Durham Research Online

Deposited in DRO:
06 April 2009
Version of attached le:
Published Version
Peer-review status of attached le:
Peer-reviewed
Citation for published item:
Rowley-Conwy, P. (2001) 'Time, change and the archaeology of hunter-gatherers : how original is the 'Original
auent society'?', in Hunter-gatherers : an interdisciplinary perspective. Cambridge: Cambridge University
Press, pp. 39-72. Biosocial Society Symposium Series. (13).
Further information on publisher's website:
http://www.cambridge.org/catalogue/catalogue.asp?isbn=9780521772105
Publisher's copyright statement:

c 2001 Cambridge University Press

Additional information:

Use policy
The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for
personal research or study, educational, or not-for-prot purposes provided that:
• a full bibliographic reference is made to the original source
• a link is made to the metadata record in DRO
• the full-text is not changed in any way
The full-text must not be sold in any format or medium without the formal permission of the copyright holders.
Please consult the full DRO policy for further details.

Durham University Library, Stockton Road, Durham DH1 3LY, United Kingdom
Tel : +44 (0)191 334 3042 | Fax : +44 (0)191 334 2971
http://dro.dur.ac.uk
 3
Time, change and the archaeology qf
hunter-gatherers: how original is the
(Original Ajfiuent Socie?)!'?

PETER RO\\-LEY-COt\\\-y

The Original Affiuent Society

This highly e'-ocative phrase first appeared in ,\lall the Hunter (Lee
and DeVore 1968a), used by Marshall Sahlins to describe hunter-
gatherers:
This was, when you come 10 think of it, the original amucllt society. By
common understanding an amucnt society is one in which all the
people's W3IHS arc easiJy satisfied . .. lbut] wams are 'caliily satisfied'
either hy producing much or desiring little, and there arc, accordingly,
two possible roads to amuencc. The Galhraithian course makes
assumptions peculiarly appropriate 10 markrl economics ... But there
is also a Zen solution to scarcity ;md affiucllcc. beginning from
premises opposite from our own, that human material ends arc few and
finitl> and technical means unchanging but on the whole adequate.
Adopting Ihe Zen strategy, a people can enjoy an unparalleled mate';al
p1eI1l); though perhaps on1\' a low standard ofli,-ing_ (Sahlins 19G8: 85)
These concepts were de,-doped in Stone Age Economics (Sahlins 1972),
the first chapter of which was entitled 'The Original Amuenl
Society'_ Sahlins rooted the Zen concept of 'want not, lack not'
(p_ I I) in the mobility of hunter-gat herers_ 1\ lost groups carry with
them all their material possessions, which must thus be kept to a
minimum_ In a word, 'mobility and property are in contradiction'
(p_ 12).
Salllins's formulation sprang directly from the definition of
hunter--gatherers provided by Lee and DeVore in Man the Hunter.
'we make two basic assumptions about hunters and gatherers: (I)
they live in small groups and (2) they mo\-e around a laC (Lee and

39
40 PETER ROWLEY-CONW,"

DeVore 1968b: II). Five outcomes of this were responsible for

shaping hunter-gatherer society: (a) little personal property and an
egalitarian social system; (b) sporadic gatherings of bands, and
much mobility of individuals between bands; (c) a fluid organisation
invoking no territorial I"ights; (d) no food storage; and (e) no group
strongly attached to a particular area (Lee and DeVore 1968b: 12).
This was the Original Amuent Society (OAS) in a nutshell, and
provided a clear vision of what hunter-gatherers were like.

Hunter-gatherer variability

The OAS was so powerful a concept that hunter-gatherer varia-

bility received little consideration until 12 years after ii/an t/ie HUllter.
Binford (1980) and Woodburn (1980) then each created a typology
of hunter-gatherers in which one type conformed to, while the
other type diverged from, the OAS model.
Binford (1980) distinguished between 'foragers' and 'collectors'
(Figure 3.1). Foragers correspond to the OAS: they move relati"e1y
often, and indi"iduals return to the residential base eaeh dar
Collectors however move less often, and are found in el1\'ironments
in which resource availability "aries in both time and space.
Resource storage helps counteract temporal variation: food collected
in a season of plenty may be stored for later use. Resource transport
helps counteract spatial variation: special-purpose field camps are
used to procure and process resources which are then transported
back to the base camp. Binford termed this a 'logistic strategy'.
Woodburn (1980) coined the tcrm 'immediate return' for groups
corresponding to the OAS because food collected on a particular
day is consumed almost immediately: the return upon labour is
immediate. In 'delayed return' societies on the other hand consump-
tion is delayed. Resources may be stored for later consumption,
and/ or work may be expended on complex items like fish traps
before the relevant resource becomes available. Woodburn explored
two further corollaries. First, a fish trap involves considerable
labour, and the trap and its catch will therefore belong to those who
constructed it. Individuals or groups thus own territories. Second,
 FORAGING COLLECTING

6 \
,•
\s\\C radiI.
,o~/' - - '.r
" '\ / ,,,
/
/
,, /

• ,
, ,,
,, /
\ "-
I
......
•
6
- ,/

It •6
/
" -'-
- /'

resource patch
8 residential base and
daily foraging radius

~
move between •
6 field cemp

4 route of logistic
~ residential bases task group

Figure 3.1. Schematic maps contrastin)( roragers (Iel\) wilh collectors (ri)(hl), based on Ih<: discussion or Binrord (1980). Each shows
the same proportion of 11YlJothcticai annual rounds, so that foragers' residential bases arc occupi.;d for shorter )>t'riods than those
or collectors.
42 PETER ROWLEY-CONW"

the ownership of both territories and stored food represents unequal

access to wealth. Food is not shared throughout the group, so society
tends to be hierarchical. Territoriality and social hierarchy are both
in Aat contradiction to the OAS.
'T'hese two typologies produce similar but not identical classifica-
tions. Inuit store food and are logistically organised, although most
are not ten;torial or hierarchical. Australian Aborigines do not store
food, but are to an extent tenitorial. Layton (1986) resolved these
anomalies: Inuit are obliged to store food due to seasonal variation
in resources, but unpredictable inter-annual and spatial variations
make territoriality unviable. Aborigines are territorial mainly with
regard to water: ritual knowledge about water sources is jealously
guarded while in other respects they practise an immediate return
strategy. This allows us to construct a four-fold typology of hunter-
gatherers (Rowley-Conwy 1999):
I. The OAS: groups \\'ilh little or no logistic movemenl or resources or
rood storage. These are mostly round in tropical regions (e.g. the
Aborigines), although some occur in higher latitude areas where
resources are a,"ailable throughout the year; people can move from OIlC
resource to the next, exploiting them in sequence ,,~thout the need for
much storagc"
2. Logistic groups that do not derend territories, such as most Inuit.
3. Logistic groups thaI do derend territories - many or Woodburn's
dclared return groups.
4. Sedentary groups who ill\"ariably ddend terrilories and store
resources, forming a continuation from type 3.
This typology applies to ethnographically known groups. Archae-
ology was similarly challenging the assumption that all hunter-
gatherers conformed to the OAS, but identifying divergent groups
was more difficult because of the problems of interpreting ule
archaeological record. Sedentism, which could be identified, did not
conform to the OAS (Price 1981, Rowley-Conwy 1983). Archaeolo-
gists use the term 'complex' to describe non-OAS groups, perhaps
unfortunate because it means that 'simple' is sometimes used to
describe OAS groups, but it will be followed here. Applying this
terminology to the four-fold typology, type I or OAS groups are
non-complex, the rest forming a continuum towards the most
comple.,x type 4 groups.
 sedentlsm sedentism
11 , 5 11 •
1
4 5 1 2
10 1 5 5 10 1 1
545 12 1
91 5 55 5 9 1 12 2
5 1
B 4 4 4 55 4 5 B 1 1 0 12 1 1
£ 7 53
4 4 5
4 7 21
0 1 1
1
~ B
33 3 4
333
4545
5 6
11
no
1 1 1202
2
'0 3 4 4 0 1 1
1i 5
2 2
2 3 333 5
0 0
0 o 110

§ 4 2 2 2 22 22 4 o0 0 00 10
Z 22 01
3 2 2022
storage 3 0 00 0 social stratification
2 0
dependence 0= none
21 2 20 2
0= none
2 0 00 0
1 = wealth classes
15= highest 2 = descent classes
1 i I I
2 2
I I I I I
1 0 0
·3 -2 ·1 0 1 2 3 ·3 ·2 ·1 0 1 2 3
population relative to ecological productivity population relative to ecological productivity

