting the decline was accompanied by a population collapse, inferred also by Collard et al. (2010). It appears that cereal cultivation briefly re-emerged in parts of the mainland during the Beaker period, around 2300-2000 cal BC (Figures 4 & 5), but the general gap in evidence for cereals within many regions, spanning well over 1000 years through the later Neolithic and Early Bronze Age, remains striking. An examination of charred assemblages from the Thames Valley has shown a similar pattern, with little evidence for cereals in the Late Neolithic, but increased numbers in the Beaker period (Hey & Robinson 2011: 258).

On the islands a different trajectory appears to have been followed, with continuing evidence for cereals and permanent occupation, and less reliance on hazelnuts, although again a potential population decline is indicated around 2900-2800 cal BC.

[FIGURE 5 OMITTED]

Were the builders of Stonehenge without cereals?

The continuation of dates on hazelnut shell is a significant factor. It demonstrates that while population levels fell, along with the potential abandonment of cereals, what might be considered food or domestic/settlement waste is still visible in the archaeological record (cf. Robinson 2000; Hey & Robinson 2011). This continuation of charred evidence attributable to occupation waste, in conjunction with the presence of radiocarbon dated cereals within the Early Neolithic and from island sites in the later Neolithic, suggests that neither taphonomy, site type nor sampling strategies are solely responsible for the observed patterns. Importantly, while quern stones are common in the Early Neolithic, they become rarer on later Neolithic and Early Bronze Age mainland sites (Curwen 1937; Stevens 2007), although they represent an under-studied line of evidence.

It is only when this lack of evidence for cultivation during the Middle and Late Neolithic across the mainland British Isles is compared against the archaeological record that the true implications can be seen. The later Neolithic to the Early Bronze Age represents a time in which many significant monuments were constructed across Britain, which in turn have been associated with large-scale agricultural societies (e.g. Case 1969; Legge 1989). Yet, for example, the most recent review of the dating of Stonehenge suggests all the main phases of construction took place between 3000 and 2400 cal BC (Parker Pearson et al. 2007), and therefore within the period in which there is little firm evidence for cereal agriculture within mainland Britain.

The association of monument building with cereal-based agricultural societies therefore appears to be misplaced, raising questions about the social mechanisms by which large numbers of people were organised to achieve such activities, and what role subsistence played within this. Parallels are perhaps better sought in the monuments of the nut-eaters of Poverty Point, Lousiana (Fritz 2007), or the Middle Jomon sites of Japan, e.g. Sannai Maruyama (Habu 2008).

The Bronze Age agricultural revolution

Given that the summed probability distribution for cereals at the commencement of the Middle Bronze Age closely matches the pattern seen in the Early Neolithic (Figure 5), it is difficult not to view it in a similar light. The innovations of the Middle Bronze Age have traditionally been interpreted as a shift from long-fallow cultivation to short-fallow with fixed plots (Barrett 1994). But perhaps they should rather be seen as a rapid population expansion associated with migrating farmers and the re-introduction of cereals to the British mainland. Several new crops--spelt, pea and bean--are all only first reliably dated in the Middle or Late Bronze Age. Additionally, the Middle Bronze Age sees the first appearance within mainland Britain of many aspects of fully-fledged agricultural societies, including storage pits, four-post granaries, waterholes, loom weights, wool and field-systems (cf. Bradley 2007: 181-93), along with increased evidence for both tillage and querns. While Early Neolithic cereal agriculture was most likely small-scale, perhaps conducted within intensively managed plots (Bogaard & Jones 2007), and supplemented by the routine collection of wild resources, the second agricultural revolution of the Middle Bronze Age represented more land-extensive and labour-intensive cereal farming, with minimal use of wild resources.

