Farming regimes in Neolithic Europe: gardening with cows and other models

Full text

is pdf of your paper in e Dynamics of Neolithisation
belongs to the publishers Oxbow Books and it is their
copyright.
As author you are licenced to make up to 50 offprints
from it, but beyond that you may not publish it on the
World Wide Web until three years from publication
(September 2014), unless the site is a limited access
intranet (password protected). If you have queries about
this please contact the editorial department at Oxbow
Books (editorial@oxbowbooks.com).

An offprint from
e Dynamics of Neolithisation in Europe
Studies in honour of Andrew Sherratt
Edited by
Angelos Hadjikoumis, Erick Robinson and Sarah Viner
© OXBOW BOOKS 2011
ISBN 978-1-84217-999-4

Contents
Acknowledgements ................................................................................................ vii
Contributors ...........................................................................................................ix
Foreword .................................................................................................................xi
John O’Shea
Introduction: the dynamics of neolithisation in Europe ...........................................1
Erick Robinson, Angelos Hadjikoumis, and Sarah Viner
1. Grand narratives and shorter stories ............................................................... 10
Alasdair Whittle
2. In constant motion? Recent advances in mathematical modelling
and radiocarbon chronology of the neolithisation of Europe ......................... 25
Marc Vander Linden
3. Time is on my side… .................................................................................... 46
João Zilhão
4. e Neolithic Revolution: an ecological perspective ...................................... 66
John C. Barrett
5. Farming regimes in Neolithic Europe: gardening with cows
and other models ........................................................................................... 90
Valasia Isaakidou
6. Interpretation of Scirpus from early farming sites in western Asia
and Europe: a cutting sedge of archaeobotanical research? ............................113
Michael Charles
7. Farming, material culture, and ideology: repackaging the Neolithic
of Greece (and Europe) .................................................................................131
Paul Halstead
8. Enchantment and enchainment in later Balkan prehistory: towards
an aesthetic of precision and geometric order ............................................... 152
John Chapman
9. Clutching at straw: the Early Neolithic and the dispersal
of agriculture ................................................................................................176
Anthony J. Legge and Andrew M. T. Moore

10. ‘Pig-menting’ the Spanish Neolithic..............................................................196
Angelos Hadjikoumis
11. e [environ-]mental contexts of earliest Neolithic settlement
and architecture in western Hungary ........................................................... 231
Eszter Bánffy and Pál Sümegi
12. Farming practice and society in the central European Neolithic
and Bronze Age: an archaeobotanical response to the
secondary products revolution model ........................................................... 266
Amy Bogaard
13. Technological traditions and ‘the dialectic of expansion’: contact,
transmission, and neolithisation along the northwestern
fringes of the LBK ....................................................................................... 284
Erick Robinson
14. Cattle and pig husbandry in the British Neolithic ....................................... 313
Sarah Viner
15. Tracing the future in the past: the introduction of the Neolithic
in eastern Scania – tracking change in a local perspective ............................ 353
Anna-Karin Andersson
16. Early farming and the creation of community:
the case of northern Europe ..........................................................................364
Magdalena S. Midgley

Introduction
‘No fodder no cattle; no cattle, no manure; no manure, no crop’
Ancient Tamil proverb ( omas 2000, 4)
 e nature of early farming in Europe has been central to a wide range of archaeological
debates. For Clark (1952), shifting ‘slash-and-burn’ cultivation was an adaptation
by early farmers to the densely wooded landscape of central Europe, while changing
frequencies of animal domesticates (from pigs to sheep) were a response to progressive
expansion of agriculture and clearance of woodland. Farming regimes have also been
inferred from or used to explain settlement patterns (e.g. Childe 1952; Sherratt 1980);
landscape change – physical (e.g. van Andel and Runnels 1995; Willis and Bennett
1994) and cultural (e.g. Barrett 1994; Bogaard 2005; Whittle 1996); the transmission of
material culture (e.g. Childe 1957); and, more generally, social interaction (e.g. Cullen
1984; Shennan 1986). Agricultural innovations, such as ‘Mediterranean Polyculture’
(Renfrew 1972) and the ‘Secondary Products Revolution’ (Sherratt 1981), have been
proposed as preconditions of social change. Farming regimes are thus implicated in
our attempts to understand a wide range of past human behaviour.
Although farming regimes proposed by prehistorians exhibit considerable diversity
in detail, a few basic models have been widely discussed and are the focus of this paper:
shifting ‘slash-and-burn’ (e.g. Clark 1952), ‘fl oodplain’ cultivation and the ‘secondary
products revolution’ (Sherratt 1980; 1981), ‘garden’ farming and extensive agriculture
(Halstead 1981). Despite their application to a variety of archaeological contexts, the
spatio-temporal relevance and broader implications of these models are by no means
agreed. For example, shifting slash-and-burn, fl oodplain cultivation and garden farming
have each been proposed as the exclusive farming regime of Early Neolithic central
Europe. Moreover, as Bogaard notes (2002, 156; 2004a, 26), the shifting slash-and-burn
model has been used to support diametrically opposed views of neolithisation: rapid
colonisation by migrant farmers from the east (e.g. Childe 1957, 15, 85, 111; Clarke
1952) and adoption by indigenous hunter-gatherers (e.g. Whittle 1996). Consensus
on the nature and signifi cance of early farming has yet to be achieved.
5
Farming regimes in Neolithic Europe: gardening
with cows and other models
Valasia Isaakidou

91
5. Farming regimes in Neolithic Europe: gardening with cows and other models
Although the importance of bioarchaeological (archaeobotanical and faunal) as well
as palaeoecological evidence has long been recognised (e.g. Clark 1952), early studies
of farming regimes had access to very little such data, essentially the presence/absence
and perhaps relative proportions of crop and livestock species. Domestic plants and
animals have broad tolerances, however, so species composition alone is not very
informative. As a result, the key models of farming regime have usually been defended
in terms of rather indirect, and often ambiguous, evidence: site density, longevity and
location; anthropogenic impact in the alluvial or palynological record; artefactual or
iconographic evidence for technological knowhow. In recent decades, however, the
volume and variety of bioarchaeological evidence has increased dramatically, as has
the range of tools for interpreting it.  e key to the latter advance has been a series of
actualistic (ethnographic, ecological and experimental) studies by bioarchaeologists of
pre-mechanised cultivation and herding in Europe and the Near East: for example, of
kill-off patterns and husbandry goals in Turkish sheep and goats (Payne 1973); work-
related skeletal pathologies of draught cattle in Hungary and Rumania (Bartosiewicz et
al. 1997); crop processing in the Near East (Hillman 1984a) and Greece (Jones 1984);
intensive cultivation of pulses in Euboea (Jones et al. 1999) and cereals in Asturias
(Charles et al. 2002); crop rotation in Jordan (Bogaard et al. 1999); experimental
cultivation of forest clearings in Germany (Bogaard 2002).
In addition to these signifi cant developments of bioarchaeological methodology
or ‘middle-range theory’ (Binford 1977), considerable eff ort has been invested in
ethnoarchaeological investigation of the practical details and socio-economic context of
pre-mechanised farming.  ese latter studies, involving observation and/or interviewing
of experienced practitioners, have been conducted both independently (e.g. Forbes
1982; Halstead 1990; Koster 1977) and in tandem with fi eld-projects devoted to
middle-range research (e.g. Halstead 1987; 1998a; 2006a; Halstead and Jones 1989;
Hillman 1984b; 1985).
Insights from such ethnoarchaeological studies of farming practices and regimes have
been integrated productively with new middle-range tools in some recent re-evaluations
of early farming in Europe and southwest Asia (e.g. Bogaard 2004a; 2005; Bogaard and
Jones 2007). In general, however, prehistorians may be more willing to absorb new
bioarchaeological evidence for individual farming practices than to engage in critical
evaluation of broader farming regimes.  is failing underlies the problem discussed by
Bogaard and noted above: that radically diff erent farming regimes may be proposed for
a single spatio-temporal context (thus claiming support from the same empirical data)
and radically diff erent social or historical implications may be claimed for the same
farming regime.  e present chapter explores this failing.  e following section reviews
some of the most commonly discussed models for the farming regime(s) of Neolithic
and Bronze Age Europe. Critical attention is drawn to: the origins of these models,
especially the extent to which they are based on actualistic studies of working farming
systems in ecologically and historically relevant contexts; the level of detail provided

