The Origins of House and Home?

Trevor Watkins
World Archaeology, Vol. 21, No. 3, Architectural Innovation. (Feb., 1990), pp. 336-347.
Stable URL:
http://links.jstor.org/sici?sici=0043-8243%28199002%2921%3A3%3C336%3ATOOHAH%3E2.0.CO%3B2-0
World Archaeology is currently published by Taylor & Francis, Ltd..

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/taylorfrancis.html.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.org
Fri Aug 24 07:21:40 2007

The origins of house and home?

Trevor Watkins

Introduction

I n the two seasoris of excavation at the site ofQea.me7 Dere in northern Iraq (SCL:F ig, i 1 thi"
centre o f interest i n one of the two areaa of excavation ha?; been a n extremely ear!) anti
unustral house. which was repeatedly rc-modellcd and i.cbuilt. 'l'lie sitc is that OK ii :,nrrali
settlement belonging to the end of the epi-palaeolitliic period anc! the very early ai:ei-;~i-r.ic
neolithic, approximately 8500 to 7900bc (uncalibratrd radiocarbon) (Warkins i~rnci\%aird
1987: Watkins, Baird and Bctt:;, forthconling: Watkins. l3aii.d. Beits an<%T?icabiei
i'orthcorning). At that early date and in that part o f the Near East the a~.chite(:tr:r.c:
Qernlez D c r e is without precetlcnt; only the site of Nernrik (see !i;ozlowi;ki and i<cnlpisry,
this vol~rnic).whose first occup;iti~)nmay be at the end of the ninth nli1lcnniui-i-r.pro\~idc:~,
relevant parallels. 7'hc excavations have been nrounteci by tllc Department of Aici-r:ic,
ology. University of Edinburgh, at thc suggestiicin or ti-lc Iraqi Directors-ic of Aniicjrritic.;
anci Heritage, because the site was seriously tlarnaged ant1 in d a n g c ~of ohli~erntior-i.
-I h e context of social and econol-illc processes -.~litIiir-i\vrjliichQ e ~ r r ~ eDr r r e s/~ol!ldbe
vicwcd, which started life as the "neolithic rc\olution', I?; !;till on? of li\cljv i-c.si:ril-ch
interest. Early research. suy-sporieil by tirc then new radio.carboi~dating !nctllod. pr:illcd
back the beginning of the neolithic t o 8000 bi: arlci beyond, but ki development o f [iio !;i!;j
decade has been the recognition that a n increasing body of data poinits to later ti;alc:.: lor tile
systematic cultivation of plants and especially tlhc hi1 domestication of animals t h a n Lvas
originally thought when sires were first labelleci 'ncolithic. Sophistic;~tedcormbin;iiiorlc, of
harvesting, gathering, hunting and culling - Flannery's 'broad spectrum economy' - c:rn
now be seen tci have continued long into the periotl technically labelled neoiiihic. anti
settlement, once thought specific to the neolithic and later periods.. has become a re!?.nri;ir
fcaturc of the p~reccdingepi-palacolithic. 'l'he distinction between the neolithic ar-lcl thlc
preceding periocl is n o longer based on changes in s~ahsistencestrategy a ~ r dthe adoption of
sedentary village life but once more on formal diffcrenccs in n~aterialculttlre: :>articrrl;irly
the chipped stone inclustries.
This essay is an atter-rapt to explore something of the nature of the differences betvS~ec:n
tlie two periods at a level deeper than those forr~raldifferences. It seeks to define orlc oi'tira:
inrportant contributory steps taken in the period before the total, i t d ~ p t i ( >ofl ~cultit alirirl
and herding. T h e assumption made is that in any g i v c ~ iarea there is no perceptiblt:
difference in tlie subsistence strategies of the later egi-paliicolithic arrd earliest neolithic
:>eriods. Consccjuently, any significant difference slioulci bc soirglit in the cultural. social or
>

The origin5 o f hoii~eaizd I~orne') 337

*-

I RAO

Figlire I

Map showing location of Qcrnicr Ilerc.

behavioural aspects of these communities. It is the thesis of this essay that Qermez I l e r e
shows important architectural developments taking placc at the beginning of t.he neolithic
period, about 8250-8000 bc. developments which signify important social changes. These
changes, it is argued. are signified in thc title of the essay, a change of view from the house
as shelter, t h e centre of certain everyday activities, to the house as home, the centre of the
family and the focus for the representation of appropriate symbolic values.