Figure 3.2. Correlations b"tween various attributes of hunter-gatherer societies. Each data poim is one ethnographic hunter-
gatherer group. S"demism is the number of months for which the winter settlement is occupied. Population relative to ecological
productivity is calculated by Keeley (198B, 1'.385). Societies with wealth cla«es arc those where individuals can acquire high
status during their lives; societies with descent classes arc those in which status is inherited. (Redrawn from Keeley 1991, Figs. 17.\
and 17.6.)
+-1- PETER ROWLEY-CONWY

Anthropological work on variability has been taken further in

se,·eral ways. For example, territoriality has been linked to environ-
mental factors: if an important resource appears in predictable
concentrations, tlH'n that area is likely to be the exclusive territory of
an individual or group; while if resources are unpredictable, Rex-
ibility is a beltn response (Dyson-Hudson and Smith 1978). Archae-
ology gains an unexpected bonus here because telTitorial hunter-
gatherers sometimes bury their dead in cemeteries whereas non-
territorial ones do not. An archaeological cemetery thus probably
indicates a terrilOrial group (Saxe 1970, Pardoe 1988). Technology
becomes more specialised when maximum resources have 10 be
procured in a short period (Torrence, this volume). The major cross-
cultural sUlyeys of Keeley (1988, 1991) demonstrate that population
density and sedentism are linked (Figure 3.2); so is dependence on
storage, and social factors: territorial descent classes characterise lhe
sedentalY groups with most storage. A hierarchical society is thus
linked to economic factors. This is a further bonus for archaeology:
if we can demonstrate a degree of sedentism or locate a cemetery,
other aspects of complexity, harder to see archaeologically, may be
confidently predicted.

Progressivist views of complexity

Anthropology in recent years has usually presented variability as

synchronic, occurring across space, in a largely adaptive context.
Archaeology tends 10 approach it "ely dilTerently, often presenting it
as diachronic, de,·eloping through time, in a largely progressive
context. l\/ore often than not this is done implicitly. While Brown
and Price (1985: 436) eschew the bandltribe/chiefdom/state pro-
gressivist typology of Sen·ice (1962), their volume is entitled Pre-
histone HUI/ter-Gatherers: The Ellle/gellee qf Cultural Comjllexi!)' (Price and
Brown 1985). The search for an 'emergence' of complexity, reRected
in so many article and chapter tilles, signals an implicit belief tha:
there was a lime bifore complexity emerged; a time, therefore, of
universal simplicity.
These two views cannot both be right. Either variability IS a
 TIme, chnllge nlld Ihe nrdwfo!og)' if !lImtrr-gnthrrrrs 45

response to loeal conditions, or it is time-dependant and tending

towards complexity. This chapter argues in favour of local responses
and local historical trajectories, and against any progressive trend,
by challenging the one aspect of the OAS that is usually accepted
without question: its originality.
Progressi"ist views of the emergence of complexity are usually
couched in one of two ways. . Both im'olve intensification, namely. an
increase in the productivity (or production) of resources. However,
demographic viewpoints see population increase as the prime
mover, while social "iewpoints see increasing social obligations as
causing increased production.
The demographic "iew holds that hunter-gatherer populations
increase as a matter of course, so groups must intensify, i.e. produce
more food from a given area. A particularly clear statement comes
from Zvelebil (1995), reproduced as Figure 3.3. Each time popula-
tion catches up with resources it 'bounces' producti"ity upwards,
firstly via greater mobility and diversification (OAS strategies), tllen
specialisation (a type 2 logistic strategy), storage (types 3 or 4), and
ultimately husbandry and domestication. The problem lies in the
time-scale: complexity is the endpoint of a long trajectory. Figure
3.3 cQ\'ers only the late Mesolithic of Zvelebil's region (the Baltic).
This impljes that during the pre"ious 5000 years since postglacial
colonisation, population was below em'ironmental canying capa-
city. Population must therefore have been growing "ery slowly. In
area's like Portugal, continuously occupied since the arrival of
modern humans, tile time lag would be 30000 years.
Would population density remain below capacity for so long?
This is widely debated, the Australian terminology being tile most
explicit: 'fast-trackers' suggest fast population increase up to car-
rying capacit); 'slow-trackers' that it remains low for a long period
(Mulvaney and Kamminga 1999: 132-3). Human populations
might well grow very fast. Discussing the hunter-gatherer occupa-
tion of North America, Diamond (1987) states that an annual
growth rate of 1.4% will turn 100 people into I 000000 people ill
olle thousnlld'yenrs. Birdsell (1957) calculates that a founder population
of 25 people could saturate Australasia with 300000 hunter-
gatherers in 2204 years. Pennington (tllis "olume) documents much
z REGIONAL
o RESOURCE
~ PRODUCTIVITY//'
::>
a. IPOPULATION
DENSITY
~ I DOMESTICATION ~-/

HUSBANDRY /'

"..
- - - -+-
SPECIALISATION STORAGE
--

/
SPECIALISATION
population control
/'
/' DIVERSIFICATION
amlgratlon
population control

amigralJon
/' populalJon control
/' greater mobill1y

TIME
Figure 3.3. Population density, resource productivity, and resource-use strategies. (Redrawn rrom Zvelebil 1995, Fig. 2.)
 Time, chmlge and Ihe arehaeolog)' '!f hunler-galherers 47

variety in growth rate even ,,·ithin the small remaining sample of

hunter-gatherers. Groups coIonising new areas like the Baltic would
probably have had rates of increase at or above the highest levels
observed recently.
Faced with such figures, slow-trackers must explain wi,,' huntcr-
gatherer populations should increase only very slowly. Voluntary
population control is usually suggested (Hayden 1972): people
perceive that increased population means increased work, and
therefore choose to keep their population well below carrying
capacity by means of infanticide, etc. There are two difficulties with
this. First, it is difficult to establish whether infanticide really has
been very important among recent hunter-gatherers (Kelly 1995:
2320:). Second, the suggestion runs counter to biological theory by
treating the group rather than the individual as the unit of selection,
by invoking 'cultural controls' as the means of decision-making.
Individuals who ignored the 'controls' would however gain an
evolutionary benefit by filling tl,e underpopulated landscape with
their descendants. Non-hierarchical OAS groups exercise little
control over individuals and would not be able to police a slow-track
demograph ic policy.
Social obligations arc tl,e other pressure sometimes im'oked for
increased production. Bender (1978) states that developing alliance
networks are crucial; there is 'a direct link between evoking social
instituti.ons and increasing pressure on production' (1978: 213). This
can transform an OAS group into a type 4 society: 'Surplus
production involves delayed return: in response to the requirements
of the alliance and leadership seasonally abundant foodstuffs and
other material items will accumulate in quantities over and above
immediate requirements' (ibid.). Under these circumstances 'clearly
there is a pay-off in Slaying put and creating permanent storage
facilities' (ibid.).
Directional and incremental change in social institurions is
assumed, but there is no reason why alliances should become more
complex and demanding through time. OAS groups in fact main-
tain some of the most complex alliance networks kno,,"n to anthro-
pology, and these are satisfactorily sen'iced without deparrure from
OAS behaviour. The !Kung of Botswana maintain numerous part-
48 PETER ROWLEY-CONWY

ners Up to 200 km away in the exehange system known as haro

(\"'iessner 1982). In southeastern Australia people travelled hun-
dreds of kilometres to attend gatherings at points of temporary
resource abundance (l\IcBryde 1984). Such activities fit well with
the flexible organisation of these OAS groups and do not appear to
cause directional change away from this. Social intensification per se
does not therefore account for when and why some societies should
become complex and others should not.

Hunter-gatherer cOIDplexity: the archaeological

record

It has been argued above that the theoretical underpinnings of the

progressi"ist view are not solid. The archaeological record of
anatomically modern humans is now examined. There are six
assumptions, often implicit, invoked in the progressivist view, which
are discussed in turn.