92 Valasia Isaakidou
for the practical workings and wider social ramifi cations of these models; and, related
to this, the degree to which these models have been subjected to empirical testing or
adopted uncritically because of their consonance with some broader vision of historical
process.  e second part of this chapter then considers a body of ethnographic data
on small-scale farming in southern Europe as the basis for a proposed revision of the
model of Neolithic land use that appears best to have withstood both critical evaluation
of its ecological viability and empirical testing.
Dissecting models of farming regimes
Shifting ‘slash-and-burn’ cultivation
In advocating slash-and-burn cultivation for Early Neolithic temperate Europe, Clark was
infl uenced by the (then recent) palynological demonstration that early farmers faced a
heavily wooded landscape. Regarding clearance of woodland as the key challenge facing
early farmers, he turned to the ethnohistoric analogies of recent farmers in north European
birch and coniferous forests and European pioneers in Canada.  ese recent forest farmers
illustrated how ring-barking and felling trees could have opened up the forest canopy and
how ash from burning of cleared vegetation could have provided nutrients to crops. Clark
noted that north European slash-and-burn took place in a ‘peripheral region’ by farmers
also cultivating permanent fi elds, and, conversely, that the loess soils favoured by Early
Neolithic settlements were fertile, as well as easily worked. Nonetheless, he extrapolated
crop yields and regular shifting of plots, in response to rapid soil-nutrient depletion, from
recent slash-and-burn to the Early Neolithic (Clark 1952, 91–94).
Over the following decades, ethnographic examples of shifting slash-and-burn in
tropical areas were mobilised to elaborate, rather than critically evaluate, Clark’s model.
Boserup assigned shifting slash-and-burn to the beginning of an evolutionary scheme
from early extensive (low input/low yield per unit area) to late intensive (high input/
high yield per unit area) systems of cultivation (Boserup 1965, 15–6, 43–44). Growing
population provided the driving force behind progressive shortening of the fallow
period (from shifting ‘forest fallow’ to permanent cultivation) and the related advances
in technology (from primitive digging stick to sophisticated plough) and methods of
fertilisation (from natural regeneration to artifi cial manuring). A corollary of this scheme,
made clear by Grigg, is the lack of integration between crops and animals in slash-and-
burn farming. Like modern shifting cultivators in New Guinea who keep a narrow
range of domestic animals and do not manure their plots, therefore, Neolithic farmers
in Europe would have made limited use of livestock and their manure, making frequent
shifts of cultivated plots inevitable (Grigg 1974, 58, 62).  e argument (contradicted
by Clark’s ethno-historic analogues) that tree stumps in cleared plots impeded use of
the plough (Gilman 1981, 6) apparently strengthened the case for limited integration
of livestock in Early Neolithic farming.

93
5. Farming regimes in Neolithic Europe: gardening with cows and other models
Rowley-Conwy (1981) questioned slash-and-burn cultivation as a model for Early
Neolithic temperate Europe, on the grounds that the underlying ethnographic analogues
were drawn from irrelevant ecological and social contexts. Recent north European slash-
and-burn was a specialised response to highly marginal conditions and an opportunistic
adjunct to more permanent farming, as noted by Clark.  e tropical examples involved
climatic and edaphic conditions in which, following clearance, the exhaustion of
soils and regeneration of natural vegetation were extremely rapid (Rowley-Conwy
1981, 88–89). Conversely, long-term experimental evidence from southern England
demonstrated that, on fertile temperate European soils, the decline in crop yields and
increase in competition from weeds were slow, undermining the supposed agronomic
rationale for shifting of plots (Rowley-Conwy 1981, 89–91). Moreover, whereas pigs
were the only livestock kept by slash-and-burn cultivators in highland New Guinea,
the combination of cattle, sheep and pigs on Early Neolithic sites in temperate Europe
would have manured and dug abandoned plots and suppressed regeneration of forest,
so making shifting cultivation even less plausible (Rowley-Conwy 1981, 94–95). In
addition, shifting settlement among tropical horticulturalists seems to be a response
to intra- and inter-communal violence and the related depredations of domestic pigs,
rather than to the pressure of slash-and-burn on cultivable land (e.g. Brown 1978; Forge
1972; Rappaport 1968). Critical evaluation of ethnographic analogues, therefore, though
not ruling out slash-and-burn in the European Neolithic, surely makes it suffi ciently
implausible that it should not be entertained without serious empirical support.
Despite these cogent criticisms, the idea of shifting farming, and hence shifting
residence in the European and especially British Neolithic, proved surprisingly resilient.
Barrett (1994) adopted Boserup’s evolutionary model for farming (early shifting/
extensive to late permanent/intensive) to argue for changing forms of land tenure. Others
(e.g.  omas 1999; Whittle 1996) favoured shifting cultivation as a practice suited to
the mobile lifestyle and ideology of Mesolithic foragers and thus supporting spread of
farming through adoption rather than colonisation (Bogaard 2002, 156). More recently,
comparison of archaeobotanical weed assemblages with data from experimental shifting
slash-and-burn cultivation in Germany has fairly conclusively pointed to long-term
fi xed plots rather than shifting cultivation in the central European (Bogaard 2002;
2004a) and British Neolithic (Bogaard and Jones 2007). One proponent of shifting
cultivation has now conceded that ‘some people in Early Neolithic Britain cultivated
cereals, and that these would have been grown in small garden plots that might have
been quite long-lived’ ( omas 2008, 72, emphasis added).
It is tempting to conclude that the argument over shifting slash-and-burn cultivation
in the European Neolithic has fi nally been resolved empirically (but cf. Rösch et al.
2002), but the importance of critical engagement with this and other models of farming
regime must also be recognised. Just as Rowley-Conwy deconstructed the ecological
and social contexts of recent analogues to argue that slash-and-burn is implausible for
Neolithic Europe, so Bogaard explored the practical workings of this regime to identify

94 Valasia Isaakidou
relevant variables for testing the model. Conversely, recent proponents of slash-and-
burn may be criticised for ignoring, rather than refuting, Rowley-Conwy’s questioning
of its relevance and for assuming an undemonstrated linkage between slash-and-burn
cultivation and particular patterns of residential mobility and land tenure.
‘Floodplain’ cultivation and the ‘secondary products revolution’
Boserup’s model of agricultural intensifi cation and technological innovation, driven
by population growth, forms the explicit theoretical basis of Sherratt’s scheme for the
development of farming in prehistoric Europe. In this scheme, Early Neolithic fl oodplain
cultivation, an extensive regime (Sherratt 1972; 1980), gives way in the later Neolithic
to intensive plough agriculture, following diff usion to Europe from the Near East of
ox-drawn ard technology and of other ‘secondary products’ uses of livestock (Sherratt
1981; 1997; 2006).  e intellectual roots of fl oodplain cultivation likewise lie in the
Near East, where aridity rather than forest clearance has tended to be the limiting
factor on which models of early cultivation have focussed.  us, Flannery based on
the location of some early farming sites in areas apparently too dry for rain-fed crops,
coupled with the discovery of wet-loving Scirpus in early grain samples, suggested that
some early cultivation exploited locally high water-tables (Flannery 1969, 81).
Sherratt (1972) initially applied this model to Anatolia and southeast Europe, where
many Neolithic sites were known in old lake basins and especially near ‘backswamp
soils’ with high water table that did not require intensive tillage to encourage moisture
absorption. Selection of this restricted niche for cultivation channelled the expanding
population of early farmers into rapid colonisation westwards. Only in the Bronze Age
did farmers in the Mediterranean, now armed with the plough, expand out of this
favoured niche onto drier, more labour-demanding soils. Sherratt subsequently (1980;
1981) adapted this model to central Europe, particularly following the demonstration
by Kruk (1973 [cited in Sherratt 1981]; 1980) that Neolithic settlement in the loess
belt of southern Poland was initially concentrated along river valleys and only later
expanded onto the higher and drier interfl uves.
Sherratt and Kruk rejected slash-and-burn cultivation for Early Neolithic temperate
Europe, partly because of growing evidence that early (LBK) settlements (as opposed to
individual houses) were much long-lived and partly because concentration of habitation
along rivers removed both the opportunity (scope for repeated clearance) and need
(rapid depletion of soil fertility) for such a farming regime (Sherratt 1980, 315–16). As
well as expanding the geographical scope of the fl oodplain cultivation model, Sherratt
also now elaborated on how it would have worked. Annual fl ooding in late winter-early
spring replenished nutrients and water supply. As the ‘self-cultivating’ soils dried out,
cracking promoted aeration and seeds were spring-sown by broadcasting in the cracks,
rendering digging to plant and cover the seed redundant. Labour-intensive weeding,
manuring, digging, and even initial forest clearance would have been unnecessary.