Archaeological backgrour~d
Although Qermez D e r e is very severely damaged and only small parts of it have survived
for excavation, it seems that the site can be divided into two distinctly different zones on
the basis of the different types of deposit. Around the margin of the wadi and the south side
of the site the deposit is fine loam containing occupation debris; the houses which form the
subject of this paper are located in the southern part of the site. T h e deposits in the central,
nortl~eastand eastern parts of the site derive from a red-brown clayey marl; in the
soundings so far undertaken there is n o trace of houses. but there is a concentration of
heavy food-processing equipment, especially boulder mortars. suggesting that this part of
the settlement may have been an open, communal activity area.
In the preserved part of the south margin of the settlement a deposit of at least 1.5m had
accurnulated before the construction of thc earliest house so far located. This house and its
two successors were subterranean. T h e str;it.igraphy of this part of the site has been defined

in terms of seven successive phases. The three phases of house-building and occupation arl:
the last three i11 the sealed stratigraphic sequence. The four early phases are rep:~rcsentcd
hy a series of super-imposed strata, of which the iowest tlvo phases are epi-palaeoiithic i n
date.

The horlses

The remains of three subterranean constructions have been excavated (Figs 2 and 3). -l'i?cy
arc very similar in C O I I S C S L I Cco~iiisti~ig
~ ~ O I I . of a single chamber formed by plastering the
sides and base of a large hole cut into the earlier- levels or the f i l l of the previous tiuii(ii11g.

,,

upright stone

FzB i n f e r r e d

Simplified plan of t h c three Ilouscs

positlnn

E;tch replaced the cllnc:r when i h c old housc wits :it-jantloned anti cies!ro>e:i. I'iic: tbrec
plans overlap, so that bhc process o f replaccrricrit invoivcd lillinp. "1' the old Iiou\e and til(:11
re-cutting a hole for the new o n e .
The shape of the chamber:; is hetwceil round ani! rcctanguiar. None i:, rruiy circ~alas
(though two are rather egg-shaped), and nonc has a clearly articul21ted corncr angle: o~ii?;
o n e sub-phase of o n e house has a reaso~iabiystraight s ~ d e There
.
is little evidence of
above-ground structure o r of rooting, sincc the flls of the abandoned chambers mostly
consist of deliberately introduced soil and other- debris. It is therefore not possible to
reconstruct the upper parts of the I2uildinlz;s. even to the extent o! knowing whether the
h o ~ i s e swere fully subterranean. Nont: of the chambers shows evidencc of the means of
access; there are no doorways o r steps dolvn the inside of the walls. T h e better preserved
houses at Nernrik, only about 60kn1 away on the Tigris, were probably entered by means of
a free ladder reaching into thc floor area (cf. p . 355).
At Qermez D e r e the original construction (so far k ~ i o w ~ ihouse
),
RAB. was built in a
roughly rectangular hole at least O.85m and perhaps m o r e than 1.5m deep. T h e surface of
the cut was faced first with a red-brown clay and then with a thin coat of fine white lime
plaster. T h e slightly concave floor was finished with the same white plaster and had an area
of around 24 square metres. Where housc R A A overlay it. much of the floor of RAN has
been damaged and is not reconstructible, \jut the west side of the house is better preserved.