[FIGURE 6 OMITTED]

An alternative model

The Neolithic spans a great many years, and archaeological data indicate a great many changes; from long-houses, long-barrows and chamber cairns to causewayed enclosures and cursus monuments in the earlier Neolithic; to palisade enclosures, round-barrows and stone and timber henges in the later Neolithic (Barrett 1994; Bradley 2007). It is now possible to tie these changes to both subsistence and demographic evidence. While the Neolithic was an initial agricultural revolution as Childe (1952) and others have argued (Rowley-Conwy 2004; Collard et al. 2010), it was only the first such subsistence revolution, and one which failed to fully take hold in the British Isles over the subsequent millennium. In this regard neither the Gradual Indigenous Transition nor the Rapid Introduced Neolithic model are supported by the current available data. We propose an alternative model, the Multiple Transference model (Figure 6), in which foraging remained important alongside earlier Neolithic cereal agriculture, but that cereal cultivation was abandoned throughout many parts of the British Isles in favour of increased reliance on pastoralism and wild resources during the Middle to Late Neolithic.

The most likely explanation for simultaneous abandonment of cereal agriculture and population collapse after 3350 cal BC is climatic deterioration: an increased reliance on wild resources, seen in later Neolithic Switzerland, has also been tied to a deteriorating climate (Schibler & Jacomet 2010). Analogous scenarios are seen archaeologically for different time periods in various parts of the world. For example, early stone monument construction in the Eastern Sahara amongst non-cultivating pastoralists occurred under conditions of declining climatic conditions (Malville et al. 1998; di Lernia 2006), while the shift away from settled agriculture towards mobile pastoralism, characterising much of peninsular India from the end of the Chalcolithic (1200-900 BC), can also be linked to climatic fluctuations (e.g. Dhavalikar 1988; Asouti & Fuller 2008:41-42). In the Gansu region of north-west China, the Dadiwan Neolithic pursued low-level millet cultivation for five centuries or more during the sixth millennium BC, before apparently fading away, with a hiatus of more than five centuries prior to the influx of more permanent millet-pig agriculture associated with the immigrant Yangshao tradition (Bettinger et al. 2010). A further case is seen in the shift from sedentary agriculturalists to nomadic-pastoralism in Late Bronze Age Mongolia, associated both with the appearance of stone monuments and possible climatic change (Allard & Erdenebaatar 2005).

The inference that cereal cultivation was abandoned runs contrary to our progressive evolutionist expectations, in which the beginning of cultivation was an advance past a point of no return. While certain levels of population density cannot be supported without agricultural production, the point at which these levels are attained must be examined in a local ecological context. Subsistence should be understood as a strategic adjustment of inherited traditions to the realities of making a living within given environmental and climatic conditions, and as behavioural ecologists infer there are situations in which gathering makes better sense than cultivating (e.g. Winterhalder & Kennett 2006; Bettinger et al. 2010). That cereal agriculture could be abandoned in favour of wild plant foods, alongside pastoralism and limited hunting, strongly supports the notion that cereals were just part of a broader subsistence strategy in the Early Neolithic (Stevens 2007). The importance of collected wild foods during the Neolithic is seen across much of Europe (e.g. Gyulai 2007; Jacomet 2007; Kroll 2007; Rottoli & Castiglioni 2009). Therefore, that wild foods formed an important part of the diet for British Neolithic peoples is unsurprising given that migrating farmers from the continent already had a subsistence strategy encompassing both cultivated and wild foods, alongside domestic animals and some hunting. Of interest is the extent to which wild resources remain prominent in agricultural subsistence, as appears to have been the case in northern Germany (Kirlies et al. 2011), or whether cereals were abandoned in favour of wild foods, as we have argued occurred in Britain.

Following Halstead and O'Shea (1982) a diverse subsistence base, in the absence of obvious storage structures in the Neolithic on the mainland, would provide the only means of coping with crop failure (Stevens 2007). In this light we might speculate that repetitive poor harvests around 3300 cal BC, linked with climate deterioration, were the probable main factors behind the reduction and possible abandonment of cereals in favour of increased reliance on domestic animals and wild resources. Similarly, the continuation of cereal agriculture within island communities may be simply because such environments offered more limited supplies of wild plant foods than the British mainland.