95
5. Farming regimes in Neolithic Europe: gardening with cows and other models
Domestic animals, the numbers of which were limited by lack of clearance, did not
play an integral role in this regime (Sherratt 1980, 317–18).
Ethnohistoric analogy played a minor role in formulation of fl oodplain cultivation as
a model for prehistoric Europe. Sherratt noted Semple’s report of spring sowing of wheat
‘on the margin of the Lake Copais basin as the winter fl oodwaters gradually receded’,
in early twentieth century Greece (1980, 317). He regarded this as a possible relic of a
once widespread regime, though Semple herself made clear that autumn sowing was the
norm in the Mediterranean in both the recent past and classical antiquity.  e key to
fl oodwater farming was compatibility between the timing of annual fl oods and growth
season of crops (Semple 1932, 382–83). Sherratt’s model assumed spring sowing of
‘three-month’ cereal varieties after late winter-early spring fl oods. Many Mediterranean
watercourses are short and torrential, however, so that fl oods may recede in time for
spring sowing but leave behind coarse material inimical to crop growth.  e longer rivers
(e.g. the Pineios in central Greece or the Danube and its tributaries in the Hungarian
plain), on the other hand, swollen by melting snow in the mountains, tend to fl ood
too late in spring for cultivation of Old World grain crops (Semple 1932, 109–10). A
more universal problem is whether spring-sown cereals existed in the Neolithic: wild
cereals are eff ectively autumn-sown (Hillman 1981, 147), as Sherratt acknowledges,
and winter-sown domestic cereals have higher and more reliable yields than those sown
in spring (1980, 316–17). In the absence of archaeobotanical evidence either way, it
would have been safer to assume autumn/winter-sowing for the Neolithic, or at least
to acknowledge that the hypothesis of Neolithic fl oodplain farming depended on the
unproven assumption of spring-sown cereals. Sherratt was aware this assumption was
problematic (2005, 8), but adhered to the model, presumably because it ‘explained’
the distribution of Neolithic settlements and speed of neolithisation – though neither
phenomenon is necessarily related to crop cultivation regime. Finally, while the model
was originally formulated to explain how cereal cultivation might have been viable in
areas of the Near East too arid for rain-fed cultivation, Sherratt does not explain how
it would have been viable and advantageous through semi-arid southern Europe with
drought a periodic rather than constant hazard, to temperate Europe where excessive
rainfall may pose more problems than drought (cf. Halstead 1989).
Floodplain cultivation, together with its claimed support for a colonist model of
neolithisation, was revived for  essaly in central Greece by van Andel and Runnels
(1995).  ey reported recent spring sowing of cereals, on the drying margins of Lakes
Ioannina and Karla, as well as Copais, but omitted to mention that cultivation on
the edge of Karla represented occasional opportunism rather than regular practice (cf.
Halstead 1984, 97).  ey denied that spring fl ooding would destroy crops sown in
fl oodplains, although their confi dence is not shared by elderly farmers with fi rst-hand
experience of this hazard (Halstead 2002).  ey also argued that early farmers did have
cereals suitable for spring sowing in wet conditions, citing a modern emmer population
from highland northern Turkey (van Andel and Runnels 1995, 491).  is population

96 Valasia Isaakidou
has been modifi ed by selection under unusual and rather extreme growing conditions,
however, and is no guide to the ecology of Early Neolithic cereals (Bogaard 2004a,
163–64; 2005, 184; Davies and Hillman 1988, 24; Hillman 1981, 147). Like Sherratt,
they ultimately espoused fl oodplain cultivation because it appeared to ‘explain’ the
location and rapid spread of Early Neolithic sites. In practice, as had become much
clearer since Sherratt’s initial formulation of the model, many Early Neolithic sites in
Greece were not close to watercourses or lakes (Wilkie and Savina 1997) so fl oodplain
horticulture was, at best, just part of the story and the argument for rapid colonisation
of a restricted arable niche was unfounded.
As with slash-and-burn, growing archaeobotanical data and more eff ective methods
of analysis have signifi cantly undermined the case for fl oodplain cultivation. Scirpus
seeds in early grain samples are probably derived from animal dung rather than being
weeds of cereals grown in wet conditions (Charles 2005, 43; this volume) and weeds
with Early Neolithic crops in central Europe clearly imply that winter sowing was
the norm (Bogaard 2004a). Signifi cantly, Bogaard’s empirical rejection of fl oodplain
horticulture is based on investigation of sowing time, the variable focussed on in earlier
attempts to evaluate the ecological viability of this proposed regime.
 e second stage in Sherratt’s scheme of agricultural development was introduced by
a fourth-third millennium BC ‘secondary products revolution’, in which the diff usion
into Europe of new breeds, technology and knowhow led to domestic animals being
exploited not only for their meat, but also now for renewable milk, wool and traction
(Sherratt 1981). As already suggested by Boserup (1965) and Sherratt (1972), the
ox-drawn ard (scratch-plough) would have facilitated more frequent and intensive
tillage, necessitated by expansion of farming from ‘self-cultivating’, well-watered and
naturally rejuvenated fl oodplains onto less tractable, drier soils of more fi nite fertility.
 e associated expansion of clearance on interfl uves provided pasture for far more
livestock, while wool and milk provided new incentives for stock rearing.  e evidence
for secondary products innovations was primarily iconographic and artefactual (scenes
of milking; models of yoked animals; waterlogged yokes, ploughs and wheels), although
ard marks sealed under burial mounds provided more direct evidence for ploughing.
 e wider impact of secondary products (use of interfl uves, poorer soils and uplands
for plough agriculture and large-scale herding) was essentially inferred from evidence
for expansion of settlement, such as that in the loess belt of southern Poland (Sherratt
1981). Following Higham and Message (1969) and Payne (1973), Sherratt discussed
the potential of kill-off patterns to distinguish the aims (meat, milk, wool, traction)
of stock rearing, but relevant bioarchaeological evidence was scarce and of minimal
importance in documenting changing husbandry practices.
Compared with early fl oodplain cultivation, Sherratt’s presentation of this second
stage of agricultural development included fairly extensive recourse to ethnographic,
ecological and biological studies of modern populations: to evaluate the costs and
benefi ts of exploiting livestock for meat or milk (Sherratt 1981, 283–84); to identify

97
5. Farming regimes in Neolithic Europe: gardening with cows and other models
obstacles to milking primitive livestock and to adult digestion of milk (Sherratt 1981,
275–79); and to explore the wider socio-economic implications of the proposed switch
from manual cultivation with minimal herding to ard-agriculture and pastoralism
(Sherratt 1981, 194–96). Ecological and biological obstacles to the use of secondary
products were thus critically evaluated, but the empirical case for a secondary products
revolution and associated changes in farming regime was rather slender (Chapman 1982;
Isaakidou 2006).  e appeal of the model, both to Sherratt and sympathetic readers,
arguably lay in its consistency with preconceived notions of technological progress
and economic intensifi cation, driven by growing human population density and an
expanding urban world system. For those disinclined to investigate farming practices,
it off ered an attractively clear, off -the-peg springboard for more interesting forays into
human social behaviour.  e extent to which data were accommodated to the model,
rather than vice versa, became increasingly clear in later expositions (Sherratt 1997;
2006) that largely overlooked the growing body of bioarchaeological evidence (e.g.
Halstead 1998b; Legge 1981; Rowley-Conwy 1997; many others summarised in Vigne
and Helmer 2007 and Halstead and Isaakidou 2011) inconsistent with a fourth-third
millennium BC secondary products revolution as originally formulated.
‘Garden’ and extensive husbandry
Like Sherratt, Halstead (1981) proposed a model of early horticulture with manual
tillage followed by larger-scale agriculture coupled with ard-cultivation and animal
traction.  is model diverged from Sherratt, however, in proposed practical workings,
terminology, use of analogy, and attempted testing against available bioarchaeological
data. Although also applied, adapted or independently proposed for the prehistoric
west Mediterranean (Bernabeu et al. 1995, 269–96), Balkans (Dennell 1978; Halstead
1989) and central Europe (Bogaard 2004a; 2004b; Halstead 1989) and the early
historic Aegean (Gallant 1991; Garnsey 1988, 93–94; Hodkinson 1988), the model
was initially proposed for Neolithic and Bronze Age Greece. Accordingly, Halstead’s
starting point was a deconstruction of ‘traditional’ Mediterranean agriculture, dominated
by low-yielding cereal cultivation with regular fallowing, and pastoralism, only loosely
integrated with arable farming and dominated by large herds of sheep and goats that
spent the summer on mountain pastures.  is traditional regime had served, implicitly
or explicitly, as a model for the prehistoric Aegean (e.g. Jarman et al. 1982; Renfrew
1972). Halstead briefl y dismissed shifting slash-and-burn as ecologically irrelevant to
the Mediterranean, on similar grounds to Rowley-Conwy for central Europe (Halstead
1981, 318–19). In outlining a generalised model of land use in prehistoric Greece,
he also ignored fl oodplain cultivation (relevant, at best, to some sites in some regions
– above), though he later questioned the viability of such a regime (Halstead 1984;
2000, 117; 2002, 106).
Halstead instead assumed rain-fed cultivation of fertile and initially wooded soils.
He argued that the low yields, low seed:yield ratios, low soil fertility and prevalence

