Elsevier

Journal of Archaeological Science

Volume 51, November 2014, Pages 216-224
Journal of Archaeological Science

The impact of the Neolithic agricultural transition in Britain: a comparison of pollen-based land-cover and archaeological 14C date-inferred population change

https://doi.org/10.1016/j.jas.2012.10.025Get rights and content

Abstract

Britain's landscapes were substantially transformed as a result of prehistoric agricultural clearance and deforestation. This process began in the Neolithic and is recorded in multiple different “archives”, notably those deriving from archaeological site excavations and from off-site pollen records. This paper assesses the extent to which these two independent sources show common trends and timing in terms of demographic and environmental change across Britain during the millennia prior to and after the appearance of the first farming communities. This period is analysed within the wider context of the 9000–3400 cal. BP time frame. We compare land-cover change aggregated from 42 pollen records employing a pseudo-biomisation approach with radiocarbon (14C) date probability density functions from archaeological sites, which have been inferred to indicate shifts in population density. We also compare these results with selected palaeoclimate records in order to test alternative drivers of landscape change. At a broad geographical scale, pollen and archaeological records reveal very similar phases between 9000 and 3400 cal. BP. Following an initial demographic shift and landscape opening during the Late Mesolithic (∼7600 cal. BP) conditions were stable until 6400 cal. BP. Around 6400–6000 cal. BP (Mesolithic–Neolithic transition) a new phase began of forest disturbance and population increase. By 6000–5300 cal. BP early Neolithic population growth is clearly evident in the archaeological record with significant impacts on woodland cover, which is evident in the pollen record, reaching a maximum between 5700 and 5400 cal. BP. Between 5300 and 4400 cal. BP the archaeological record is inferred to indicate reduced mid-late Neolithic landscape impact, and this is matched by evidence of woodland re-establishment in the pollen record. Between 4400 and 3400 cal. BP renewed late Neolithic woodland clearance coincided with further population increase, which continued into the early Bronze Age. A very similar pattern is evident using a smaller-scale dataset from Scotland alone. Thus, rather than being slow and progressive, the initial Neolithic cultural transformation of Britain's landscape was relatively rapid and widespread; however, after several centuries of expansion, this process was halted and even reversed, before recommencing during the later 3rd millennium BC (after ∼4500 cal. BP). The impact of Neolithic forest clearance is clearly detectable as a driver of regional-scale mid-Holocene landscape change, alongside variations in climate. The Neolithic agricultural transition began a long process of anthropogenically-driven land-cover change in the British Isles, which has continued up until the present-day.

Highlights

► Spatially-aggregated demographic and land-cover change for the whole of Britain. ► Transformed pollen records are compared with archaeological 14C date densities. ► Neolithic population growth was accompanied by decreased woodland cover. ► Human impact is the main pace-maker of mid-Holocene land-cover change.

Introduction

Prehistoric agricultural land clearance, burning and deforestation have transformed the landscapes of the British Isles since the Mesolithic–Neolithic agricultural transition around 6000 years ago (∼4000 BC). Evidence of cultural landscape transformation has been preserved in numerous different archives, including archaeological sites and fossil pollen records, which provide information on past vegetation and land-use change. The expansion of agriculture in Europe was strongly linked to the Neolithic demographic transition (Bocquet-Appel et al., 2012). Both demic diffusion by migration and cultural diffusion of ideas are thought to have been important factors in initiating this agricultural shift (Lemmen et al., 2011). The Mesolithic–Neolithic transition in the British Isles (see e.g. Darvill (2010) for standard definitions of these and other period names in Britain) is characterised by a number of key environmental changes and economic shifts (Cayless and Tipping, 2002). These include anthropogenic vegetation disturbance, fluctuating patterns of burning, and the elm (Ulmus) decline (∼5650–5600 cal. BP). Cereal cultivation was one of the earliest elements of Neolithic life to be introduced to Northwest Europe, with the arrival of farming occurring over a 200–300 year period around 6000 BP. Whittle et al. (2011) describe the process of spread of the Neolithic in Britain as time-transgressive and regionalised, beginning in southeast England and taking over two centuries to spread to other areas. However, Bocquet-Appel et al. (2012) demonstrate that compared with most other regions of Europe the spread of farming to and then through Britain was extremely rapid and must have involved sea routes, not just overland expansion (cf. Collard et al., 2010). There are numerous factors that are likely to have influenced the nature and spread of the agricultural transition, and these will have had different impacts and consequences in different regions. Among them, Gronenborn (2009) suggests that the emergence and spread of farming resulted from climate-induced crisis periods.