The period in which cereals were largely absent from much of the British Isles mainland lasted until c. 1500 cal BC, although some possibly localised reintroductions occurred, in particular around 2300 cal BC. Rather than the Neolithic, it is perhaps the Middle Bronze Age that represents the real agricultural revolution for the British Isles, a period that established the subsistence base that supported the persistence of landscape modifications, agricultural population and social hierarchy over the millennia that followed. This Bronze Age 'agricultural revolution' was referred to by Childe over 50 years ago, and in line with the ideas expressed above, he suggested that it "provided subsistence for a vastly increased population" (1952: 187).

Conclusion

Profiting from the growing quantity of AMS dates across the British Isles, this paper offers an alternative line of enquiry through the use of independently-summed probability distributions for radiocarbon determinations on wild and cultivated plant foods as a method of tracking subsistence and population change through time. We have painted Britain with a broad brush, suggesting only differences between the mainland and islands (Figure 5), but further work should investigate the likely regional and local variation on the persistence or abandonment of agriculture during the Late Neolithic. In the absence of conclusive evidence we should not accept the existence of cultivation based upon deeply-entrenched beliefs that cereals could not be abandoned, or that they are intrinsically linked to the construction of monuments. For those who favour continuous cereal cultivation, archaeobotanical sampling and AMS-dating of grains provides a ready test of the hypothesis outlined above. The resulting peaks and troughs suggest that the importance of cereal cultivation was extremely variable throughout the first two to three millennia and across the British Isles. Further, the paper demonstrates the limitations of using traditional period divisions and highlights the need to escape the tyranny of the 'Neolithic' label, and its assumption of unchanging village farmers.

Acknowledgements

The authors would like to thank Alistair Barclay, Helen Lewis, Ruth Pelling, Sarah Wyles, Michael Grant, Sue Colledge, Andrew Fitzpatrick, Pippa Bradley, Mike Allen, Adrian Gollop, Rebecca Nicholson, and Neil Brodie, for discussion of some of these ideas and drawing our attention to sources of dated material over the years, although not all may agree with all the conclusions reached. We thank our reviewers, Glynis Jones and Wendy Carruthers for their constructive suggestions.

Received: 7 December 2010; Revised: 14 July 2011; Accepted: 7 October 2011

References

ALLARD, F. & D. ERDENEBAARTAR. 2005. Khirigsuurs, ritual and mobility in the Bronze Age of Mongolia. Antiquity 79: 547-63.

ASOUTI, E. & D.Q. FULLER. 2008. Trees and woodlands of south India. Archaeological perspectives. Walnut Creek (CA): Left Coast Press.

BARRETT, J.C. 1994. Fragments from antiquity. Oxford: Blackwell.

BETTINGER, R., L. BARTON & C. MORGAN. 2010. The origins of food production in north China: a different kind of agricultural revolution. Evolutionary Anthropology 19: 9-21.

BOGAARD, A. & G. JONES. 2007. Neolithic farming in Britain and Central Europe: contrast or continuity? Proceedings of the British Academy 144: 357-75.

BRADLEY, R. 1978. The prehistoric settlement of Britain. London: Book Club Associates.

--2007. The prehistory of Britain and Ireland. Cambridge: Cambridge University Press.

BRONK RAMSEY, C. 2005. OxCal program v3.10. Available at: http://c14.arch.ox.ac.uk (accessed 11 April 2012).

BROWN, A. 2007. Dating the onset of cereal cultivation in Britain and Ireland: the evidence from charred cereal grains. Antiquity 81: 1042-52.

CASE, H. 1969. Neolithic explanations. Antiquity 43: 176-86.

CHILDE, V.G. 1952. Prehistoric communities of the British Isles. London: W.R. Chambers.

CLEAL., R.M.J. 2004. The dating and diversity of the earliest ceramics of Wessex and south-west England, in R.M.J. Cleal & J. Pollard (ed.) Monuments and material culture: papers in honour of an Avebury archaeologist: Isobel Smith: 166-92. East Knoyle: Hobnob Press.