H e r e there were foilr holes in the plaster iloor in a line parallel wich the long axis or thc
chamber. None of these has yet bcen investigated, but one. labelleti RCLon the plar~(Fig.
2 ) , is fairly certainly the place from which had been broken a tall standing slab of stone
coated in red clay and white plaster (restored in Fig. 3 ) . Around the foot of the pillar the
iloor had been modelled to stnootli the {ransition from floor tr:, pillar., ancl a ridge in [lie
floor was carried from the base of the pillar towards the ccntrc of the charnbcr.
T h e pillar ivas found in pieces mrhcrc it had Fallen o r been felled when the house was
demolished and filled in. It had bccn re-modelled. for inside its outer coat of plaster a n d
tauf \$?asfouaid another thin layer of plaster. In its fi~ialform the pillar was a simple slah
(Fig. 4b). but in its earlier Form (IYig..$a) it had tapered ~ ~ p w ; l r dand
s been roux~dedat the
sidis. Near the top of the slab the c~irlicrcoating of tauf h a d bern modelled into ;I definite
shape, rather like a pair of shouiders \+,it11raised arms truncated to the merest stumps. I n
both ith l'orms the pillar had something modelled in plaster-coated tauf upon i t s tog,, thc:
shattered pieces of ivhich \$?erefound beyond the end of the i'al1a:n pillar.
b'hcn house RAB had bcen demolished a n d its chamber filled, it was I-ep!accd b!
another. similar liouse (RAD) constructed in a hole which was Icss deep a n d somc:\vhai
smaller i i i floor area (about 18 square metres). Only ;rbotit half ol' house RAD hatJ
survived, for when its successor (RAA)was constructed its southerlr half \+as diig ;1n7;ib,.
-I'liis 5urvicing part is elliptical in plan.
Across tlle long a x i \ o l I I ~ L I SRAP)
~
therc were three constructions in a line. At thc
centre was a large, rough slab ol'stonc. set on edge into thc plaster of the floor. ' l o t.it1:e.r.
sick \$?asa clay pillar built around a stone core and finished with the usual white piiar;ti:r,
The cores of the two pillars were long slabs oi'limcstonc, firm!y lie1 into pits \vcIl beiow rhi:
iloor of the house. T h e gre),-green clay m~odcliedaround ilie stone cores was the sai:ice ;is
that used iii the construction of the sicles and Aoor of ti-ii: ci-~ar~ibci-.
Both pillars hatl brcis
truncated, almost certainly bq bulldozing. ' l ' l i ~ ~i ts is no1 possible to recont;trua:i 11ieir.
original heights o r the Features of their upper parts. A6 some time the floor ~t,;is
re-plastered and the two pi11;1r4 enlarged.
When house R A D was filleci i11. its successor (RAA)n r ; i construcieci ova:r its sou~,lier-a~
part and the eastern part of RAB; the floor level of' RAA svas dug lower than that crf 111c
earlier R A D . At first the norlh encl of R A A was formed ia-x~mcdiatel!.m u t h of thc 1i:ie oi'
pillars and standing stone i n RPPLI. its side walls t~clopti;lgext~ctlythe same alignment 21s
those of its pretlecessor. A t a later stage R A A mas re-.modelled. it:<north w;~Ilbccorrring n
broad curve. giving the chamber an egg-shapecl plan about 4.8by 4in and soine Ihclrrark:
metres iai area.
House R A A was coated with grey-green clay finished ~ v i t hthe usual fine \vhite plaster,
In the centre of the floor was a scorched area with a few stones set irs the piaster arcprxnci i I
011an earlier floor there was an equivalent hearth a little 2 0 the west, consisting o l ;i neat
circle of scorched plai;ter about :iOcrn across. Close to tilc licarth was a sirsgle clay pillar
built about a stone core. It cvasprcscrved only to a low height and had been trunc;itetl. T h e
upper part of the stone was in fact found, still with solme clay adherin,? to it, in a nearby pile
of bul1dore1-debris. 'Chis pillar seems to laaye bccn of a diffcrer-it shapc from the otila:rs: tile
stone core was a columnar piece of rock rather than a slab, and the pillar itself v:;li; 01 aia
irri:gular shape.

Elgrur, 4 Thc rccon5tr ucted

origin'il ('1) nnii f~nnl (b)
torrn5 ot the pl,istercd clap
pillar in house K A B

In 1987 six human crania were found in thc northern half of this chamber. T h e had
been placed in the lowest fill, not o n the plaster of the floor itself.