Assumptioll I: There is a trendji'DIll sim/lle to complex

It is often argued that aquatic resources are a late addition to tlle

human diet. Binford (1991) states that 'Pleistocene people ... , it is
well known, lavoured terrestrial resources' (p. 134); at the Pleisto-
cene/Holocene boundary there was an 'aquatic resource revolu-
tion'; the appearance of logistic (type 2 or 3) strategies was
dependent on this revolution, and on the development of transpon
technologies such as water transpon and pack animals (p. 138).
Pleistocene marine strategies are hard to examine because the
sea has risen > I00 m since the last glacial maximum and has
flooded most coastlines, but work in various areas has revealed
aquatic interests. ew Ireland (northeast of New Guinea) was
occupied by 33000 OP, requiring a sea crossing of c. 50 km (Allen
el al. 1988). The Solomon Islands were occupied before 28000 OP,
requiring a voyage of over 130 km, involving sailing out of sight of
land (Wickler and Spriggs 1988). Several sites have produced
 Time, change and the archaeology <if hunter-gatherers 49

marine fish and molluscs; these are in steep areas where the sea
was never far away. The evident use of watercraft removes one of
Binford's objections, and at 19000 BP live marsupials were trans-
ported to New lreland and released to lGund populations which
could be hunted (Flannery and White 1991) - a delayed-return
activity if ever ulere was one! On the Australian mainland marine
evidence is available from at least 25000 BP (Morse 1988), and it
has even been suggested that the entire first occupation of
Australia was coastally oriented (Bowdler 1977). In southern
Africa marine resources were exploited still earlier. The shell
midden at Heralds Bay is 125000 years old (Brink and Deacon
1982). Many others are known, and marine foods may have
played an important role in ule evolution of modern humans
(Parkington in press).
Binford's statement was heavily based on Europe, where cvi-
dence of marine exploitation is indeed concentrated in the post-
glacial. However, some hints from the glacial period are found in
areas of steep topography. Thousands of Upper Palaeolithic
limpets come from the northern Spanish sites of EI Juyo (?\'ladar-
iaga de la Campa and Fermindez PaLO 1985) and La Riera (Ortea
1986). Seal bones have been recovered from Nerja (?\'Jorales et at.
1998), Gorham's Cave and Altamira (Cleyet-Merle 1990), all in
Spain. Bones of tunny were recovered from the Grimaldi caves in
northwestern Italy, which suggests offshore boating, and occasional
marine fish turn up elsewhere (ibid.). Artistic representations of
marine fauna include the famous baton from Montgaudier, an
inland site in southern France, which shows two seals (Figure 3.4).
Seals may occasionally swim up rivers and be seen inland, but the
Montgaudier specimens are well depicted and clearly engraved by
someone who knew seals well. Under the nose of the leading seal
is what is sometimes described as a salmon, but this is more likely
to be a whale as it appears to have a spout above its head. Fish
depictions are generally difficult to identify, but flatfish are exclu-
sively marine and are shown at Mas d'Azil and Altxerri in ule
Pyrenees.
These cases do not of course demonstrate logistic strategies, but it
is hard to see how hunter-gatherers could have survived in ice-age
50 PETER ROWLEY-CONWY

..

..,
! \

', \ ;',1\
.-...;.o-f.;)

Figure 3.4. 11,e r-tontgaudier baton (top), and the flatfish from Altxem Oeft)
and r- las d'Azil (right). (Redrawn from Cle)'et-Merle 1990.)

Europe without them. Reindeer and salmon were the major terres-
trial and aquatic resources, and ethnographically they are classic
targets of logistic strategies. What could have prevented Upper
Palaeolithic people li'om developing maritime adaptations? Specia-
lised multi-component technology for hunting land mammals
appeared by 30 000 BP (Knedlt 1993), and recent dates on art
re\'eal a complex ritual system just as early (Clottes 1999), so
technological and ideological flexibility were evidently not lacking.
There arc some hints of coastal specialisation. Clustering of art
sites along the north Spanish coast may indicate territorial group-
ings (Layton 1987), and marine shell ornaments predominate at
Riparo ~10chi in northwestern Italy but hardly penetrate inland
(Stiner 1999). These examples suggest communities oriented
towards the sea - but of whose territories we can only see a small
part. Coastlines dating from the Pleistocene/Holocene boundary
are above water in western Sweden and Norwav• due to isostatic
rebound, and they have e\·idence of settlement. Neither the sites in
Sweden (Schmill 1995) nor those in Norway, some of which are
 Time, change alld the archaeolog)' ~ hUllter-gatherers 51

large and lie north of the Arctic Circle (Thommessen 1996), have
preserved fauna, but they must be coastally oriented. l\!lost of
Norway was still under ice, with only islands and headlands
available for setUemenl. This Arctic maritime adaptation must
have been logistically organised.
This evidence is sufficient to indicate the likelihood of maritime
adaptations during the Upper Palaeolithic, and the nature of U1C
resources makes type 3 or even type 4 groups most probable. The
only time we can examine a relevant coast (Norway and Sweden),
settlement is demonstrated - and most glacial Atlantic and led iter-
ranean coasts would have been less demanding than this because of
the presence of hinterland resources not available in Non\'ay due to
glacial coverage.
The postglacial howe"er remains the accepted time that com-
plexity emerged, in particular the later Mesolithic. Cemeteries are
often late Mesoliulic, and it has been argued Ulat 'some sort of
demographic threshold was crossed in some parts of western
Europe at around 6500 years ago' (Clark and Neeley 1987: 124).
Most cemeteries are however near the coasts, which is where we
should expcct Ulem: the territorial groups that use cemeteries would
appear in productive and reliable resource areas, whicll mostly
means the coasts. Because earlier coasts are now under the sea, any
earlier Mesolithic cemeteries would remain inaccessible. Europe
does have one terminal Pleistocene cemetery, at Arene Candide in
northwestern haly, containing 20 individuals (Cardini 1980); this is
an area where topography means that the shoreline has not moved
far, and as at nearby Riparo Mochi there are many shell ornaments.
This is a strong indication that complex groups may have existed
elsewhere as well.
The hypothesis Ulat cemeteries result from demographic increase
must be examined against evidence in other areas. In Ule Murray
Valley in southeastern Australia cemeteries are known as carly as
13000 BP (Pardoe 1988) - some 40 000 years after the first colonjsa-
tion of Australia. In orlh Africa a series of late glacial cemeteries
(Wendorf 1968, Lubell et al. 1984) postdates modern human occu-
pation by maybe 100000 years. In the Mississippi Valley however
the logistically organised Dalton culture used cemeteries as early as
52 PETER RQWLEY·CONWY

423 squlfTel
caudal wrtebnte

O chough

-"'" N
/
-,
. ~~

,
child.' VII ,:
pertoreted ,
sheila
,
, ..."
.- .

Fi~lIre3.5. Plan of the terminal Pleistocene cemetery from Arene Candide,

Liguria, Italy. Positions of some grave goods marked. (Redrawn from Cardini
1980, Fig. 2.) (Published by kind permission of the )stituto Italiano di
Paleontolo~ia Umana.)

10750 BP - just 750 years after the first occupation of the region
(Morse et al. 1996). Such hugely diverse time-lags argue strongly
against the demographic steamroller.
This evidence from around the world demonstrates much Pleisto-
cene interest in marine resources (contra Binford, above). Much
e\~dence is dist0l1ed by coastal loss, but enough remains from
Norway to New Ireland to make the case for type 3 groups in many
areas (see also Kuhn and Stiner, this volume). There is no archae-
 Time, challge alld the archaeolog)' if IlUlller-gatherers 53

ological evidence that hunter-gatherers display an inherent trend

from simple to complex.

Assumption 2: Only simple groups were present among

early anatomically modern humans
This assumption is virtually ubiquilOus, but caution is suggested by
the arguments based on lvlesolithic cemeteries. Skhul and Qafzeh
caves in Israel, both around 90000 years old, contain 10 and 18
skeletons of modern humans respectively. If they are cemeteries
their implications for territoriality should not be ignored because of
their antiquity.
Gargelt (1999) has argued that the Qafzeh assemblage is a
natural accumulation of skeletons rather than deliberate burials,
because the skeletons are fi'agmentary, incomplete and disarticu-
lated, and there are no c1earcut grave pits. However, the number of
individuals is more than could accumulate by chance, and Gargelt
does not consider whether these anatomically modern people might
also have been behaviourally modern. Disarticulation and breakage
are common in cave cemeteries, and grave pits are often unneces-
sary. At Arene Candide, Cardini states that:
Some oCthe inhumations were insencd directlv into modest natural
depressions formed between the .umbled bloeks from the Q\'edring
deposits, others appeared to be protected both by these same blocks
already present in the deposits and by the dclibcralc placing of olher
lumps or slabs of limestone, which in a few cases encircled the
inhumation and in others partially con:red it. ~Ian)' gran~s were
disturbed or complelcly destroyed by the placing of subsequent bllrials,
with the consequent dispersal of the bones. Only in a few cases ... do
we ObSCIYC a deliberate placing. something like a secondary deposition
of groups of bones coming from pre,-ious disturbed inhuma.ions. (1980:
13, my .ransla.ion)
This disturbance appears similar to that at ~'lfzeh, but Arene
Candide is an indubitable cemetery replete with grave goods and
ochre (Figure 3.5); utilised ochre was also found associated with one
of the Qafzeh skeletons (Vandermeerscll 1969).
Much more information is needed regarding the beha"iour of
early modern humans. In the meantime we cannot dismiss Ule
54 PETER ROWLEY-CONWY