COLLARD, M., K. EDINBOROUGH, S. SHENNAN & M.G. THOMAS. 2010. Radiocarbon evidence indicates that migrants introduced farming to Britain. Journal of Archaeological Science 37(4): 866-70.

CURWEN, E.E. 1937. Querns. Antiquity 11: 133-51.

DHAVAHKAR, M. 1988. The first farmers of the Deccan. Pune: Ravish.

DI LERNIA, S. 2006. Building monuments, creating identity: cattle cult as a social response to rapid environmental changes in the Holocene Sahara. Quaternary International 151: 50-62.

ENTWHISTLE, R. & A. GRANT. 1989. The evidence for cereal cultivation and animal husbandry in the southern British Neolithic and Bronze Age, in A. Milles, D. Williams & N. Gardner (ed.) The beginnings of agriculture (British Archaeological Reports international series 496): 203-15. Oxford: British Archaeological Reports.

FRITZ, G. 2007. Keepers of Lousiana's levees: early mound builders and forest managers, in T. Denham, J. Iriarte & L. Vrydaghs (ed.) Rethinking agriculture: archaeological and ethnoarchaeological perspectives: 189-209. Walnut Creek (CA): Left Coast Press.

GKIASTA, M., T. RUSSELL, S. SHENNAN & J. STEELE. 2003. Neolithic transition in Europe: the radiocarbon record revisited. Antiquity 77: 45-62.

GYULAI, F. 2007. Seed and fruit remains associated with Neolithic origins in the Carpathian Basin, in S. Colledge & J. Conolly (ed.) The origins and spread of domestic plants in Southwest Asia and Europe: 125-40. Walnut Creek (CA): Left Coast Press.

HABU, J. 2008. Growth and decline in complex hunter-gatherers societies: a case study from the Jomon period Sannai Maruyama site, Japan. Antiquity 82: 571-84.

HALSTEAD, P. & J. O'SHEA. 1982. A friend in need is a friend indeed: social storage and the origins of social ranking, in C. Renfrew & S. Shennan (ed.) Ranking, resource and exchange: 92-99. Cambridge: Cambridge University Press.

HEY, G. & A. BARCLAY. 2007. The Thames Valley in the late fifth and early fourth millennium cal BC: the appearance of domestication and the evidence for change. Proceedings of the British Academy 144: 399-422.

HEY, G. & M.A. ROBINSON. 2011. Neolithic communities in the Thames Valley: living in the natural landscape, in G. Hey, P. Garwood, M. Robinson, A. Barclay & P. Bradley (ed.) The Thames through time: volume I, part 2; Mesolithic to Early Bronze Age: 221-60. Oxford: Oxbow.

JACOMET, S. 2007. Neolithic plant economies in the northern Alpine Foreland from 5500-3500 cal BC, in S. Colledge & J. Conolly (ed.) The origins and spread of domestic plants in Southwest Asia and Europe: 221-58. Walnut Creek (CA): Left Coast Press.

JONES, G. & A.J. LEGGE. 2008. Evaluating the role of cereal cultivation in the Neolithic: charred plant remains from Hambledon Hill, in R. Mercer, & F. Healy (ed.) Hambledon Hill, Dorset, England: excavation and survey of a Neolithic monument complex and its surrounding landscape, volume 2: 469-76. Swindon: English Heritage Archaeological Reports.

JONES, G. & P. ROWLEY-CONWY. 2007. On the importance of cereal cultivation in the British Neolithic, in S. Colledge & J. Conolly (ed.) The origins and spread of domestic plants in Southwest Asia and Europe: 391-420. Walnut Creek (CA): Left Coast Press.

JONES, M. 1980. Carbonized cereals from grooved ware contexts. Proceedings of the Prehistoric Society 46: 61-63.

KIRLIES, W., S. KLOOB, H. KROI.L & J. MULLER. 2011. Crop growing and gathering in the northern German Neolithic: a review supplemented by new results. Vegetation History and Archaeobotany 21: 221-42.