Recurrent features

Several features of the buildings. their construction, fitting and use, are striking and merit
discussion. O n e is the activities surrounding the replacement of one building by another,
which embrace both the destruction and construction processes. A second is the design of
the interior. which can inform us about the uses which were thought appropriate.
All three phases of building and use were terminated by the obliteration of the old
building. It is notable that replacement was not effected by the simple expedient of finding
a new site for the new chamber. but apparently required that the old chamber be filled s o
that the new chamber could be constructed on very much the same spot. It is also notable
that the fill of the old chamber was not derived from debris from the destroyed
superstructure and roof (with the exception of the first house, R A E , where some possible
roof and superstructure debris was identified); nor were discarded chambers used as
convenient dumping places for midden. Of course, if the houses were completely
subterranean there would have been no superstructure t o remove; but at least in the cases
of houses R A D and R A A the fill was homogeneous. and there was n o trace of any material
which might have been a mud o r clay roof.
T h e implication is that the process of demolition was deliberate and expensive of effort,
involving the importation of many tons of soil. Because the new chamber was t o b e
constructed substantially in the same position, it also required that the fill of the old
chamber was sufficiently compact not t o cause the collapse of part of the new chamber's
wall. It also deserves mention that the lower part of the fill may contain objects not found

commonly elsewhere on the site and therefore presumably incorporated deiibernteiy. 111
house WAR a n~amberof large bones of large anirniils, which wzre otheris,isr .,.cry
uncommon, werc recorded close lo ( h i t not on) thc floor of ttir: chamber; there 1,ver.eal?;cj
several bonc pins and x~ccdles,uncommon ~ l s c \ ~ ~ ~ lIir rc lior~sc
r r ~ . Rr\A the only slviic i?c:acis
Eound in the 198'7 seasoti occurred in ihc lowest part of the fill vf the (charnbcr. A 1thtrugC1
t.here was n o ohscrvablc difference iri the gc:ner-a1 mata:ri;:i n:,ed throughout !.he trli. i t
~toulclappear that the beginning o f the filling of an abandoncil charnber was the <>ccasloi!
for depositing ccrtain npecial things. Thin conciunior~is supported by the obsenxtion t h a ~
at the heginning of the filling process i n houses RAA and RAE3 a xiragle slab oE s?a>nc: e v ; ~ s
propped upright with other, smaller stones in the soa~thernhalf oC the chamber.
None of the houses was abandoned and rcplaced l?cc;rusc of structural we;rki~esso r
decay so far as could be determined. Indecd, i t was riotiibic: il-iat thc houses wcrc ncii only
carefully made but were also carefully rn;iintainccl. Each of the three houses exhibited
repeated patching o r replacing of the plaster and even the prirnary clay coatirrgg. iiiou5c
MAA in particular showed at least three :lub5t;intial sub-phases, each of which involvcd
re-modelling of the northern end of the house. T w o of these are illustratcct in Figu1.e 3 . Sn
gcl~era!,,thcn. there was a constant attention to details of' maintenance and a fonclncss for
re-modelling.
Another token of this constant attention was t h e cleanliness of all three chambers. Great
care was taken in the excavation to isolate the last few centirnetres of material above a
plaster floor in order to pick up any occupation debris which nlight be evidence for the
activities which went on within a chamber. In no case was it possible to distinguish an
occupation deposit from t h e general fill of a chamber, and there was n o record of artefact!,
found on the floor (as opposed to within the fill). In addition to the lack of debris related to
domestic occupation there was a general absence of t h e kind of equipment associated with
normal don-restic activities such as food-processing. T h e n-rortars, pestles and \,arious
rubbers and grinders which are found elsewhere on the site \yere totally absent from the
houscs; in the last of the thrcc houscs, R A A , \yhich was the only o n e to have been found
with its floors intact, there was a small, central hearth, but that amounted only to a
scorched area of plaster.
While domestic equipment may have been extraordinarily sparse, there was o n e type or
fixture which was present in all thrcc of t h e houses. Each house had at least o n e pla5tered
clay pillar somewhere in the middle of its floor. Although the pillars in lrouscs R A A and
R A D had been truncated by recent surface damage to the site, t h e fallen pillar in house
R A E gives some idea of the original imposing appearance of these fittings. T h e pillar in
R A B was at least 1 . 4 0 ~ t1all, discounting whatever was modelled in clay on its top. This
pillar tells us that such fixtures were not structural. and it is hard to avoid the conclusion
that their function was not connected with the execution of the everyday house-keeping
activities of the occupants. 'l'he pair of pillars and the single standing stone in house RAD
show us how these fittings werc intended to h e grouped: their importance must have been
considerable for they stood in the centre of the chamher in a row. which did nothing to
improve circulation o r facilitate the utilisation of space,
Finally. the deposition of human crania in the lavt building in thc series when it was
abandoned and destroyed is another act expre5sing an attitude towards the ]louse which
goes far heyond that rieccssary at the simple ~atiiitariai~
level of removing ;in un~viinledholc