possibility of complexity merely because it is 'too early'; this should

be investigated, not assumed in ad\'ance,

Assumption 3. Change towards complexity occurs slowly

Archaeologists often assume that a trend towards complexity occurs
slowly over the long term, perhaps millennia. Archaeological evi-
dence however suggests that such changes can take place much
faster when the opportunity arises.
A good example comes from southern Scandinavia, Late Meso-
lithic coastal settlements have long been taken to indicate com-
plexity (Rowley-Conwy 1983), but the middle Mesolithic was little
known because sea level rise had flooded contemporary coastlines.
Underwater archaeologists have recently examined these coastlines
and put archaeological developments into ecological context. The
rising sea crossed the - 27 m tllreshold into the present watenvays
at 7000 BC (Figure 3.6), and due to flat topography much land was
then flooded very rapidly. As soon as sea levels began to stabilise,
indications of complexity appeared in the form of large scmi-
sedentary selliements, cemeteries and large fish traps. The chron-
ology of these (Figure 3.6) shows no time-lag; within the limits of
dating resolution they appear immediately.
Otller examples have already been mentioned. The Dalton
culture in the Mississippi Valley was logistically organised and used
cemeteries less than a millennium after people first arrived in the
area (Morse el al. 1996). The rapid logistically organised occupation
of the Nonvegian coast occun"ed as soon as the reu"eating glaciers
had exposed the islands and headlands (fhommessen 1996), Thus
complexity may appear rapidly in response to environmental oppor-
tunities, This does not support the suggestion that complexity
emerges slowly as the endpoint of internal development.
How rapid can responses be? These archaeological examples
suITer from the imprecision of our dating methods - uncertain by at
least tile standard deviation of a radiocarbon date. Various anthro-
pological studies have prO\'ided tighter resolution. The Mbaka of
central Africa, formerly OAS-type hunter-gatherers, settled down
to become farmers and rapidly developed a hierarchical wealt1l-
 Time, challge alld the archaeolog)' qf hUllter-gatherers 55

MESOLITHIC NEOLITHIC
EARLY MIDDLE LATE agriculture
I I I I
I
0-
(I)
rfastest'l
~
III
I rising
sea
I
Q)
_rn 5-
c
(I)
rn
~
0. 1
~ 10-
o I
(I)
.0 '-I,
rn •• '- lU • I
~ ~
Q) 15- • lU 1
-.I
E
• -I MUSHOLM BAY settlement

20- : 1 STAVRESHOVED settlement

-.~ BLAK II fish trap

1- VEDSK0LLE fish trap

25- ,
threshold . '
-
.=I
1=
TAGERUP 7 A:1 settlement
cemetery
30-
--
fishtrap

35- •
-
-I
~ SEGEBRO settlement
cemetery
I I I I I
7000 6000 5000 4000 3000 BC
Figure 3.6. Sea level rise and hunter-galherer complexity in southern
Scandinavia; black bars are calibrated raruocarbon ages. The - 27 m
threshold is the level at which the southern Scandina,·ian waterways began to
be flooded. (Sea level cun'e from Christensen et al. 1997, Fig. 2; I\'lusholm Bay
from Fischer and Maim 1997; StavreshO\'ed frOIll Fischer 1997 a; B1ak 11 from
Sarensen 1996; Vedskalle from Fischer 1997 b; lagerup frolll Andersson and
KnalTSlrom 1997; Segebro from Larsson 1982.)
56 I'ETER ROWLEY-CONWY

based society - 'not due to imitative processes, but to the require-

ments of their new activities' (I\'lcillassoux 1973: 197). Aboriginal
groups near Lake Eyre oscillated between type 3 territoriality and
storage in producti"e years, and OAS-type sharing and nexibility in
lean years (Layton eI al. 1991: 258-9). This suggests that archaeolo-
gists should not view 'social factors' as rigid and inhibitors of
change, particularly among hunter-gatherers.

Assumption 4. Change towards complexity is irreversible

The progressivist view that complex groups appear towards the end
of hunter-gatherer history carries with it the assumption (often
implicit) that agriculture is the next 'stage' of directional develop-
ment. From this perspective, change towards simplicity is un-
expected - but it is far from uncommon.
A good example comes from Labrador, Canada. The Maritime
Archaic tradition began about 7500 8P. Numerous settlements are
known, such as Aillik (Figure 3.7). The AiJlik sequence starts at 6000
8P with struclUres 6 and 7, and ends at 4300-3600 8P with structure
2, a 28-metre longhouse. This is by no means the largest: Nulliak
Cove has 27 longhouses, some measuring 100 metres in length -
though all were not occupied at once. The AiUik specimen has seven
or more compartments each of a size appropriate for a nuclear
family, and about 10 stone-lined storage pits around it; this may
indicate storage by the individual families (Fitzhugh 1984). These
longhouses were probably foundations for tents, suggesting seasonal
occupation in summer. The large aggregation size means that 'some
kind of social mechanisms for organizing large groups (for example,
leadership hierarchies) may have been in operation' (Hood 1993:
170). This level of social integration disappeared at the end of the
l\'laritime Archaic tradition, which overlaps chronologically with the
immigrant Paleoeskimo responsible for structure I at Aillik (Figure
3.7).
The demise of other apparently complex adaptations is also
documented. Late Palaeolithic occupants of the lower Nile Valley
may have smoked and stored large numbers of catfish; the environ-
ment was stable and productive and sites quite large (Vermeersch e/
 Time, change and the archaeology if hUlller-gatherers 57

() r.
V o c o
~
2 melres

• poslhole

o 25 melres
------',
'--'

BAR HARBOR

~\(
• J'
. ----
-'II};? ""'I '~:" 1
! ff"

11.5m_
't
~
..~
....,,/'/1'
-.......
,:..:...... ~-:
.:;
0

+ [II 1111) o ~ : ledges

2 000 oc?o
---
~
$

steep
:I-f /
cliffs /.

/,
~=Jm-
......,.."'-<-+ 06 '-
23..m~~~~ 07 '-.. ....
.:r
--------------------
CREST Of BEACH SERIES

Figure 3.7. AiUik, Labrador, eastern Canada. Botlom: plan of thc cm·e,
showing the raiscd beaches and structures (m asl = metres above sea level).
Top: enlarged plan of structure 2. (Redrawn from Fitzhugh 1984, Figs. I and
2.)
58 PETER ROWLEY-CONWY

al. 1989: I II). The 'Wild Nile', a period of high and violent floods
starting at II 500 BP, brought this to a close, and the valley was
apparently abandoned until 8000 BP. Thus 'for tlle inhabitants of
the lower Nile \ alley, the Pleistocen Holocene transition would
seem to ha\'e been an unmitigated disaster' (Close 1996: 54). In the
IVlississippi Valley, the cemeteries and logistic adaptation of the
Dalton culture ended at 10000 BP, to be followed by an OAS group
(lvlorse et al. 1996).
Archaeology therefore provides sufficient examples to demon-
strate that change can go either way: simple to complex, or complex
to simple. Directional change towards incremental complexity is not
supported by the empirical record.

Assumption 5. Change towards complexity is a step

towards agriculture
This assumption is often explicit. An argument that complex
hunter-gatherers in southern Sweden at 4400 BC were not de\'e1-
oping towards agriculture (Rowley-Com\')' 1998) has been criticised
by the editors of tlle volume in which it appeared! 'Rowley-ComV)'
is wrong . . . Even though the movement of agriculture across
central Europe at 4400 BC was a coincidence unrelated at tllat point
to the lives of south Scandinadan hunter-gatherers, tl/{JI were SOOIl to
feel its impact . .. ' (Zvelebil et at. 1998: 2, added emphasis). This
endows the hunter-gatherers with 20-20 foresight, as well as a
curious desire to 'get ready' for the arrival of farming on ilie
opposite side of the Baltic Sea - over 100 Ion away. Farming did not
actually appear in southern Sweden until 1500 years later. Another
clear statement comes from Hodder (1990), who argues that a
major step towards agriculture was the 'domestication of society'.
This was essentially the appearance of a hierarchical delayed-return
organisation iliat Hodder terms 'the domus' (from tlle Latin word
meaning 'home'). After this, domestication occurred: 'it was
through the domus that the OI'igins of agriculture were thought
about and conceived ... The domus became the conceptual and
practical locus for the transformation of wild into cultural ... [and]
 Time, c1tangt and lite arcltaealog)' if IUl/ller-galherers 59