KROLL, H. 2007. The plant remains from the Neolithic Funnel Beaker sites of Wangels in Holsatia, northern Germany, in S. Cnlledge & J. Conolly (ed.) The origins and spread of domestic plants in Southwest Asia and Europe: 349-58. Walnut Creek (CA): Left Coast Press.

LEGGE, A.J. 1986. Seeds of discontent: accelerator dates on some charred plant remains from the Kebaran and Natufian culture, in J.A.J. Gowlett & R.E.M. Hedges (ed.) Archaeological results from accelerator dating. 13-22. Oxford: Oxford University Committee for Archaeology.

--1989. Milking the evidence: a reply to Entwhistle and Grant, in A. Milles, D. Williams & N. Gardner (ed.) Beginnings of agriculture (British Archaeological Reports international series 496): 217-42. Oxford: British Archaeological Reports.

MALVILLE, J.M., F. WENDORF, A.A. MAZAR & R. SCHILD. 1998. Megaliths and Neolithic astronomy in southern Egypt. Nature 392: 488-91.

McKINLEY, J. 2008. The human remains, in R. Mercer & F. Healy (ed.) Hambledon Hill, Dorset, England: excavation and survey of a Neolithic monument complex and its surrounding landscape, volume 2: 477-521. Swindon: English Heritage Archaeological Reports.

MOFFETT, L., M. ROBINSON & V. STRAKER. 1989. Cereals, fruit and nuts: charred plant remains from Neolithic sites in England and Wales and the Neolithic economy, in A. Milles, D. Williams & N. Gardner (ed.) Beginnings of agriculture (British Archaeological Reports international series 496): 243-61. Oxford: British Archaeological Reports.

PARKER PEARSON, M., R. CLEAL, P. MARSHALL, S. NEEDHAM, J. POLLARD, C. RICHARDS, C. RUGGLES, A. SHERIDAN, J. THOMAS, C. TILLEY, K. WELHAM, A. CHAMBERLAIN, C. CHENERY, J. EVANS, C. KNUSEL, N. LINFORD, L. MARTIN, J. MONTGOMERY, A. PAYNE & M. RICHARDS. 2007. The age of Stonehenge. Antiquity 81: 617-39.

PIGGOTT, S. 1954. The Neolithic cultures of the British Isles. London: Cambridge University Press.

REIMER, P.J., M.G.L. BAILLIE, E. BARD, A. BAYLISS, J.W. BECK, P.G. BLACKWELL, C. BRONK RAMSEY, C.E BUCK, G.S. BURR, R.L. EDWARDS, M. FRIEDRICH, P.M. GROOTES, T.P. GUILDERSON, I. HAJDAS, T.J. HEATON, A.G. HOGG, K.A. HUGHEN, K.F. KAISER, B. KROMER, F.G. McCORMAC, S.W. MANNING, R.W. REIMER, D.A. RICHARDS, J.R. SOUTHON, S. TALAMO, C.S.M. TURNEY, J. VAN DER PLICHT & C.E. WEYHENMEYER. 2009. IntCal09 and Marine 09 radiocarbon age calibration curves, 0-50 000 years cal BP. Radiocarbon 51(4): 1111-50.

RICHARDS, M.P. 2000. Human consumption of plant foods in the British Neolithic: direct evidence from bone stable isotopes, in A.S. Fairbairn (ed.) Plants in Neolithic Britain and beyond (Neolithic studies seminar paper 5): 123-35. Oxford: Oxbow.

RICHMOND, A. 1999. Preferred economies: the nature of the subsistence base throughout mainland Britain during prehistory (British Archaeological Reports British series 290). Oxford: Archaeopress. ROBINSON, M.A. 2000. Further considerations of Neolithic charred cereals, fruits, and nuts, in A.S. Fairbairn (ed.) Plants in Neolithic Britain and beyond (Neolithic studies seminar paper 5): 85-90. Oxford: Oxbow.