The origit1.s of house and home:) 343

in the ground. The crania await reconstruction and detailed study; as recovered fro111the
soil matrix of the house's fill they were too fragile to handle. The six crania represent
individuals of various ages (one at least is a juvenile). None has a mandible and many of the
maxillary teeth are also missing, having fiillcn out post mortcm. All that can be said at
present is that they must have been exposed for a considerable time before they were
brought to their final resting-place, and that final resting-place was in the primary
deposition of fill on the floor of the last house to be occupied. How that house-site relates
to other houses and thc settlement as a whole is unclear at present, but will be the subject
of further investigation in a future season. It is hard to avoid wondcring whose were these
skulls which were thought appropriate to take up occupation of a house whose living
inhabitants were finally leaving it.

Discussion
At present there is little if any epi-palaeolithic material from northern Mesopotamia with
which the earliest neolithic of Qermez Dere can be compared. Qermez Dere shows that a
final epi-palaeolithic is locally present, and the earliest stratum at Der Hall, a multi-period
tell-site on the Tigris northwest of Mosul which was sountled as a salvage operation by a
Japanese expedition, was laid down earlier in the epi-palaeolithic (Ohnuma and
Matasumoto 1989). It is only in the Levant that extensive excavation has taken place on
open settlements of the advanced epi-palaeolithic period. Comparison over such a great
distance can be justified only because no geographically close conlparison is possible and
also because it may be argued on the evidence of Qerrnez Dere and Tell Der Hall that the
cpi-palaeolithic of north Iraq was at the same cultural stage of development as that in the
Levant.
The Natufian culture of the south Levant and its congeners exhibit a general
architectural similarity. I-louses were substantial constructions which involved the digging
of shallow semi-subterranean floor areas and their enclosure with circular or oval walls
(Bar-Yosef 1981: 401). Although 1il-r-~eplaster \vas already employed in the epi-.
~~alaeolithic,
Natufian houses were not finished throughout with a plaster surface as were
the houses at Qermez Dere. More significantly, they were not used i n the same way. It is a
regular characteristic of Natufian houses to have plenty of food-processing equipment
around a prominent, central hearth (cf. Eynan ('Ain Mallaha): Perrot 19h6), and it seems
also that their floors were allowed to accun~ulatca great deal of occupational debris of all
kinds. Edwards (forthcoming) has analysed tens of thousands of pieces of chipped stone
from the floor area of one house at Wadi Hammeh 27, and Valla (forthcoming) is able to
undertake a quantified locational analysis of the different classes of tinds from one house at
Eynan.
The construction of sytnbolic fittings and wall decorations and the general architectural
elaboration of the house are features which made Catal Hiiyiik in central Anatolia unique
in the 1960s (Mellaart 1967; Hodder 1988), but which are now becoming common-place in
the later acerarnic neolithic from the south of Jordan (e.g.. Basta: Nissen, Muhcisen and
Gebel, forthcoming; and 'Ain Ghazal: Rollefson and Simmons 1988; Banning and Byrd
1987) to south-eastern Anatolia (e.g.. hievali Cori: l-lauptmann, forihcoming; or Caydnii:

&dogan and ozdogan. forthcoming; Schirmer, this volume), hut their discovery at
Qerlnez Dere pushes back the daie oitheir lnceprion to a point where tilt contrast in rasc oC
buildings between the cpi-palaeolithic and the ncolitiiic bi:conies stark. Aurcnche (1380)
has also drawn attention to the al.chiicctural elaboration at Tell Mureybet ear!?;.in th4:
aceramic neolithic (noting. incicientail!;, the ;ilmost tola! abssnce of tools a n d ecluipmcni
isom the building). Similarly the revcrrnt and careful treainnent of tlctached skuiis and
their placing in houses is so~net'rsingknown from many sitcs in the Levantine aceran~ic
neolithic, but now Q e r ~ n e zDere and the Khiamian scttlen-icnt of Netiv Hagducl ir! aha
.Tordarr Valley (Rar-Yosef. forthcoming: Bar-Yosef, Gopher and Goring-hforris 1380)
show that the tradition goes back to the \.cry beginning of that period. It should bs clear iy
now that the authos takes the view that a building housing s).mbolic elemcnrs is noi
necessarily a shrine. and that 21 house can be consecrated to everyday doialestic I-rurposcs
\+ithout leaving convenient archaeological tr.;ices. The casc for ihc Qermez Xkre houses
re~nainsambiguous; of course., until further excavation shotvs whether these structures are
the domestic norm or the exception.
The architectural and associated changes which occur at the beginning of the earl)
aceranric neolithic at Qcrrnez Dere. them: may be sum~narisedas being concerned with the
re-definitior~of thc role of the house, involvink: the exi~lusionoi' cert;rim food-processing
activities and an emphasis on labour-intensive carc and cleanliness. 'The housc was noit thci
appropriate place for sy~nbolicconstn~ctions.ancl ultimately ior the reception o i thc sktllls
o i the long tleacl. I n short the house: fornnerly the shclier for general everycjay activ:l:ics,
was bcing perceived as something more than utilitarian,, as the hon~i:. the private and
concrete expression of a particular family group. Studies of contemporary sedentary
h~intcr-gatherershave shown that they differ considcrabl\; horn the ;.~nobilchilnlcr-gatherer bands ~vhichhave become familiar to prehistoric i~rchacologists;variotrs writers
(Testart 19XZa. 19X2b: Woodburn 1980. 1982; Binford 1980) have stressed their intcreat irs
investment, or dela\ied return. and their clear (and to us -cry rccogt-riaabie) perceptions oi'
property and territory. Bender (1978) discusses hosv changes in attitude io\ii:irds
production can be of significance among hunter-gatherer- groups. In a rcccnt study h,lnrray
(1086) has shown how such studies of recent groups can be applied to ihe epi-palaeoIithic
groups of the Levant. but it i s suggested here that the implications of the transition from
mobile. fluici: hunter-gatherer bands to permanent, seikntary village cornmunitics which
took place in the cpi-palaeolithic period were bcing realiscd and articulated only latc. ir, the
ninth millennium be at the beginning of the neolithic period.
Another aspect of the recognition of propcrty and territory is the potential L'or conflicts
of interest. I t is perhaps not a coincidence, therefore, that thc beginning of 111saceralnic
neolithic throughout the Lcvant, and now in north Mesopo~amiri.is the time \vhcn t l ~ c
project~lcpoint series begins with the Khiasnian point ;IS its tirst rnanii'estation. Since t11i:r-e
does not serrn to be a changr in the species of animals exploited or in the strategies of
exploitation, it is necessary to seek an explanation for the projectile points in some cuItur.;li
or. social rather than economic change. That the change may be secn in the ernergeiii.6: oi'
formalised inter-group conllict is suggested aiso by the t1ii;covery of sever;~lskeletons
among those buried in the seitlerilent at Nemrili with projcciiie point5 cr-.l~bcddei.l
in [!!ern
(Kozlom:ski. pers. cornrn.: sec also p. T.1.9, t11is volunrc) and the. massivc conslrtr6:tion of
cs-Suil;ln: arrcic~~t
Jericho, early in thc 1:igheli milieniiinrn b i ,
dcfcnces ad

The originr of h o ~ l . ~11n(1

e home?