98 Valasia Isaakidou
of fallowing over crop rotation in traditional Mediterranean agriculture were products
of long-term cultivation from nucleated settlements of land often marginal for arable
farming. Traditional use of summer grazing in the mountains needed complementary
winter pasture in the lowlands, available only after extensive clearance for arable farming.
Traditional agriculture and pastoralism were thus shaped not only by Mediterranean
climate and topography, but also by historically contingent, anthropogenic or cultural
variables. Subsequent elaborations of this deconstruction emphasised inequalities of land
ownership and urban markets, rather than settlement nucleation, as the key cultural
variables (Halstead 1987; 2000).
Halstead argued that, in contrast to the context of traditional agriculture, Neolithic
‘villages’ were small, concentrated in relatively fertile and well-watered terrain, and so
could have been supported by cultivation on a small-scale in the immediate vicinity of
the settlement. Initial Neolithic cultivation would have resembled traditional in-fi eld
gardening more than out-fi eld agriculture, therefore, and would have been characterised
by cereal-pulse (rather than cereal-fallow) rotation, hand-cultivation and weeding,
dibbling rather than broadcast-sowing, and fertilisation by middening with household
refuse. Limited clearance would have confi ned small numbers of livestock to arable
land, especially at night, so ensuring close integration of crop and animal husbandry
and relatively intensive manuring. Such a gardening regime (similar to what Rowley-
Conwy proposed in place of slash-and-burn for central Europe) would have favoured
high yields and high seed:yield ratios, would have depleted very slowly the initially
high fertility of forest soils and so would have provided an ecologically and socially
stable subsistence base (Halstead 1981, 317–20).
 is gardening regime was disrupted around the end of the Neolithic, by settlement
nucleation. Larger communities demanded cultivation on a larger scale and hence
at greater distance from the settlement, so that labour-intensive weeding, manuring
and pulse growing were restricted to the nearer plots and low-yielding, cereal-fallow
agriculture developed further afi eld.  e expansion of fallow land would now have
created suffi cient winter grazing to support large, seasonally mobile fl ocks, favouring
looser integration between crop and animal husbandry and so reinforcing the low-
yielding and spatially extensive nature of cereal cultivation. Finally, Halstead followed
Sherratt in linking animal traction (for transport and tillage) to the proposed expansion
of cultivation, but emphasised its role as a solution to problems posed by settlement
nucleation (Halstead 1981, 334; also Gamble 1982) rather than as a stimulus to
agricultural change. Similarly, whereas Sherratt emphasised diff usion of technology,
the ard (Sherratt 1981) and draught-pole for attaching this to a pair of oxen (Sherratt
2006), Halstead stressed the high costs of maintaining draught cattle and consequent
inequality of access to plough teams (Halstead 1981, 329–30).
Halstead labelled Neolithic gardening intensive and Bronze Age agriculture extensive
because they were characterised, respectively, by high and low human labour input per
unit area (Bogaard 2004b, 50; Halstead 1992, 110).  is terminology avoided the

99
5. Farming regimes in Neolithic Europe: gardening with cows and other models
confusion inherent in Sherratt’s description of the shift, from (localised) ‘intensive’
fl oodplain cultivation to ‘extensive’ plough agriculture (1981, 292), as an ‘intensifi cation’
(1981, 262). It also highlighted the crucial role of draught oxen in Bronze Age and later
surplus production: oxen enabled a radical expansion in the scale of cultivation, thus
compensating for the lower area yields of extensive agriculture, coupled with a major
reduction in the amount of human labour for tillage and also for weeding (replaced
by cultivated fallow). On the other hand, this strategy of surplus production required
access to temporary harvest workers, making social inequality a precondition, as much
as consequence, of cultivation with oxen (Halstead 1995).
In addition to these diff erences of substance and terminology, the intensive gardening/
extensive agriculture model diverged from the slash-and-burn and fl oodplain cultivation/
secondary products revolution models in underlying methodology: its practicalities were
worked through in greater detail (and increasingly so in later versions); this elaboration
was based on more extensive analogical research (ethnographic, agronomic, ecological
and ethological); and these analogies were overwhelmingly drawn from the ecologically
relevant context of the Mediterranean. As with Rowley-Conwy’s critique of the slash-
and-burn model for temperate Europe, Halstead also critically evaluated the socio-
historical context (settlement nucleation, market economy, etc.) of recent analogues to
establish their potential relevance to prehistoric Greece. One consequence of Halstead’s
relatively detailed presentation was that it generated empirical expectations, in terms
of the relative proportions of cereal and pulse crops and of diff erent livestock species
(Halstead 1981), the degree of anthropogenic impact on vegetation and landscape
(Halstead 2000), and the relationship between these variables and settlement size
(as opposed to regional population density, emphasised by Sherratt and Boserup).
As a result, within the limitations imposed by low-resolution archaeobotanical and
faunal data and by the broad ecological tolerances of domestic crops and livestock,
this scheme was tested against bioarchaeological data to a much greater degree than
the slash-and-burn or fl oodwater/secondary products schemes. Another consequence,
arguably, of the greater detail and more thorough analogical foundations of Halstead’s
scheme is that Neolithic gardening, closely integrated with small-scale stock husbandry,
is supported by emerging evidence from new middle-range tools using arable weed
ecology (Bogaard 2004a; 2004b; 2005; Bogaard and Jones 2007) and livestock dental
microwear (Mainland 2003; Mainland and Halstead 2005).
Although more clearly formulated and thoroughly tested than previous models,
Halstead’s scheme has two partly related weaknesses. First, whereas his later ‘extensive
agriculture’ was modelled on traditional Mediterranean practice, the early ‘intensive
gardening’ was, like Sherratt’s fl oodplain cultivation, a largely extinct system of land use
no longer observable in its entirety. Rather the system comprised a suite of practices,
each well attested in pre-mechanised agriculture, but bolted together as a heuristic
alternative (Halstead 1981, 310; 2000, 113; Hodkinson 1988, 69) to traditional
extensive agriculture. Secondly, although the logic of such a heuristic opposition must

100 Valasia Isaakidou
be that past land use could fall anywhere on the continuum from intensive manual
gardening to extensive plough-agriculture, Halstead in practice accommodated the
admittedly coarse bioarchaeological data to one or other of the two extremes. His later
publications (Halstead 1995; 1999) off ered a more nuanced account, in which Late
Bronze Age palatial estates were ploughed with oxen, while the rest of the population
cultivated by hand or with draught cows, but he regarded traction animals as unnecessary
in small-scale Neolithic husbandry (Halstead 1989, 30).
Discussion
 is brief critical review of three of the most widely encountered models of early farming
in Europe highlights the very variable extent to which they were based on contextually
relevant, recent analogues and had been thought through in the detail necessary to
assess their practical viability, to determine suitable variables for empirical testing, and
to identify their likely wider implications for human society.  e popularity of the rival
models has not been commensurate either with the robusticity of their formulation or
with the strength of their empirical support, but seems rather to have been determined
by their apparent (and often circular) consistency with broader agendas of cultural and
social change.
Recent archaeobotanical investigation of weed assemblages suggests that Neolithic
cultivation in Europe was permanent rather than shifting, with intensive tillage
and fertilising rather than extensive husbandry, with primarily autumn- rather than
spring-sowing, and hence rain-fed rather than fl oodwater-irrigated. Faunal analysis
(taxonomic composition, kill-off patterns, biometry, and dental microwear) suggests
small-scale, mixed-purpose animal husbandry, more or less closely tied to arable land
(Halstead 2006b).  ere is also growing evidence of stress-related skeletal ‘pathologies’,
suggesting use of cattle as draught animals in Neolithic Europe, from Greece to Britain
and Scandinavia (Bogaard this volume; Halstead and Isaakidou 2011). Some at least
of this evidence relates to cows rather than oxen, especially at Neolithic Knossos on
Crete where a very large sample of ‘traction pathologies’ spans from the seventh to
the fourth millennia BC (Isaakidou 2004; 2006). Such evidence has in turn triggered
reconsideration of the role of draught-animals in Neolithic gardening (Bogaard 2005,
179; Isaakidou 2008), both as an aid to tillage and as a source of, and means of,
transporting stall-manure (Isaakidou 2008).
As yet, however, weed evidence for intensity of Neolithic cultivation is only abundant
for central Europe; taxonomic, demographic and biometric faunal data are widespread,
but dental microwear analyses restricted to southeast Europe; and pathological evidence
for draught-cattle is patchy in space and time, and much richer in Crete than elsewhere.
Ongoing research will doubtless fi ll many present gaps, and isotopic analysis of crop
seeds and of human and animal bone should provide an independent measure of
intensity of manuring and the role of manured plots in both human and animal
nutrition (Bogaard et al. 2007). However, individual categories of evidence will surely