Research that attempts to address whether the Neolithic transition resulted from demic migration or cultural diffusion, and whether this was gradual or more rapid, has been limited by a focus on single sources of spatially-aggregated data, either archaeological or palaeoecological. Whittle et al. (2011), for example, commented that few archaeological studies of Neolithic subsistence incorporate pollen data, partly due to the geographical disjunction between the varying sets of evidence. Each approach on its own has potential biases and limitations. For example, archaeological radiocarbon (14C) dates derive from a combination of different contexts and site types including domestic settlements, burials and monumental sites, and the balance between these varies temporally and spatially within the archaeological record in later prehistory. Similarly, pollen-inferred vegetation changes during the mid-Holocene had multiple drivers, including climate change and long-term ecological dynamics, as well as anthropogenic impact. It has also been argued (e.g. Brown, 1997) and demonstrated in simulation experiments (Tipping et al., 2009) that regional pollen diagrams lack sufficient spatial resolution to detect isolated clearings and the effects of subtle human impacts. In order to overcome these archive-specific biases, in this study we compare two approaches – pollen-inferred land-cover reconstruction and 14C date distributions from archaeological sites – for Britain as a whole. One advantage of this multi-archive approach is that changes in vegetation and land-use can be directly compared with inferred demographic change on a common calibrated radiocarbon timescale. We also compare these datasets with selected palaeoclimate records in order to explore the prospective roles of human activity and climate in influencing Holocene land-cover change in Britain.

Section snippets

Pollen-inferred land-cover change

Various approaches have been employed to utilise pollen in reconstructing land-cover change (e.g. Behre, 1986; Berglund, 1991; Gaillard et al., 1994; Sugita et al., 1999; Sugita, 2007; Favre et al., 2008). Within the pseudo-biomisation (PBM) approach used here, pollen count data from 14C-dated sediment sequences are transformed into records of natural and culturally-modified land-cover change according to the shifting dominance of different taxa through time (Fyfe et al., 2010). This method

Britain

Significant changes in the pollen-inferred land-cover record are identified as those that persisted for more than one 200 year time slice, and phases that correspond with changes in inferred-population are highlighted. The land-cover reconstruction for the whole of Britain over the last 9000 years (Fig. 3a) shows an overall trend from dominance of deciduous woodland during the early Holocene to increasing landscape openness and diversification during the late Holocene. The percentage of pollen

The Neolithic agricultural transition in Britain

Common trends and timing in land-cover and population density change for the period 9000–3400 cal. BP imply that increases in human population levels and the advent of Neolithic farming systems were the primary cause of the mid-Holocene decline in woodland and increase in open land in Britain. Although this conclusion is far from new at a local scale, it has not been universally accepted at the scale of Britain as a whole, where climate change has been seen by some authors as the most important

Conclusion

The comparisons between pollen-inferred land-cover and archaeo-demographic change presented here have implications for understanding of the Neolithic transition and for the approaches used to investigate anthropogenic activity and environmental change during the mid-Holocene. The multi-archive approach employed here has allowed large-scale human population and vegetation change to be investigated together and provides insights into the impacts of changing demography upon land-cover. The close

Acknowledgements

The development of the pseudo-biomisation approach was supported through Plymouth University Geography PUP funds and Leverhulme Trust grant F/568W. The radiocarbon date work by Downey, Edinborough and Shennan was funded by European Research Council Advanced Grant #249390 for the project EUROEVOL: Cultural Evolution of Neolithic Europe. Thanks are also given to contributors to the European Pollen Database and the authors of the palaeoclimate datasets used in this paper. We are also grateful for

References (44)

  • R. Tipping et al.

    Modelling land use around an early Neolithic timber ‘hall’ in north east Scotland from high spatial resolution pollen analyses

    Journal of Archaeological Science

    (2009)
  • L. Verrill et al.

    Bronze Age agriculture at Belderg Beg, Co. Mayo, Ireland: a palynological and geoarchaeological investigation

    Journal of Archaeological Science

    (2010)
  • A.N. Williams

    The use of summed radiocarbon probability distributions in archaeology: a review of methods

    Journal of Archaeological Science

    (2012)
  • K.E. Behre

    Anthropogenic Indicators in Pollen Diagrams

    (1986)
  • K.E. Behre

    The interpretation of anthropogenic indicators in pollen diagrams

    Pollen et Spores

    (1981)
  • B.E. Berglund

    The cultural landscape during 6000 years in Southern Sweden: the Ystad Project

    Ecological Bulletins

    (1991)
  • A.G. Brown

    Clearances and clearings: deforestation in Mesolithic/Neolithic Britain

    Oxford Journal of Archaeology

    (1997)
  • A. Brown

    Dating the onset of cereal cultivation in Britain and Ireland: the evidence from charred cereal grains

    Antiquity

    (2007)
  • C.J. Caseldine

    Archaeological and environmental change on prehistoric Dartmoor – current understanding and future directions

    Quaternary Proceedings

    (1999)
  • S.M. Cayless et al.

    Data on mid-Holocene climatic, vegetation and anthropogenic interactions at Stanshiel Rig, Southern Scotland

    Vegetation History and Archaeobotany

    (2002)
  • T. Darvill

    Prehistoric Britain

    (2010)
  • J.A. Fossitt

    Late Quaternary vegetation history of the Western Isles of Scotland

    New Phytologist

    (1996)
  • Cited by (0)

    View full text