ROTTOL, M. & E. CASTIGLIONI. 2009. Prehistory of plant growing and collecting in northern Italy, based on seed remains from the early Neolithic to the Chalcolithic (c. 5600-2100 cal BC). Vegetation History and Archaeobotany 18: 91-103.

ROWLEY-CONWY, P. 2004. How the West was lost: a reconsideration of agricultural origins in Britain, Ireland and southern Scandinavia. Current Anthropology 45(S4): S83-S113.

--2011. Westward ho! The spread of agriculture from Central Europe to the Atlantic. Current Anthropology 52(S4): S431-S51.

SCHIBLER, J. & S. JACOMET. 2010. Short climatic fluctuations and their impact on human economies and societies: the potential of the Neolithic lake shore settlements in the Alpine Foreland. Environmental Archaeology 15(2): 173-82.

SHENNAN, S. & K. EDINBOROUGH. 2007. Prehistoric population history: from the late Glacial to the Late Neolithic in Central and Northern Europe. Journal of Archaeological Science 34: 1339-45.

SHERIDAN, A. 2011. The Neolithization of Britain and Ireland: the 'big picture', in B. Finlayson & G. Warren (ed.) Landscapes in transition: 89-105. Oxford: Oxbow.

STEVENS, C. J. 2007. Reconsidering the evidence: towards an understanding of the social contexts of subsistence production in Neolithic Britain, in S. Colledge & J. Conolly (ed.) The origins and spread of domestic plants in Southwest Asia and Europe: 375-90. Walnut Creek (CA): Left Coast Press.

THOMAS, J. 1999. Understanding the Neolithic. A revised second edition of rethinking the Neolithic. London: Routledge.

WENDOF, F., R. SCHILD, A.E. CLOSE, D.J. DONAHUE, A.J.T. JULL, T.H. ZABEL, H. HANNA WIECKOWSKA, M. KOBUSIEWICZ, B. ISSAWI & N. EL HADIDI. 1984. New radiocarbon dates on the cereals from Wadi Kubbaniya. Science 225: 645-46.

WINTERHALDER, B. & D.J. KENNETT 2006. Behavioral ecology and the transition from hunting and gathering to agriculture, in D.J. Kennett & B. Winterhalder (ed.) Behavioral ecology and the transition to agriculture: 1-21. Berkeley (CA): University of California Press.

Supplementary material is available online at www.antiquity.ac.uk/ProjGall/stevens333/

Chris J. Stevens (1) & Dorian Q Fuller (2)

(1) Wessex Archaeology, Portway House, Old Sarum Park, Salisbury SP4 6EB, UK (Email: [email protected])

(2) UCL Institute of Archaeology, 31-34 Gordon Square, London WC1H 0PY, UK (Email: [email protected])

Table 1. Numerical summary of radiocarbon dates. Seed
(cultivated) include flax, grape, pea and broad bean. Seed (wild)
comprises sloe stones and crab apple pips. Nuts are all hazelnut
and acorns. Numbers of intrusive cereals are based on authors'
comments and includes material that dated to later than 700 cal
BC, but was initially reported as Neolithic/Bronze Age.

Period/material cereal seed (cult) nut

Early Mesolithic 0 0 37
Early Mesolithic/Late Mesolithic 0 0 52
Late Mesolithic 1 0 55
Late Mesolithic/Early Neolithic 0 0 3
Early Neolithic 52 3 109
Early Neolithic/Middle Neolithic 36 1 15
Middle Neolithic 14 0 25
Middle Neolithic/Late Neolithic 4 0 22
Late Neolithic 2 0 9
Late Neolithic/Early Bronze Age 20 0 11
Early Bronze Age 9 0 5
Early Bronze Age/Middle Bronze Age 19 0 3
Middle Bronze Age 45 1 5
Middle Bronze Age/Late Bronze Age 41 2 0
Late Bronze Age 45 3 0
Late Bronze Age/Early Iron Age 17 0 0
Late Bronze Age to Middle Iron Age 6 0 0
Early/Middle Iron Age 1 0 0
Late Iron Age/Romano-British 1 0 0
Romano-British 0 0 1
Romano-British/Saxon 3 0 0
Saxon 3 0 0
Saxon/medieval 5 0 0
Medieval 7 0 0
Medieval/post-medieval 9 4 0
Post-medieval/modern 10 4 0