101
5. Farming regimes in Neolithic Europe: gardening with cows and other models
continue to be patchy in availability, so there is no realistic prospect of reconstructing
early farming regimes purely inductively, any more than there is of fi nding the ‘right’
model by critical evaluation of analogues without testing against archaeological data.
For the foreseeable future, we will have to tack backwards and forwards between data
and models. In this spirit, the following section explores the role of draught cows, as
aids to tillage and transport and as producers of manure, in traditional farming in
southern Europe.
Cows in recent south European cereal farming
 e role of draught cows in recent southern European farming may usefully be compared
for two diff erent regions: Greece, more or less subject to a Mediterranean climate, with
drought the principal hazard to crop growth; and upland Asturias in northern Spain,
subject to an Atlantic climate where excessive summer rainfall may pose problems.
Greece
In a survey of published historical and ethnographic data from across the Mediterranean
and of fi rst-hand oral accounts from Greece, Halstead (1995) reported broadly consistent
fi gures for the costs and, especially, benefi ts of tillage by hand or with cattle-drawn
ards. Ploughing with a pair of cattle was estimated to be 2–15 times faster than manual
tillage by a single person – thus ensuring timely sowing of crops – but the considerable
costs of breeding (or buying) and maintaining working cattle restricted their use. Oxen
(castrated males) especially, provided only their labour – and, eventually, carcasses
– and tended to be kept by farmers cultivating 3–4 ha or more of fi eld crops (Halstead
1995, 16). Farmers with less land were more likely to plough with cows, which were
less powerful but also less costly (they bore their own replacements) and also provided
milk and/or calves (Halstead 1995, 17).
Subsequent interviews in many parts of Greece, by Halstead and the present author,
have confi rmed and elaborated on this picture, which is also consistent with published
information from other regions (e.g. Bala et al. 1992; Harris 1966; Lawrence and Pearson
2002; Mishra and Pandey 2000). Cows, being smaller and sometimes handicapped by
pregnancy or lactation, usually ploughed signifi cantly smaller areas and less thoroughly
(shallower tillage) than well-fed oxen, but remained signifi cantly more effi cient than
manual tillage. In practice, draught cows were more widely used than oxen in the recent
past, to a degree disguised by the tendency across the Mediterranean to describe all
working cattle (regardless of sex) as ‘oxen’ and to clarify sex only when specifi cally asked
to do so. Even a pair of cows could cultivate more than enough land for household
subsistence, while draught oxen were capable of surplus production on a large scale
– and were usually only considered worth keeping for this purpose.  e linkage between
use of draught cattle and land ownership is reinforced by the practicalities of providing

102 Valasia Isaakidou
fodder. Cattle needed something more nutritious than straw while working (and so
unable to graze) and across most of the Mediterranean this meant grain: roughly as
much per pair as 1–4 persons would consume in a year. Moreover, the grain consumed
by a family household might not yield the chopped straw needed as maintenance rations
for the cattle in winter, when not working.
Farmers able to support only a single cow joined forces with a neighbour in the same
position and, provided their land was not excessively heavy or infertile or distant, were
often self-suffi cient in staple grains. Both partners cultivated on a smaller scale than the
owner of a pair of cows: even if the draught animals had the strength for an extended
ploughing season, the onset of winter was likely to bring their labours to a halt. As in
other parts of the Mediterranean, however, much of the rural population was unable
to support any draught animal. Some households compensated for shortage of land by
intensive manual cultivation of cash crops in market gardens or vineyards. Others, in
hilly and mountainous regions, worked small, steep or boulder-strewn plots by hand
even for staple grains. For the most part, however, those owning land but not draught
animals hired the plough team of a neighbour, often in return for their own manual
labour or a share of the harvest. Manual cultivation of staple grains was avoided if
possible, underlining the extreme drudgery involved. Indeed, although self-suffi ciency
in grain is theoretically possible with manual cultivation, oral accounts suggest that
households would have struggled to achieve this, especially at unfavourable points in the
‘domestic cycle’ (low ratio of workers to consumers) and especially if the workforce was
undernourished or tired by other seasonal tasks. Moreover, neither manual cultivators
nor those dependent on hired plough-teams could ensure timely tillage so they risked
not completing sowing, if hostile winter weather began early, or sowing too late for a
reliable harvest.
Draught-cows were to some extent associated with more intensive husbandry
than oxen. First, farmers using cows were more likely to practice crop rotation than
fallowing, because fallowing presupposes ‘surplus’ land, because cows often lacked the
stamina to plough fallow as well as sown fi elds, and because cows could not break
heavy soils that had been left as unploughed fallow. Secondly, an ard pulled by cows
was often followed by human workers who broke clods and cultivated fi eld edges and
around boulders.  irdly, farmers ploughing with cows often hand-weeded their more
valuable crops in spring. Farmers with oxen were less likely to deploy manual tillage
or weeding, because they cultivated on a larger scale, more thoroughly, and on large
fi elds with fewer obstacles.
 e level of manuring tended to be limited by ownership of animals: some big
landowners ran large herds and, otherwise, their fallow fi elds attracted the fl ocks of
transhumant pastoralists in winter. Manure from grazing animals and stall litter were
almost invariably less than desired and, in southern Greece, the latter was often spread
mainly on vegetable gardens and cash crops, such as olives. Stall litter is also very heavy
and, if available for cereal crops, was usually spread on fi elds within a few hundred

103
5. Farming regimes in Neolithic Europe: gardening with cows and other models
metres of the byre – even when transported by ox-drawn carts in gentle terrain. When
a mild and wet growing season followed manuring, cereals often grew too vigorously
and, unless lightly grazed or mown to retard their development (Halstead 2006a), risked
‘lodging’ (fl attening by wind or rain). In the long term, however, manuring improved
yields and made heavy soils more tractable.
Asturias
 e role of cows in intensive ‘garden’ cultivation was explored in interviews of elderly
farmers, again by Halstead and the present author, in hill villages (400–800 m asl)
of central Asturias, an area with high rainfall (≤ 1200 mm) distributed fairly evenly
throughout the year and low temperatures, frost and snowfall during winter (Charles
et al. 2002). Farmers grew glume wheat (emmer and spelt) in rotation with maize
and potatoes.  e scale of cultivation was modest: perhaps 0.5–1.5 ha per household,
including 0.2–0.5 ha of emmer/spelt. Individual plots were small (ca. 0.04–0.3 ha) and
dispersed around the village – usually 10–30 minutes walk from the house. Emmer/spelt
crops were often weeded and benefi ted from intensive tillage and manuring associated
with the preceding maize or potato crop. Dung was mainly collected from byres in
the village, where working cows were stall-fed year-round together with any milking
or suckling cows; distant fi elds and meadows were manured from outfi eld byres that
housed other cows (those not working or lactating) grazing beyond the zone of arable
land.  anks to intensive tillage, heavy manuring and regular weeding, yields of emmer/
spelt ranged from 1 ton/ha of clean grain in a bad year to 2.5 tons/ha in a good year
– double the yields under extensive cultivation (with modest manuring and little or
no weeding) in lowland areas of the Mediterranean with intrinsically more fertile soils.
High emmer/spelt yields, coupled with growing of New World maize and potatoes and
with wage labour in local mines, enabled villagers to survive on very modest holdings
of rather infertile land.
Cows made several important contributions to this intensive regime. First, they
ploughed the grain fi elds, with manual labour normally playing just a supplementary
role (clod-breaking, weeding of growing crops).  e area notionally ploughed per ‘day of
oxen’ was 800 m2, at the low end of the range for draught-cows in the Mediterranean,
but still well in excess of what could be cultivated eff ectively by the manual labour alone
of two fi t adults. Purely manual cultivation seems to have been restricted to vegetable
plots and, recently, to ‘hobby’ farming on a similarly small scale of emmer/spelt.
Draught cows were also used for the lighter and quicker task of harrowing that was
particularly valuable to households lacking suffi cient manual labour for clod-breaking
and weeding.
Secondly, the same animals pulled sledges, which transported the harvest from
fi elds to village, and manure from byres to fi elds.  ese sledges also carried hay from
outlying meadows to village byres, enabling cows to work without grazing and produce
stall-manure. Of these tasks, transport of the modest emmer/spelt harvest was least

104 Valasia Isaakidou
burdensome, as the ears were reaped separately from the straw and were sometimes
carried by human labour.  e hay crop was much bulkier and its transport – in mid-
summer – from the meadows, which were further from the village than the arable plots,
sometimes brought draught-cows to their knees. Stall manure was also bulky and very
heavy: 10–25 sledge loads (depending on distance, soil quality, availability of manure
and extent to which this was bulked up with litter) were spread on a plot of 800 m2,
and a single load was said to weigh ca. 250 kg. A farmer with one hectare of arable
land, rotating un-manured emmer/spelt with manured maize or potatoes, would thus
have transported 15–40 tons of manure each year. In one day, a pair of cows might
manage several trips with the dung sledge to plots adjacent to the village, but only two
round-trips to more distant fi elds; only the nearest plots could be manured as rapidly
as they could be ploughed. Not surprisingly, this task was never attempted with just
human labour. Draught cattle thus played a central role in intensive farming and
transport occupied these animals for more of the year than tillage.
 irdly, cattle produced the stall manure used to maintain high levels of soil fertility.
Depending on the amount of stall-litter added, a stalled cow fi lled a dung-sledge
perhaps every 8–10 days and so should have produced something like 40 loads per
year – enough to fertilise well only 0.1–0.3 ha of land. Stall-manure of the working
pair was at best barely suffi cient, therefore, even for the modest scale of cultivation
normal in Asturias, and most farmers seem to have stall-fed something like 10 cows
per hectare of arable land.  ese additional cows provided milk, calves and also labour
when one of the regular working animals was injured or calving.
Discussion: contextual evaluation of recent practice and lessons for the Neolithic
Recent use of draught cows in both study areas was shaped by historical contingencies.
In Greece, even farmers with a pair of draught cows usually produced some surplus
to pay tithes, buy consumer goods, and provide for their children’s future. Conversely,
manual cultivation of staple grains usually supplemented plough agriculture or another
source of income, and was invariably regarded as a necessity rather than choice.
Moreover, given farmers’ widespread insistence on the importance of timely sowing,
manual cultivation alone would have been a risky strategy at unfavourable points in
the domestic cycle and especially in years when tardy autumn rains or early onset of
winter shortened the sowing season. Manual cultivation would have been more viable
(lighter and less time-stressed tillage on a smaller scale) with heavier manuring and, in
the absence of cash crops, more of the available manure would presumably have been
devoted to staple grains.
In upland Asturias, interviewees (or their parents) had bought much of their land
in the early twentieth century from absentee landlords, for whom they had previously
been share-croppers, tenants or day-labourers (e.g. Fernández 1988), but change of
ownership did not aff ect methods of cultivation greatly. More signifi cantly, rotation of
glume wheats with high-yielding maize and potatoes reduced the scale of cultivation