provided a way of thinking about the control of the wild

(Hodder 1990: 38-9).
To what extent al'e complex hunter-gatherers potential agricul-
turalists? In both northern Europe and eastern Turkey, such groups
,,'ere probably sedentary and collected nuts; wild cereals grew in
neither area. Domestication claims are based on pigs: there are
many juveniles in the archaeological assemblages, and this is
believed to result from domestic culling patterns (Zvelebil 1995,
Hongo and Meadow 1998, Rosenberg el 01. 1998). But is a juvenile
age structure likely to indicate domestication? Large sedentary
hunter-gatherer groups would increase hUllling pressure, and this
can lead to a more juvenile kill without any domestication being
involved (Elder 1965). The hunting ofjuveniles can markedly increase
the wild population, even though this seems counter-intuitive to
''''estern notions of sportsmanship. This is shown for moose in
Figure 3.8. (Grenier 1979). The effect will be e"en more marked in
pigs because of their multitudinous offspring. l"lan} juveniles in an
archaeological pig population is probably not about domestication;
it is simply about hunting more pigs.
It is refreshing to examine the best-researched complex hunler-
gatherer tradition knoml to archaeolog)\ and see how it has dealt
with agriculture. The Jomon ofJapan comprises mainly type 3 or
4 groups which lasted from the late Pleistocene until the first
millennium Be (Imamura 1996). l'vlany large sites are kno"~l,
including coastal shell middens and inland villages, and houses are
substantial (D and E in Figure 3.9). Sedentary occupation has
been demonstrated ,·ia shellfish (Koike 1980) and fish (Akazawa
1981). Salmon was a major quarry, involving a logistic strateg)'
and storage (Matsui 1996). In the interior, storage pits for nuts are
large and common (Imamura 1996 pp. 104-6). Some still contain
nuts; the oldest of these is from Higashi-Kurotsuchida, dated to
II 300+ 130 BP (Miyaji 1999). There are many cemeteries, and
society may have been hierarchical (Kobayashi 1992). This led to
early suggestions of agriculture (e.g. Kamjkawana 1968), but a
great deal of subsequent work has demolished this claim; suitable
plants are in fact absent (Imamura 1996: 106-9). Agriculture
 Figure 3.8. Elfect of hunting
DECISION
NOW r( choice on a moose popula-
tion. Encountering a female
and two calves, a hunter
year 1
rr may kill the adult; on
average one calf will die

year 2
rr wilhout the protective
female, while the other and
its progeny total five
animals after five years. Al-
year 3
rtr' f;(
"
ternatively the hunter may
IUU a calf, in which case the

year 4
rr\ rr ~'
adult female continues to

year 5
"
.", I 'rr rr rr fi1
breed, as docs the other
caU; producing 15 animals
after five years. (Modified
from Grenier 1979, Fig. 2.)

~u~e~~~RS'"
3 ADULTS, 2 CALVES
. . . .~
10 ADULTS, 5 CALVES
 Time, change and the archaeolog)' r!f hunter-gafherers 61

A. EARlY C. THUlE SEMISUBTERRANEAN DWEUING

CLASSIC
THULE
SETTlEMENT
...
2"'§})..AI
cqQV\N\1
-

B. LATE-
CLASSIC

"".,
THUlE
SETT\.EMENT

.
_....
~_.' ,._,~
C~ 19hI--.o-
<;7 .-.d no. .... ,I lIInW_'~

Q "0...,.........
".,0'" brlgf..-d
0
.........

E. MIDDLE JOMON HOUSE

• "" j/-.
°0
•
•
@ 0 (J

-...."

•
©J

Figure 3.9. Settlements and houses. A-B: Early and late Classic 'nlllie Inuit
settlemenls at Creswell Bay, northern Canada. (Redrawn rrom Sa-·clle 1987,
Figs. 87 and 90b). C: Early Classie Thule winter hOllse. (Redrawn rrom
Dumond 1977, Fig. III). D: Middle Jomon shell middcn at Takanc-Kido.
(Redrawn rrom Barncs 1993, Fig. 28). E: 1\ fiddlc Jomon housc rrom Idojili.
(Redrawn rrom Aikens and Higuchi 1982, Fig. 3.31). In A, Band D nOl all
houses need ha\'e been occupied at Ihe same lime; rom building phases are
represented i.n D.
62 PETER ROWLEY-CONWY

arrived from the Asian mainland in the frrst millennium BC, when
domestic pigs appear ishimoto 1994). The Jomon was not
'gelling ready' for this to occur' in most areas ofJapan population
was decreasing in the period before this. Only in the northeast,
the last area to go agricultural, was population incrcasing (Koyama
1978).
Eight thousand years of complexity did not lead to an indigenous
Jomon agriculture. .lomon studies have freed themselves from this
predestination, so groups can be examined for their own sake - not
for what they might become.

Assumption 6. The most interesting hunter-gatherers

are those who becamefarmers
Natufian type 3 or 4 groups inhabited the Near East for 3000 years
before the emergence of agriculture. They collected wild grass
seeds. The idea that this acti\'ity is a prelude to agriculture is
widespread. Arguing for autochthonous intensification in Australia,
Lolll'andos (1983: 92) describes 'the process being nipped in the bud
by the coming of the Europeans'; but as grass seeds were being
collected there before 25000 BP (McConnell and O'Connor 1997)
one may ask how much longer one would have had to wait. Harlan
(1989) argued that modern African seed collection with beating
baskets was a prelude to domestication, but this collection method
causes no genetic change in the grass (Hillman and Davies 1992);
wild sorghum has been collected in Africa since at least 8500 BP
(Magid 1995), so collection evidently need not lead to domestica-
tion. The belief that seed collectors were proto-agriculturalists led to
the Wadi Kubbaniya fiasco, where domestic barley, lentils and
chickpeas in contexts dated to 18000 BP were uncritically accepted
as evidence of Palaeolithic agJ;culture (Wendorf et al. 1982). These
items all proved to be misidentifications or recent contaminants
(Hillman et al. 1989).
So was the Natufian in any sense pregnant with agriculture?
Claims of Natufian domesticated cereals have proved groundless
(Legge 1986), and the collection of wild cereals at Abu Hureyra
 Time, change and Ihe mdlOeologl' r!f hllnler-galhtrtrs 63

decreased through time (Hillman et al. 1989). \Vild wheat and barley
were collected by 19000 BP in the Near East (Kislcv el al. 1992),
so here too there was a very long lead-in time. \'Vild grass seed
collection is by no means restricted to complex groups. In Keeley's
(1992) survey of 93 ethnographic hunter-gatherer societies, com-
plexity and \\~Id grass seed collection revealed no correlation
whatsoever.
There was tJlerefore nothing about the Natufian that made
agriculture ine\~table. Agriculture appears to have resulted from the
meshing of a series of unrelated factors of which the Natufian
delayed-return economy was only one; climatic change and plant
genetics were just as crucial (Hillman 1996). Had these factors nol
all come together, tJle Natufian might have continued hunting and
gathering indefinitely.
The Arctic is tJle area that all hunter-gatherer archaeologists
should know about. The impossibility of agriculture in the Arctic is
an antidote to tJleories of its ine\'itability elsewhere. Groups may use
ceramics and ground stone (Gusev el al. 1999) without being
suspected of Neolithic activities. Thule Eskimo in the central
Canadian Arctic were presented with a remarkable opportunity
around AD 1000: a warming climatic trend meant less sea ice, and
an increase in bowhead whales. Thule Eskimo actively hunted
these; even juveniles weighed up to 10 tons (Sa"elle and I\lcCartney
1991). This practice led to the construction of villages as large and
as sedentary as those of the Natufian (Figure 3.9A), consisting of
many heavy semisubterranean houses built of whalebones, stone
and turf and equipped with a coldtrap entrance tunnel (Figure
3.9C) - though not all houses need have been occupied at once.
Ceremonial kong; houses and circles of whale crania appear. Sites
are quite far apart, WitJl smaller camps and storage locations in
between tJlem, indicating a logistic strategy with a large logistic
radius (cf. Figure 3.1) (Savelle 1987; Sa\·dle and McCartney 1988).
Of all the type 3 and 4 groups described here, this is probably the
most remarkable. A cooling trend from AD 1200 decreased bowhead
whale availability; settlements became smaller, houses less perma-
nent, and kang; and whale skull circles ceased being constructed
64 PETER ROWLEY·CONWY

(Figure 3.9B). After AD 1600 further cooling decreased whale

availability even more. At European contact the Netsilik people in
the area used many short-lived settlements to exploit sequentially
available rcsources. They were among the least storage-dependent
of any Inuit groups (Balikci 1970) and so most closely approached
the OAS.
Agriculture could never have been an outcome of Thule com-
plexity; but had such a spectacular archaeological manifestation
occurred anywhere else in the world it would surely have been
suggested.

Conclusions

There is no directional trend among hunter-gatherer socIetIes.