TOTAL 350 18 352

Region/material cereal seed (cult) nut

South-east England (30 sites) 55 3 29
South-west England (15 sites) 38 6 18
Central England (26 sites) 48 0 27
East Anglia (13 sites) 10 0 39
North England (17 sites) 27 6 70
Scotland (44 sites) 62 2 90
Wales (9 sites) 1 0 13
Ireland (13 sites) 15 0 40
Scottish Isles (26 sites) 80 0 26
Channel Islands (4 sites) 4 1 0
Isle of Man (1 site) 10 0 0
TOTAL 350 18 352

Period/material seed (wild) Intrusive

Early Mesolithic 0 0
Early Mesolithic/Late Mesolithic 1 0
Late Mesolithic 0 0
Late Mesolithic/Early Neolithic 0 0
Early Neolithic 2 0
Early Neolithic/Middle Neolithic 0 0
Middle Neolithic 0 0
Middle Neolithic/Late Neolithic 1 0
Late Neolithic 1 0
Late Neolithic/Early Bronze Age 1 0
Early Bronze Age 0 0
Early Bronze Age/Middle Bronze Age 0 1
Middle Bronze Age 1 4
Middle Bronze Age/Late Bronze Age 0 1
Late Bronze Age 0 0
Late Bronze Age/Early Iron Age 0 0
Late Bronze Age to Middle Iron Age 0 0
Early/Middle Iron Age 0 1? *
Late Iron Age/Romano-British 0 1? *
Romano-British 0 1 nut *
Romano-British/Saxon 0 2+1? *
Saxon 0 3 *
Saxon/medieval 0 5
Medieval 0 7
Medieval/post-medieval 0 13
Post-medieval/modern 0 14

TOTAL 7 47

Region/material seed(wild) Intrusive

South-east England (30 sites) 5 7
South-west England (15 sites) 0 8
Central England (26 sites) 1 4
East Anglia (13 sites) 0 3
North England (17 sites) 0 16
Scotland (44 sites) 1 6
Wales (9 sites) 0 0
Ireland (13 sites) 0 0
Scottish Isles (26 sites) 0 2+1 nut
Channel Islands (4 sites) 0 0
Isle of Man (1 site) 0 2
TOTAL 7 47

Period/material Total by period

Early Mesolithic 37
Early Mesolithic/Late Mesolithic 53
Late Mesolithic 56
Late Mesolithic/Early Neolithic 3
Early Neolithic 166
Early Neolithic/Middle Neolithic 52
Middle Neolithic 39
Middle Neolithic/Late Neolithic 27
Late Neolithic 12
Late Neolithic/Early Bronze Age 32
Early Bronze Age 14
Early Bronze Age/Middle Bronze Age 22
Middle Bronze Age 52
Middle Bronze Age/Late Bronze Age 43
Late Bronze Age 48
Late Bronze Age/Early Iron Age 17
Late Bronze Age to Middle Iron Age 6
Early/Middle Iron Age 1
Late Iron Age/Romano-British 1
Romano-British 1
Romano-British/Saxon 3
Saxon 3
Saxon/medieval 5
Medieval 7
Medieval/post-medieval 13
Post-medieval/modern 14

TOTAL 727

Region/material Total by region

South-east England (30 sites) 92
South-west England (15 sites) 62
Central England (26 sites) 76
East Anglia (13 sites) 49
North England (17 sites) 103
Scotland (44 sites) 155
Wales (9 sites) 14
Ireland (13 sites) 55
Scottish Isles (26 sites) 106
Channel Islands (4 sites) 5
Isle of Man (1 site) 10
TOTAL 727

* In some cases it is probable that dated material was not
intrusive, but other dating evidence residual.