105
5. Farming regimes in Neolithic Europe: gardening with cows and other models
needed per household, as did employment in the mines and sale of calves. Despite the
small scale of crops (and hand tillage of plots too small for the plough), cultivation was
mostly undertaken with draught-animals.  is small-scale, intensive regime was only
viable, however, thanks to heavy inputs of stall-manure from cattle fed on hay from
outlying meadows; draught cows enabled transport of both hay and manure. In the
absence of New World crops, mining and a market for calves, cultivation would have
been on a larger scale and manuring less intensive (as required by Old World grain
crops), but poor soils on steep slopes and enforced winter-stalling of livestock would
arguably have maintained the need and opportunity for relatively intensive manuring
and so the need for draught animals.
A notional self-suffi cient Neolithic household probably cultivated grain crops on a
smaller scale (with fewer demands for overproduction beyond domestic subsistence)
than recent Greek farmers with draught cows, but on a similar or slightly larger
scale than their Asturian counterparts. Assessment of the feasibility of purely manual
cultivation in the Neolithic is diffi cult, as recent farmers utilised metal hand tools and
even when they did not own work animals, normally had the option of exchanging
their own labour for a borrowed or hired plough team. On the other hand, feeding
surplus grain to draught cows would certainly have reduced the drudgery and risk of
crop husbandry and facilitated overproduction for both risk buff ering and hospitality
(Isaakidou 2008). Feeding surplus grain to work animals would, as in the recent past,
have been an eff ective means of ensuring future production. Domestic cattle were a
normal component of the Early Neolithic ‘package’ of domesticates (perhaps partly
for this reason) and both yokes and ploughs were commonly home-made in the recent
past, suggesting that technological knowledge is unlikely to have been an obstacle to
adoption of cattle for tillage. In sum, there is no reason to preclude use of cows for
transport and ploughing from an early stage of the Neolithic, and there is good reason
to imagine that, depending on contingent ecological and social circumstances, such
use by some households would have been emulated by their neighbours or would
have led to asymmetrical exchanges of human and animal labour. Recent practice
in Greece and, especially, Asturias also makes clear that tillage with cows could have
reinforced rather than undermined intensive gardening of staple grains. Ironically,
manual cultivation would have been more practicable with high levels of manuring,
but draught cows would have revolutionised the capacity to distribute stall-manure.
Recent practices suggest that intensity of manuring is likely to have been a function
of availability (number and type of animals kept, degree of stall-feeding) and ability
to transport (access to draught-animals), and not of progress in agronomic know-how
(contra Boserup 1965).

106 Valasia Isaakidou
Conclusions
Models of early farming regimes have played a major role in attempts to understand
processes of neolithisation, the nature of early farming society and the dynamics of
cultural change through and beyond the Neolithic. Unfortunately, such models are often
treated uncritically, not as heuristic tools to be tested, but as historical facts accepted
as the basis for further signifi cant inferences (e.g. shifting cultivation – Johannsen
2006, 44;  omas 2003, 71; ‘secondary products revolution’ – Bakels 1997; Bogucki
1993). For example, Hedges and Reynard (2007, 1245) interpret high nitrogen values
in Neolithic human remains as refl ecting meat consumption, because ‘manuring
seems rather unlikely in the early Neolithic’. In support of this statement, they cite
Bakels (1997, 444), who sees stalling of draught or milk animals as a prerequisite for
manuring and so considers the latter unlikely to predate Sherratt’s ‘secondary products
revolution’ of the fourth-third millennium BC.  e shakiness of the foundations
of some popular models is made clear by the lack of agreement on which farming
regime(s) prevailed when and where and on what, if any, were the broader social
and ecological implications of each regime. Dispute as to which regime(s) prevailed
in particular spatio-temporal contexts may partly be attributed to the scarcity, until
recently, of high-quality bioarchaeological evidence and of actualistic studies designed
to aid its interpretation. On the other hand, some models of farming regime were
constructed so vaguely as to defy eff ective empirical testing, which instead took the
form of appeals to datasets (e.g. settlement patterns) only tangentially related to the
target regime.  e construction of models was often also fl awed in being based on
insuffi cient (e.g. fl oodwater farming) or contextually inappropriate (e.g. slash-and-burn)
analogical foundations. As a result, prehistorians have built grand narrative edifi ces
on assumed farming regimes that are at best ecologically implausible and empirically
groundless (e.g. slash-and-burn) and at worst so inviable that they would rapidly have
led any Neolithic practitioners to extinction (e.g. fl oodplain cultivation). One aim of
this chapter, therefore, has been to highlight the need to examine more critically the
structure and analogical underpinnings of competing models of Neolithic farming. In
addition to screening model regimes for viability in a Neolithic ecological and social
context and identifying relevant variables for empirical testing, such scrutiny exposes
the extent to which models are embedded in wider paradigms of technological advance
and/or demographic forcing, such as that of Boserup (Leach 1999; Morrison 1994),
originally formulated within a diff erent disciplinary context, with very diff erent aims
(Sunderlin 2003) to those concerning archaeologists.
Perhaps not fortuitously, intensive gardening is the farming regime that has best
withstood recent empirical testing for Neolithic Europe and also that most extensively
supported by ecologically relevant analogical research. Growing faunal evidence for
Neolithic draught-cattle, however, suggests that the original model may draw too sharp
a contrast between Neolithic gardening with hand tools and Bronze Age use (at least

107
5. Farming regimes in Neolithic Europe: gardening with cows and other models
by elite farmers) of animal-drawn ard-ploughs. Recent pre-mechanised cultivation in
Greece and upland Asturias makes clear the compatibility of draught cows with intensive
gardening. As aids to tillage and transport and as a source of stall-manure, draught
cows would have signifi cantly eased the human labour costs of Neolithic intensive
gardening and made such a regime viable on a larger scale, and thus compatible with
larger settlements, than manual labour alone. Accordingly, although Neolithic weed
evidence for intensive gardening and faunal evidence for draught cattle are presently
strongest in central Europe and Crete, respectively, there is no practical reason why
draught cattle should not have been widely integrated in a gardening regime. Recent
practice also highlights the importance of timely sowing of staple crops and, in this
respect, owners of draught teams were at a signifi cant advantage over those reliant
on manual labour or on hired, borrowed or shared work animals.  e constraints on
timing of sowing vary, however, with latitude. On Crete, prompt sowing is particularly
important, because severe summer drought both delays autumn sowing of grain crops
and curtails their growing season early and rapidly. In Asturias, by contrast, wet summers
leave a longer autumn-winter window for sowing and farmers may even delay sowing
to reduce the risk of excessive growth and lodging.
Future research will clarify how widely draught cows were embedded in Neolithic
gardening. Either way, recent experience highlights the considerable costs of feeding
working animals and suggests that ‘ownership’ thereof will have been uneven between
and within Neolithic communities. Maintenance of draught cattle would have been
discouraged where severe winters increased costs of stall-feeding, but encouraged where
large settlement size or a short sowing window exacerbated the costs and risks of manual
cultivation. In the recent past, uneven ownership of draught animals variously promoted
symmetrical collaboration, especially between owners of single draught animals, or
asymmetrical exchanges of human for animal labour, between owners of two or more
draught animals and owners of none.  e broader social implications of draught cattle,
therefore, will have been shaped both by practical constraints on the maintenance of
such animals and by ideological constraints on competition and collaboration within
Neolithic communities.
Acknowledgements
As a response to previous scholarship, this paper would not have been possible without the
stimulus and direction provided by those whose work it critiques and, in particular, the
inspirational models of Andrew Sherratt. I also owe a lot to Paul Halstead, who introduced
me to Asturias and taught me the value of listening to people with fi rst-hand experience
of non-mechanised farming. I am grateful to Amy Bogaard for comments on a draft of
this paper. Last but not least, I thank the numerous patient informants in Asturias and
Greece who shared with me their knowledge and wisdom on matters agricultural.