Numerous examples reveal complexity coming and going frequently
as a result of adaptive necessities. The adaptationist view cannot be
reconciled with progressivist theories, whether demographically or
socially based, but is in stark opposition to them. Most hunter-
gatherers who became farmers have done so as the result of stimuli
from agricultural neighbours. Hunter-gatherers with no agricul-
tural neighbours originated agriculture very rarely, perhaps only
three or four times - empirical evidence for the low likelihood of
such an event occurring. Iv10st hunter-gatherer historical trajec-
tories would never have resulted in agriculture had that way of life
not impinged on them from the outside.
In 10 000 years agriculture and its outcomes have come to
dominate the world. Agricultural economies, despite catastrophes
and reversals, can usua]Jy be intensified: new animals can be
domesticated, ploughs can be made more elfective, locally adapted
crops such as oats can be added, animals can provide traction and
other secondary products, transport and redistribution systems can
be imprO\·ed, pests controlled, industrial fertilisers produced, and
crops genetically modified. Hunter-gatherer economies were not
like this. Salmon, nuts, reindeer or grass seeds cannot be intensified
indefinitely; the limits to growth will be reached unless domestica-
tion follows.
 Tillie, rhange alld the archaeolog)' if IlUnter-gathfrers 65

If the Original Arnuent Society is not 'original', what is it?

Archaeologists have onen regarded terroriality, rigid group mem-
bership and social hierarchy as stepping stones from the OAS
towards ourselves, but these features arc all found among chimpan-
zees. From this perspective the flexibility, mobility and social
equality of the Original Affluent Society may be the most remark-
able and special.ised social form that humans ha"e ever e,·olved. It
has no claim to be the original human condition.

Acknowledgexnents

Thanks to Tim l'vlurray, David Frankel and the Department of

Archaeology at La Trobe Uni'·e.rsity in Melbourne where this
chapter was written, to 1\'lary Stiner and John Parkington for
unpublished work, to Pamela Rose, Phil Edwards and Roberto
Maggi for literature, and to A1essandra Buonfino for checking my
Italian translat.ion. An early version was delivered as the Meyerstein
Lecture in Oxford in 1997; thanks to Helena Hamerow.

References

Aikens, C.M. and Higuchi, T (1982). PrehisloT)' rif]apall. Nell' York: Academic
Press.
Akazawa, T (1981). Maritime adaptation of prehistoric hunter-gatherers and
their transition to agriculture inJapan. Sen'; EI/lIIotogiral SllIdies 9,213-258.
AlIen,.J., Gosden, C.,Jones, R. and White,J-P. (1988). Pleistocene clates for the
human occupation of Nell' Ireland, northern ~Ielanesia. Na/llre 331:
707-709.
Andersson, M. and Knan'Strom, B. (1997). Plals 7A: I - etl mesolilisk
boplalSkomplex vid sammanOodel Braan-Saxan. In SMile, Malmohlls WII,
Jonwiigall ViisUmslballall, A/'SI/illfl ulI/dsloona-hnl'finCt /996- J997, Ad;mlogisk
FOnmt!ersolo/ing. (vol. I), ed. ~1. Svensson and P. Karsten, 75-99. Lund:
Riksantikvarieambctet.
Balikci, A. (1970). 771f.Nflsilik Eskimo. New York: atuml History Press.
Barnes, G. (1993). ChulI/, /(orta alld Japall: Ihf Rise rif CiI,ilisalioll ill Easl Asia.
London: Thames and Hudson.
66 PETER ROWLEY~CONWY

Bender, B. (1978). Gatherer-hunter to farmer: a social pe",pective. 1V0rid

ArrllOeologl, 10: 20+-222.
Binford, L.R. (1980). Willow smoke and dogs' tails: hunter-gatherer sellle-
l1''1ent systems and archaeological site formation. Amm·coll Alltiqui!)t 45:
4-20.
Binford, L.R. (1991). I\lobilit~; housing. and environment: a comparative
study. JOIln/al '!fAlllhropologiral Research 46: 119-152.
Birdsell, j.B. (1957). Some population problems invoh-ing Pleistocene man.
OJId SP,;lIg Harbor S."mposia 011 O!.wllii/otil" Biologl' 22: 47-69.
BowdlcJ; S. (1977). The coastal colonisation of Australia. In SlIlIda alld Sal/lll,
cd. j. Allen. j. Golson and R. Jones, pp. 205-246. London: Academic
Press.
Brink,j.S. and Deacon. HJ. (1982). A Sludy of a last interglacial shell midden
and bone accumulation at Herolds Bay, Cape Pro\'iner, South Mrica. In
Proceedillgs '!f Ihe rJl/' Binmial CO/yermc< held a/ Ihe TralLS/'aal Alustum, Preloria,
26-29 Mqy 1981, ed.j.C. Vogel, E.A. Voigt and TC. Partridge, pp. 31-39.
ROllerdam: A.A. Balkcma.
Bro\\1l,j.A. and Price, T.D. (1985). Complex hunter-gathere"': retrospect and
prospect. In Prehisloric Hmllcr-Ga/lmm. 77" Emagfllet '!fClllhtral Complexi!" cd.
TD. Price andj.A. Brown, pp. 435-442. ew York: Academic Press.
Cardini, L. (1980). La necropoli mesolitica delle Arenc Candide (Liguria).
Memorie deWlsli/lllo Iialiallo di PalfOlI/ologia Umalla 3: 9- 31.
Christensen, C., fischer, A. and I\lathiassen, D.R. I. (1997). The great sea rise
in the Storeba:h. In 77" DOllish Sloreb..11 sillce Ihe Ice Age, cd. L. Pede",en,
A. Fischer and B. Aaby, pp. 45-54. Copenhagen: A/S Storeba:h fixed
Link.
Clark, G.A. and Neeley, M. (1987). Social differentiation in European
mesolithic burial data. In Alesoli/hic Northwesl Ellro/If: Reeml Trellds, cd.
P. Rowley-CoJl\\'Y, M. Z\'c!cbil and H.P. Blankholm, pp. 121-127. Shef-
field: Uni\'e",ity of Sheffield, Department of Archaeology and PrehistOlY
Cleyet-I\!erle,j.:l. (1990). fA Plihisloire de 10 Pi,/'e. Paris: Editions Errance.
Close, A.E. (1996). Plus ,a change: the Pleistocene-Holocene transition in
northeast Africa. In Hllmalls 01 Ihe £I,d '!f tite Ice Age, cd. L.G. Straus, B.V.
Eriksen, j.1\1. Erlandson and D.R. Yesner, pp. 43-60. ew York: Plenum
Press.
Clollcs,j. (1999). Twenty thousand yea", of palaeolithic cave art in southern
France. In World Pr,hislol)" SllIdies ill Ml1110~1' '!f Grahame Clark, ed. j. Coles,
R. Bewley and P. I\lella"', pp. 161- 175. Oxford: Oxford Uni\'e",ity Press.
(Proceedings of the British Academy 99).
Diamond,j.1\1. (19117). Who were the fi"'t Americans? Nalure 329: 580-581.
Dumond, D. (1977). 77" Eskimos alld Aleu/s. London: Thames and Hudson.
Dyson-Hudson, R. and Smith, E.A. (1978). Human territoriality: asl ecolo-
gical reassessment. Amerirall Alllhropologisl 80: 21 -41.
 Time, change and the archaeology qf hUlltrr-gatherers 67

Elder, W.H. (1965). Prime\'al deer hunting pressures re\'ealed by remains from
American Indian middens. ]ollmol qf 11r,1d1if- MonagmrfTlt 29: 366- 370.
Fis<her, A. (19970). People and the sea - selliement and fishing along ti,e
mesolithic coasts. In The Danish Sto"b"lt sinff the lee Age, cd. L. Pedersen, A.
Fischer and B. Aab); pp. 63-77. Copenhagen: A/S Storeb",lt Fixed Link.
Fischer, A. (1997b). Monneoruologiske jonmdersoge/ser jorud jor rtab/ering qf til fast
Oresnlldiforbindelse, vol. 2. Copenhagen: l\Jiljo- og Energiministeriet, Skov-
og Naturstyrelsen.
Fischer, A. and Maim, T (1997). The seHlement in the suhmerged forest in
l\lusholm Bay. In The DOllish Storeb"lt since the lee Age, cd. L. Pedersen,
A. Fischer and B. Aaby, pp. 78-86. Copenhagen: A/S Storeh",1t Fixed
Link.
Fitzhugh, W. (1984). Residence paHern development in the Labrador Mar-
itime Archaic: longhouse models and 1983 surveys. In Arc/lOeologl' in New-
fOlll/dland and Labrador - J983, ed.JS. Thomson and C. Thomson. pp. 6-4 I.
Sl. John's, Newfoundland: Historic Resources Di\'ision, Department of
Culture, Recreation and YOUU1.
Flannery, TF. and Whi,e.JP (1991). Animal translocation. Notional Geogrol,hi,
Research and ExplorotiOll 7: 96- I 13.
Gargell, R. (1999). Middle palaeolithic burial is not a dead issue: the view
from Qafzeh, Saint-Ccsaire, Kebara, Amud, and Dederireh. ]ollrnol <if
Hllmon £<"llItion 37: 27-90.
Grenier, P 1979. New regulations on moose hunting in Quebec. CmlOdion
1Ir,ldlif- Administration 4: 28-31.
Guse\, S. \~, ZagoTOulko, A. V. and POTOto\; A.V. (1999). Sea mammal hunters
of Chukotka, Bering Strait: recent archaeological results and problems.
II orld AldlOtologl' 30: 354-369.
Ilarian.JR. (1989). "'ild-grass seed han'esting in the Sahara and Suh-Sahara
of Aliiea. In Foraging and Fanning, ed. D.R. Han;s and G.C. Hillman,
pp. 79-98. London: U,lI\;n Hyman.
Hayden, B. (1972). Population control among hunter-gatherers. II orld Archoe-
olog)' 4: 205-221.
Hillman, G.C. (1996). Late Pleistocene changes in wild plant-foods available
to hunter-gatherers of the northern Fertile Crescent: possible preludes to
cereal domestication. In D.R. Hams (cd) TIe O,igins and Spread <ifAgriClJture
and Pastoralism in Enrasia, 159-203. London: UCL Press.
Hillman, G.C. and Da\.;s, M.S. (1992). Domestical'ion rate in wild wheats and
barley under plimitivc cultivation: preliminary results and archaeological
implications or field measurements or selection coefficient. In Aihislo;,.e de
l'AgliCIIltllre: Naill/tiles Approches Etpbimelltnles et EtllIlogral,hiqlles. Centre de
Recherches Archeologiques l\lonographe 6, cd. PC. Anderson.
pp. I 13 - 158. Pal;s: CNRS.
Hillman, G.C. Colledge, S.I"!. and Hams, D.R. (1989). Plant-food economr
68 PETER ROWLEY-CONWY