108 Valasia Isaakidou
Bibliography
van Andel, T. and Runnels, C. (1995)  e earliest farmers in Europe. Antiquity 69, 481–500.
Bakels, C. C. (1997)  e beginnings of manuring in western Europe. Antiquity 71, 442–45.
Bala, B. K., Satter, M. A. and Biswas, B. K. (1992) Utilization pattern of animal draft power
and cow dung in Bangladesh. Bioresource Technology 41, 1–8.
Barrett, J. C. (1994) Fragments from antiquity: an archaeology of social life in Britain (2900–1200
BC). Oxford, Blackwell.
Bartosiewicz, L., Van Neer, W. and Lentacker A. (1997) Draught cattle: their osteological
identifi cation and history. Tervuren, Musée Royal de l’Afrique Centrale.
Bernabeu, J., Aura, J. E. and Badal, E. (1995) Al oeste del Eden: las primeras sociedades agrícolas
en la Europa Mediterranánea. Madrid, Editorial Síntesis.
Binford, L. R. (1977) General introduction. In L. R. Binford (ed.) For theory building in
archaeology, 1–10. New York, Academic Press.
Bogaard, A. (2002) Questioning the relevance of shifting cultivation to Neolithic farming in
the loess belt of western-central Europe: evidence from the Hambach forest experiment.
Vegetation History and Archaeobotany 11, 155–68.
Bogaard, A. (2004a) Neolithic farming in central Europe. London, Routledge.
Bogaard, A. (2004b)  e nature of early farming in central and southeast Europe. Documenta
Praehistorica 31, 49–58.
Bogaard, A. (2005) ‘Garden agriculture’ and the nature of early farming in Europe and the
Near East. World Archaeology 37, 177–96.
Bogaard, A., Palmer, C., Jones, G., Charles, M. and Hodgson, J. (1999) A FIBS approach
to the use of weed ecology for the archaeobotanical recognition of crop rotation regimes.
Journal of Archaeological Science 26, 1211–24.
Bogaard, A. and Jones, G. (2007) Neolithic farming in Britain and central Europe: contrast
or continuity? Proceedings of the British Academy 144, 358–75.
Bogaard, A., Heaton, T. H. E., Poulton, P. and Merbach, I. (2007)  e impact of manuring
on nitrogen isotope ratios in cereals: archaeological implications for reconstruction of diet
and crop management practices. Journal of Archaeological Science 24, 335–43.
Bogucki, P. (1993) Animal traction and household economies in Neolithic Europe. Antiquity
67, 492–503.
Boserup, E. (1965)  e conditions of agricultural growth. New York, Aldine Publishing
Company.
Brown, P. (1978) Highland peoples of New Guinea. Cambridge, Cambridge University Press.
Chapman, J. (1982)  e secondary products revolution and the limitations of the Neolithic.
Institute of Archaeology Bulletin 19, 107–22.
Charles, M. (2005) East of Eden? A consideration of Neolithic crop spectra in the eastern
Fertile Crescent and beyond. In S. Colledge and J. Conolly (eds.)  e origins and spread of
domestic plants in southwest Asia and Europe, 37–51. Walnut Creek, Left Coast Press.
Charles, M., Bogaard, A., Jones, G., Hodgson, J. and Halstead, P. (2002) Towards the
archaeobotanical identifi cation of intensive cereal cultivation: present-day ecological
investigation in the mountains of Asturias, northwest Spain. Vegetation History and
Archaeobotany 11, 133–42.
Childe, V. G. (1957)  e dawn of European civilisation. London, Routledge and Kegan Paul.
Clark, J. G. D. (1952) Prehistoric Europe: the economic basis. London, Methuen.
Cullen, T. (1984) Social implications of ceramic style in the Neolithic Peloponnese. In W. D.
Kingery (ed.) Ancient technology to modern science, vol. 1, 77–100. Columbus, American
Ceramic Society.

109
5. Farming regimes in Neolithic Europe: gardening with cows and other models
Davies, M. S. and Hillman, G. (1988) Eff ects of soil fl ooding on growth and grain yield of
populations of tetraploid and hexaploid species of wheat. Annals of Botany 62, 597–604.
Dennell, R. W. (1978) Early farming in south Bulgaria from the VI to the III millennia BC.
Oxford, British Archaeological Reports.
Fernández, J. G. (1988) Sociedad y organización tradicional del espacio en Asturias. Gijón,
Silverio Cañada.
Flannery, K. V. (1969) Origins and ecological eff ects of early Near Eastern domestication.
In P. J. Ucko and G. W. Dimbleby (eds.)  e domestication and exploitation of plants and
animals, 73–100. London, Duckworth.
Forbes, H. (1982) Strategies and soils: technology, production and environment in the peninsula of
Methana, Greece. PhD thesis, University of Pennsylvania. Ann Arbor, University Microfi lms
International.
Forge, A. (1972) Normative factors in the settlement size of Neolithic cultivators (New Guinea).
In P. J. Ucko, R. Tringham and G. W. Dimbleby (ed.) Man, settlement and urbanism,
363–76. London, Duckworth.
Gallant, T. W. (1991) Risk and survival in ancient Greece. Cambridge, Polity Press.
Gamble, C. (1982) Animal husbandry, population and urbanisation. In C. Renfrew and M.
Wagstaff (eds.) An island polity: the archaeology of exploitation in Melos, 161–71. Cambridge,
Cambridge University Press.
Garnsey, P. (1988) Famine and food supply in the Graeco-Roman world. Cambridge, Cambridge
University Press.
Gilman, A. (1981)  e development of social stratifi cation in Bronze Age Europe. Current
Anthropology 22, 1–23.
Grigg, D. (1974)  e agricultural systems of the world: an evolutionary approach. Cambridge,
Cambridge University Press.
Halstead, P. (1981) Counting sheep in Neolithic and Bronze Age Greece. In I. Hodder, G.
Isaac and N. Hammond (eds.) Pattern of the past: studies in honour of David Clarke, 307–39.
Cambridge, Cambridge University Press.
Halstead, P. (1984) Strategies for survival: an ecological approach to social and economic change
in the early framing communities of  essaly, N. Greece. Cambridge, University of Cambridge
PhD thesis.
Halstead, P. (1987) Traditional and ancient rural economy in Mediterranean Europe: plus ça
change? Journal of Hellenic Studies 107, 77–87.
Halstead, P. (1989) Like rising damp? An ecological approach to the spread of farming in
southeast and central Europe. In A. Milles, D. Williams and N. Gardner (eds.)  e beginnings
of agriculture, 23–53. Oxford, British Archaeological Reports.
Halstead, P. (1990) Waste not, want not: traditional responses to crop failure in Greece. Rural
History 1, 147–64.
Halstead, P. (1992) Agriculture in the Bronze Age Aegean: towards a model of palatial economy.
In B. Wells (ed.) Agriculture in ancient Greece: proceedings of the seventh international symposium
at the Swedish Institute at Athens, 105–16. Stockholm, Swedish Institute at Athens.
Halstead, P. (1995) Plough and power: the economic and social signifi cance of cultivation with
the ox-drawn ard in the Mediterranean. Bulletin on Sumerian Agriculture 8, 11–22.
Halstead, P. (1998a) Ask the fellows who lop the hay: leaf-fodder in the mountains of northwest
Greece. Rural History 9, 211–34.
Halstead, P. (1998b) Mortality models and milking: problems of uniformitarianism, optimality
and equifi nality reconsidered. Anthropozoologica 27, 3–20.
Halstead, P. (1999) Surplus and share-croppers: the grain production strategies of Mycenaean
palaces. In P. Betancourt, V. Karageorghis, R. Laffi neur and W.-D. Niemeier (eds.)

110 Valasia Isaakidou
MELETHMATA. Studies presented to Malcolm H. Wiener as he enters his 65th year, 319–26.
Liège, University of Liège.
Halstead, P. (2000) Land use in postglacial Greece: cultural causes and environmental eff ects.
In P. Halstead and C. Frederick (eds.) Landscape and land use in postglacial Greece, 110–28.
Sheffi eld, Sheffi eld Academic Press.
Halstead, P. (2002) Agropastoral land use and landscape in later prehistoric Greece. In E. Badal,
J. Bernabeu and B. Martí (eds.) El paisaje en el Neolitico Mediterráneo, 105–13. València,
Universitat de València.
Halstead, P. (2006a) Sheep in the garden: the integration of crop and livestock husbandry in
early farming regimes of Greece and southern Europe. In D. Serjeantson and D. Field (eds.)
Animals in the Neolithic of Britain and Europe, 42–55. Oxford, Oxbow Books.
Halstead, P. (2006b) What’s ours is mine? Village and household in early farming society in Greece.
Amsterdam, University of Amsterdam.
Halstead, P. and Jones, G. (1989) Agrarian ecology in the Greek islands: time stress, scale and
risk. Journal of Hellenic Studies 109, 41–55.
Halstead, P. and Isaakidou, V. (2011) Revolutionary secondary products: the development and
signifi cance of milking, animal traction and wool gathering in later prehistoric Europe and
the Near East. In T. Wilkinson, S. Sherratt and J. Bennet (eds.) Interweaving worlds: systemic
interactions in Eurasia 7th–1st millennia BC. Oxford, Oxbow Books.
Harris, M. (1966)  e cultural ecology of India’s sacred cattle. Current Anthropology 7,
51–66.
Hedges, R. E. M. and Reynard, L. M. (2007) Nitrogen isotopes and the trophic level of
humans in archaeology. Journal of Archaeological Science 34, 1240–51.
Higham, C. and Message, M. A. (1969) An assessment of a prehistoric technique of bovine
husbandry. In D. Brothwell and E. S. Higgs (eds.) Science in archaeology, 315–30. London,
 ames and Hudson.
Hillman, G. (1981) Reconstructing crop husbandry practices from charred remains of crops.
In Mercer, R. (ed.) Farming practice in British prehistory, 123–62. Edinburgh, Edinburgh
University Press.
Hillman, G. (1984a) Interpretation of archaeological plant remains: ethnographic models
from Turkey. In W. van Zeist and W. A. Casparie (eds.) Plants and ancient man, 1–41.
Rotterdam, Balkema.
Hillman, G. (1984b) Traditional husbandry and processing of archaic cereals in recent times:
part 1, the glume wheats. Bulletin of Sumerian Agriculture 1, 114–52.
Hillman, G. (1985) Traditional husbandry and processing of archaic cereals in recent times:
part 2, the free-threshing cereals. Bulletin of Sumerian Agriculture 2, 1–31.
Hodkinson, S. (1988) Animal husbandry in the Greek polis. In C. R. Whittaker (ed.) Pastoral
economies in classical antiquity, 35–74. Cambridge, Cambridge Philological Society.
Isaakidou, V. (2004) Bones from the labyrinth: faunal evidence for the management and consumption
of animals at Neolithic and Bronze Age Knossos, Crete. London, Institute of Archaeology
PhD thesis.
Isaakidou, V. (2006) Ploughing with cows: Knossos and the ‘secondary products revolution’. In
D. Serjeantson and D. Field (eds.) Animals in the Neolithic of Britain and Europe, 95–112.
Oxford, Oxbow Books.
Isaakidou, V. (2008) ‘ e fauna and economy of Neolithic Knossos’ revisited. In V. Isaakidou
and P. Tomkins (eds.) Escaping the labyrinth: the Cretan Neolithic in context, 90–114. Oxford,
Oxbow Books.
Jarman, M. R., Bailey, G. N. and Jarman, H. N. (1982) Early European agriculture. Cambridge,
Cambridge University Press.

111
5. Farming regimes in Neolithic Europe: gardening with cows and other models
Johannsen, N. N. (2006) Draught cattle and the south Scandinavian economies of the 4th
millennium BC. Environmental Archaeology 11, 35–48.
Jones, G. (1984) Interpretation of archaeological plant remains: ethnographic models from
Greece. In W. van Zeist and W. A. Casparie (eds.) Plants and ancient man, 43–61. Rotterdam,
Balkema.
Jones, G., Bogaard, A., Halstead, P., Charles M. and Smith H. (1999) Identifying the intensity
of crop husbandry practices on the basis of weed fl oras. Annual of the British School at
Athens 94, 167–89.
Koster, H. A. (1977)  e ecology of pastoralism in relation to changing patterns of land use in
the northeast Peloponnese. PhD thesis, University of Pennsylvania.
Kruk, J. (1980)  e Neolithic settlement of southern Poland. Oxford, British Archaeological
Reports.
Lawrence, P. R. and Pearson, R. A. (2002) Use of draught animal power on small mixed farms
in Asia. Agricultural Systems 71, 99–110.
Leach, H. M. (1999) Intensifi cation in the Pacifi c. Current Anthropology 40, 311–39.
Legge, A. J. (1981) Aspects of cattle husbandry. In R. J. Mercer (ed.) Farming practice in British
prehistory, 169–81. Edinburgh, Edinburgh University Press.
Mainland, I. (2003) Dental microwear in modern Greek ovicaprids: identifying microwear
signatures associated with a diet of leafy hay. In E. Kotjabopoulou, Y. Hamilakis, P. Halstead,
C. Gamble and P. Elefanti (eds.) Zooarchaeology in Greece: recent advances, 45–50. London,
British School at Athens.
Mainland, I. and Halstead, P. (2005)  e diet and management of domestic sheep and goats
at Neolithic Makriyalos. In J. Davies, M. Fabis, I. Mainland, M. Richards and R.  omas
(eds.) Diet and health in past animal populations: current research and future directions,
104–12. Oxford, Oxbow Books.
Mishra, A. K. and Pandey, A. S. (2000) Seasonality of bullock power use in rainfed areas.
Draught Animal News 32, 11–13.
Morrison, K. D. (1994)  e intensifi cation of production: archaeological approaches. Journal
of Archaeological Method and  eory 1, 111–59.
Payne, S. (1973) Kill-off patterns in sheep and goats: the mandibles from Aşvan Kale. Anatolian
Studies 23, 281–303.
Rappaport, R. A. (1968) Pigs for the ancestors. London, Yale University Press.
Renfrew, C. (1972)  e emergence of civilisation: the Cyclades and the Aegean in the third
millennium BC. London, Methuen.
Rösch, M., Ehrmann, O., Hermann, L., Schulz, E., Bogenrieder, A., Goldammer, J. P., Hall, M.,
Page, H. and Schier, W. (2002) An experimental approach to Neolithic shifting cultivation.
Vegetation History and Archaeobotany 11, 143–54.
Rowley-Conwy, P. (1981) Slash and burn in the temperate European Neolithic. In R. Mercer (ed.)
Farming practice in British prehistory, 85–96. Edinburgh, Edinburgh University Press.
Rowley-Conwy, P. (1997)  e animal bones from Arene Candide: fi nal report. In R. Maggi
(ed.) Arene Candide: a functional and environmental assessment of the Holocene sequence,
153–277. Rome, Il Calamo.
Semple, E. C. (1932)  e geography of the Mediterranean region: its relation to ancient history.
London, Constable.
Shennan, S. (1986) Central Europe in the third millennium BC: an evolutionary trajectory
for the beginning of the European Bronze Age. Journal of Anthropological Archaeology 5,
115–46.
Sherratt, A. (1972) Socio-economic and demographic models for later prehistoric Europe. In
D. L. Clarke (ed.) Models in archaeology, 477–542. London, Methuen.

112 Valasia Isaakidou
Sherratt, A. (1980) Water, soil and seasonality in early cereal cultivation. World Archaeology
11, 313–30.
Sherratt, A. (1981) Plough and pastoralism: aspects of the secondary products revolution. In
I. Hodder, G. Isaac and N. Hammond (eds.) Pattern of the past: studies in honour of David
Clarke, 261–305. Cambridge, Cambridge University Press.
Sherratt, A. (1997) Economy and society in prehistoric Europe: changing perspectives. Edinburgh,
Edinburgh University Press.
Sherratt, A. (2005) Diverse origins: regional contributions to the genesis of farming. In S.
Colledge and J. Conolly (eds.)  e origins and spread of domestic plants in southwest Asia
and Europe, 1–20. Walnut Creek, Left Coast Press and University College London Institute
of Archaeology Publications.
Sherratt, A. (2006) La traction animale et la transformation de l’ Europe néolithique. In P.
Pétrequin, R.-M. Arbogast, A.-M. Pétrequin, S. van Willigen and M. Bailly (eds.) Premiers
chariots, premiers araires. La diff usion de la traction animale en Europe pendant les IVe et IIIe
millénaires avant notre ère, 329–60. Paris, CNRS Editions.
Sunderlin, W. D. (2003) Ideology, social theory, and the environment. Lanham, Rowman and
Littlefi eld.
 omas, C. K. (2000)  e role of draught cattle and buff aloes in sustainable agriculture in
India. Draught Animal News 32, 4–11.
 omas, J. (1999) Understanding the Neolithic. London, Routledge.
 omas, J. (2003)  oughts on the ‘repacked’ Neolithic Revolution. Antiquity 77, 67–74.
 omas, J. (2008)  e Mesolithic-Neolithic transition in Britain. In J. Pollard (ed.) Prehistoric
Britain, 58–89. Oxford, Blackwell.
Vigne, J.-D. and Helmer, D. (2007) Was milk a ‘secondary product’ in the Old World
neolithisation process? Its role in the domestication of cattle, sheep and goats.
Anthropozoologica 42, 9–40.
Whittle, A. (1996) Europe in the Neolithic: the creation of new worlds. Cambridge, Cambridge
University Press.
Wilkie, N. C. and Savina, M. E. (1997)  e earliest farmers in Macedonia. Antiquity 71,
201–7.
Willis, K. J. and Bennett, K. D. (1994)  e Neolithic transition fact or fi ction? Palaeoecological
evidence from the Balkans.  e Holocene 4, 326–30.