during Ihe epipalaeolilhic period al Tell Abu Hureyra, Syria: dietary

diversily, seasonality and modes of exploilation. In Foragillg alld Fanllillg,
ed. D.R. Harris and G.C. Hillman, PI'. 240-268. London: Unwin
Hyman.
Hillman, G.C., Madeyska, E. and Halher,J. (1989). Wild planl foods and diet
al lale palaeolithic Wadi Kubbaniya: the evidence from charred remains. In
TIe PrehisloT)' 'If I"odi li;,bballiJ'a, vol. 2, Stratigra#!l', PaleoecollolI!Y alld Em,iroll-
menl, ed. F. Wendorf, R. Schild and A.E. Close, PI'. 162-242. Dallas:
Southern l\lcthodisl Uni\'ersity Press.
Hodder, I. (1990). 171e Domesticatioll 'If Ellrope. Oxford: Blackwell.
Hongo, H. and l\leadow, R.H. (1998). Pig exploilalion at neolithic Qayonli
Tepesi (soulheastern Analolia). I\L~SCtI Research Papm ill SciellCf alld Archae-
ologl' 15: 77 -98.
Hood, B. (1993). The Marilime Archaic Indians of Labrador: investigating
prehistoric social organisalion . .Ne"folllldialld Stlldies 9: 163- 184.
Imamura, K. (1996). PrehisloricJapall. London: UCL Press.
Kamikawana, A. (1968). Siles in middle Yamanashi-ken and Middle Jomon
agriculture. Asiall Perspeclil~s II: 53-68.
Kedey, L.H. (1988). Hunler-gatherer economic complexity and 'population
pressurc': a cross-cuhural analysis. JOllmal 'If Allthropological Archaeolog)' 7:
373-411.
Kcele); L.H. (1991). Ethnographic models for late glacial hunter-galherers. In
171e Lale Glacial ill Norllw',st Ellropr, cd. N. Barlon, AJ. Roberts and D.A.
Roe, PI'. 179-190. London: Council for Brilish Archaeology.
Keeley; L.H. (1992). The use of plant foods among hunter-galherers: a cross-
cultural survey. In Prehistaire de tAg,;wllllre..NolII~lIes A/JProches Experimentales er
£tlmogml,hiqlles Centre de Recherches Archeologiques Monographe 6, cd.
PC. Anderson, PI'. 29-38. Paris: CNRS.
Kelly, R.L. (1995). 17" Foragillg Spertmm: Dillern!)' ill Hllllier-Galherrr Lifru,,!J's.
Washington DC: Smithsonian Institution Press.
Kisle\; M.E., Nadel, D. and Carmi, I. (1992). Epipalaeolithic (19000 HP)
cereal and fruit diet at Ohalo II, Sea of Galilee, Israel. Relliew 'If PalaeobolOl!Y
alld Pa!yllologJ' 73: 161 -166.
Knecht, H. (1993). Early Upper Palaeolilhic approaches 10 bone and antler
projectile lechnology. In Hlilltillg alld AII;mal Exploilatioll ;11 Ihe Laler Palaeolithic
alld Mesolithic 'If Ellms;a, cd. G.L. Peterkisl. H. Bricker and P Mellars,
PI'. 33-47. Archaeological Papers of the American Anlhropological Asso-
ciation 4.
Kobayashi, T (1992). Patterns and levcls of social complexil)' inJomonJapan.
In Pacific N0/1heost As;a ill i'lehisto1)\ ed. C.M. Aikens and S.N. Rhee,
pp. 91-96. Pullman: Washinglon Stale Uni\'ersily Press.
Koike, H. (1980). SPasolial Datillg ~)' Growth-Lille CO/llllillg 'If the Clam, l\ leretrix
lusoria. Tokyo: University Museum, University of Tokyo.
 Time, change and Ihe archaeolog)' ~ h/lnler-galherers 69

Koyama, S. (I 978).Jomon subsistence and population. &mi Elhnological SllIdies

2: 1-65.
Larsson, L. (1982). Segebro: E71 Tidigalla71lisk &plalJ "in Segf Aj "'!wlIlillg. l\lalmo:
l\-Ialmo IIluseum.
Layton, R. (1986). Political and temtorial structures among hunter-gatherers.
Ma71 (N.S.) 21: 18-33.
La)10n, R. (1987). The usc of ethnographic parallels in interpreting upper
palaeolithic rock art. In Comparalie. Alllhropolog)\ ed. L. Holy, pp. 210-239.
Oxford: Blackwell.
Layton, R. Foley, R. and Williams. E. (1991). The transition between hunting
and gathering and the specialised husbandry of resources. Cllrrflll Anlhro-
pololJl' 32: 255-263.
Lee, R.B. and DeVorl', I. (eds.) (19680). Man Ihe J-Itmler. Chicago: Aldine.
Lee, R.B. and DeVore, I. (I 968b). Proh1cms in the study of hunters and
gatherers. In Man tilt /-I1I1I1a, cd. R.B. Lee and I. DeVore, pp. 3-12.
Chicago: Aldine.
Legge, AJ. (1986). Seeds or discontent: accelerator dales on some chan'cd
plant remains rrom the Kebaran and Natu/ian cultures. In Archaeological
RU71lls from Aeeelfralor DOli71g, ed. J-AJ. Gowlett and R.E.1I1. Hedges,
pp. 13-21. Oxford: Oxrord Unh'ersity Committee ror Archaeology.
Lourandos, H. (1983). Intensification: a late Pleistocene-Holocene archaeolo-
gical sequence from southwestern Australia. ArrlloeologJ' i71 Orea1lia 18:
81-94.
Lubell, D., Sheppard, P. and Jackcs, II I. (1984). Continuity in the epipalaeo-
lithic of North Arrica with emphasis on the Maghreb. Anl'071(fj ill World
ArchafololJl' 3: I43- I91.
IIladatiaga de la Campa, B. and Fernandez Pato, C.A. (1985). Esludio
malacologico de la CUC"a 'EIJuyo'. In Erravaaoll'J nl 10 Ci,el:O del )19'0, cd. I.
Barandiaran, L.G. Freeman, J-G. Echegaray and R.G. Klein, pp. 75-95.
Madrid: IIlinisterio del Cultura.
IIlagid, A.A. (1995). Plant remains rrom the sites or Aneibis, Abu Darbcin and
EI Damer and their implications. In Aq71alilhic Siles along Ihf Ri,.rs Nilf a71d
Albara, S71dan, cd. R. Haaland and A.A. Magid, pp. 147-177. Bergen: Alma
IIlater.
Matsui, A. (1996). Archaeological investigations or anadromous salmonid
fishing inJapan. World ArchalOlolJl' 27: 444-460.
McBryde, I. (1984). Kulin greenstone quanies: the social contexts or produc-
tion and distribution ror the Mt William site. II arid ArchafololJl' 16:
267-285.
McConnell, K. and O'Connor, S. (1997). 40000 year record offood plants in
the Southern J(jmberle)' Ranges, Western Australia. Ardralian Arc/wlOlog!'
45: 20-31.
Meillassoux, C. (1973). On the mode of production or the hunting band. In
70 PETER ROWLEY-CONWY

Frn/,h Persp«lil~s ill Africall SllIdies, cd. P. Alexandre, PI'. 187-203. Oxford:
Oxford Uni"ersilv Press.
1\ Iiyaji, A. (1999). SlOrage pits and Ihe development of plant food manage-
ment in japan during the jomon period. In Bog Bodies, So,ud Siles alld
II tllalld Archatalagll ed. B. Coles, J Coles and M. Schou-Hansen,
PI'. 165-170. Exeler: Department ofAschaeology.
I\I0rales, A, Rosello, E. and Hermindez, E (1998). Late upper palaeolithic
subsistellCC strategies in southern Ibel;a: tardiglacial faunas from Cue"a de
NeIja (I\hilaga, Spain). Eilroptall]ollmal qfArchneolog,. I: 9-50.
I\lorse, D.E Anderson, D.G. and Goodycar, AC. (1996). The PleislOcene-
Holocene transition in the eastern United Stales. In Hllmnlls Ihe Ihe Elld qflhe
I" Ag', ed. L.G. Straus, B.V Eriksen, J,M. ErlaJ1(lson and D.R. Yesner,
PI'. 319-338. Ne\\' York: Plenum Press.
1\Iorse, K. (1988). I\landn I\landu Creek rock.,helter: Pleistocene human
coastal occupation of orth West Cape, Western Australia. Arc/lOtalog)' ill
Orenllin 23: 81 -88.
I\luh-aney, DJ. and K.amminga,J, (1999). Prehislo1)' qf Allslmlin. St. Leonards,
SW: Allen amI Un,,;n.
Nishimoto, 'I (199-l). Domesticated pigs in the early agricultural period in
japan. Archifo.(oologin 6: 57-70.
Ortea, J, (1986). The malacology of La Riera cave. In UJ Riera Cal~ Anthro-
pological Research Papers 36, ed. L.G. Straus and G.A. Clark,
pp. 289-313. Tempe: Arizona State Uni,·ersity.
Pardoe, C. (1988). The cemetery as symbol. The distribution of prehisloo;c
Aboriginal bUI;al grounds in southeastern Ausrralia. Ar'haeolOgJ' ill Oceallia
23: 1-16.
Parkington, J, (in press). The impact of the systematic exploitation of marine
foods on human emlution. Proceedillgs qflhe Dllal COllgresS, SIIII Ci!r 1998.
Price, T.D. (1981). Complexity in 'non-complex' soeietie . In Archaeologiml
Approa,hes 10 Ihe S11I11' qf Complexi!)', ed. S.E.\'an der Leeuw, pp. 55-97.
Amsterdam: University Press.
Price, T.D. and Bro\\'n, JA (eds.) (1985). Prehislori, lillllier-Galh""rs: 77"
Emergellre qfClIllllral Complenl)' New York: Academic Press.
Rosenberg. 1\1., j esbill, R., Redding, R.W. and Peasnall, B.L. (1998). Hallan
C;:emi, pig husbandl')\ and post-Pleistocene adaptations along the Taurus-
Zagros arc (furkey). PaliO/1e111 24: 25-41.
Rowley-Com,,)', P. (191l3). Sedentar), hunters: the Ertebolle example. In
filmler-Calhmr £fOIlOIl!t' ill Prehislo~l\ ed. G.N. Bailey, pp. 111-126. Cam-
bridge: Cambridge Universily Press.
Rowley-Conw)', P. (1998). Cemeleries, seasonality and complexily in the
Ertebolle of southern Scandina,~a. In lial1~slillg Ihe Sen, Fanlling Ihe Foresl, cd.
1\1. Z"e1cbil, L. Domanska and R. Dennell, Pl'. 193-202, Sheffield: Shef-
field Academic Press.
 71me, challge alld Ihe archaeology q! 1lIlIller-galherers 71

Rowley-Conwy, P (1999). Ecollomic prehistory in southern Scandina,~a. In

Il'orld Prehislol)': Sh,di,s ill MflIIO~l' if Gmhame Clark, Proceedillgs if the British
Acadfl/~)\ ed. JI\'1. Coles, R.I\J. Bcwley and PA. I\Ieliars, pp. 99,
125-159.
Sahlins, M. (1968). Notes on the original amuent socicty. In ,'vfan the HUllter, cd.
R.B. Lee and I. DeVore, pp. 85-89. Chicago: Aldine.
Sahlins, M. (1972). Slane Age &o/lOl1Iics. Chicago: Aldine.
Savelle, J (1987). Collectors alld Foragers: Subsistence-Selllel1lmt Sptt/II Change ill the
Cel/lral Canadiau Arclic, A.D. /000-/960. Oxford: British Archaeological
Reports.
Savelle,JI\I. and McCarmey, A.P (1988). Geographical and temporal ,'ana-
tion ill Thule Eskimo subsistence economics: a model. &search ill &ouol1lic
AUlhropolofJ' 10: 21 - 72.
Savelle, JM. and I\IcCartney, A.P (1991). Thule Eskimo bowhead whale
procurement and selection. In Humall Predators alld Prry MOitali~)\ ed.
1\1. Stiner, pp. 201-216. Boulder: Westview Press.
Saxe, A. (1970). Social dimensions of mortuary practices. Unpublishcd PhD
thesis, Uni"ersity of Michigan .
Schmitt, L. (1995). The West Swedish Hcnsbacka: a maritime adaptation and
a seasonal expression of the North-Central European Ahrensburgian. In
Mall alld Sea ill Ihe Mesolilhic, Oxbow Monograph 53, ed. A. Fischer,
pp. 161- 70. Oxford: Oxbow Books.
Service, E.R. (1962). Primitive Sacial Orgallisation: All Evolulionary PfI"p€cli"e. New
York: Random House.
Sorensen, S.A. (1996). KougflllOsekultureu i Sj,dskalldilll1l';fIl. ]a:gerspris: Egnsmu-
see I Fa:rgcgarden.
Stiner, M.C. (1999). Palaeolithic mollusc exploitation at Riparo Mochi (Balzi
Rossi, haly): food and ornaments from the Aurignacian through Epigra"et-
tian. AUliquity 73: 735-754.
Thommessen, T. (1996). The early settlement of northern Non,·av. In 171t
Earliest Selllelllflli if Scmldilla,.ia, cd. L. Larsson, pp. 235-240. Stockholm:
Almqvist and Wiksell. (Acta Archaeologica Lundensia, Series in 8°, 24)
Vandermeersch, B. (1969). Decouverte d'un objet en ocre avec traces d'utilisa-
tion dans Ie Moustenell de Qafzeh (Israel). Bullelill de 10 Sociili Prihislarique
Fmllraise 66: 157- 158.
Vcrmcersch, P.M., Paulissen, E. and Van Neer, W (1989). The late palaeolitl'ic
Makhadma sites (Eg)1)t): environment and subsistence. III Late Prehistol)' if
tire Nile Basin and Ihe Saham, cd. L. Krzyzaniak and M. Kobusie\\~cz,
pp. 87-114. Poznan: Archaeologicall\luseum.
Wendorf, E (1968). 171t Prehistol)' if Nubia. Dallas: Southern Methodist Uni-
versilv Press.
Wendorf, E, Schild, R. and Close, A.E. (1982). All ancient han'est on the Nile.
SCzlrlce 82: 68-73.
72 PETER ROWLEY-CONWY

Wickler, S. and Spriggs, 1\.1. (1988). Pleistocene human occupation of the

Solomon Islands,1\lelanesia. Allliqui!v 62: 703-706.
Wiessner, P. (1982). Risk, reciprocity and social influences on !Kung San
cconomics. In Polilics alld Hislo~,' ill Balld SocielitS, ed. E. Leacock and R.B.
Lee, pp. 61-8·t Cambridge: Cambridge University Press.
"'oodburn, J. (1980). Hunters and gatherers today and reconstruction of the
pas\. In Sollitl alld II (simi AlllhropologJ', ed. E. Gellner, pp. 95-117. London:
Duckworth.
Zvclebil, 1\1. (1995). Hunting, gathering, or husbandry? Management of food
resources by the late 1\ Icsolithic C"ommunities of temperate Europe. AJASCA
RfStardl Papers ill Srifllct alld ArchatologJ\ 12 (Supplement): 79-104.
Z,·e1ebil. 1\1. Domanska. L. and Dennell. R. (1998). Introduction: the Baltic
and the transition to farming. In Hon',slillg Ihf Sea, Famlillg Ih, FortSl, ed.
1\ I. Z,·e1ebil. L. Domanska and R. DennelJ, p. 1- 7. Sheffield: Sheffield
AcademiC" Press.