345

However, the main interest of this essay 11asbeen the novel architecture of Qermez Dere
and in particular the possible interpretation of the contrast between these buildings and
their immediate epi-palaeolithic predecessors of the Levant. There is no dramatic change
of plan such as Flannery (1972) sought to interpret in contrasting circular ancl rectangular
structures, which also forms an important clivision in Aurenche's classification (Aurenche
1981). Here, following on the pioneer exainple of Barbara Bender (1978), who exhorted
archaeologists to view the hunter-gathererlfarmer transition in 'a social perspective', the
enquiry has started from a desire to use architectural form and detail of the buildings in
order to discern aspects of social and cultural behaviour (cf. Banning and Byrd.
forthcoming). When the Cauvins examined the question of the origins of agriculture
(Cauvin and Cauvin 1983), they argued that no technological, biological or demographic
explanation on its own was enough. They concluded that a complex, multi-factorial cause
should be sought. whose key elements wo~alclbe found in social and cultural change. This
essay has attempted to suggest that a social and cultural change bcgun in the
epi-palaeolithic was consolidatecl at the very beginning of neolithic times and finds its
expression in significant changes in domes1.i~architecture, the treatment of the dead and
the development of armaments.

Department of Archaeology
University of Edinburgh

References
A number of the references cited below originatcti as papers at the collocluium PrChistoirc du Levant
2, organiscd by Olivicr Aurenche and Jacques and h4aric-Claire Cauvin at the Maison de I'Oricnt in
h4ay 1988. PrChistoire du Levant 2 is to be published as two issues of the journal Paleorient.

Aurenche, 0.1980. LJn cxa~nplede l'architecturc domesticlue en Syric au VIIIe millenairc: la

maison XLVII de Tell Murcybct. In L e ~MoyrnEJupl~rrite:z o n e de cnntact et d'gchnnges (ed. J .
MarguCron). UnivcrsitC des Sciences Hu~nainesde Strasbourg. Travaux du Centre dc Recherche sur
le Proche-Orient ct la GrCce Antiques, no. 5.
Aurenche, 0.1981. L a Maison Orientclle: l'architecture d u F'roclze-Orient Ancien d e ~origitzes au
rrzilieu d u qucltrierne millknclire. Paris: Gcuthncr.
Banning. E. B . and Ryrd. R . F. 1987. Housing and the changing residential unit: dornestic
architecture at PPNR 'Ain Ghazal. Jordan. Proceedings o f t h e Prehi~toricSociety 53: 309-25.
Banning, E. R . and Byrd. . B . F. Forthcoming. The architectural variability at 'Ain Ghazal in
relationship to other sites. Palkorient.
Bar-Yosef, 0.
1981. The cpi-palaeolithic complcxcs in the southern Levant. In Prehistoire d u Levant
(eds J . Cauvin and P . Sanlaville). Paris: Editions du CNRS.
B a r - Y o ~ c t0.
. Forthcom~ng.PPNA ~ i t inc the
~ Jordan Valley. Puleorient
Rar-Yoscf. 0 . . Gopher, A . and Goring-h4orris. N . 1980. Nctiv Hagdud: a Sultanian mound in the
Lower Jordan Valley. Paleorient 6: 201-6
Render. B . 1978. Gatherer-hunter to farmcr: a social perspective. World Archaeology 10
(2) : 204-22.
Binford. L. R . 1980. Willow smoke and dogs' tails: hunter-.gatherer settlcmcni systems and
archeological site information. Anlericcln Anticjrrity 45: 4-20.

346

Trevor W(~tkiul.5

Cauvin. J. and Cauvin. M.-C. 1983. Origincs dc l'agriculturc au Levant: facteurs biologiqi~cscr
socio-culturcls. In The Hillj' Flanks anti Reyonci. (cds T . C:. Young, P. E . L. Smith and
P. Mortensen). Chicago: Thc Oriental Institute.

g
settlement In Wadi al-tlammeh. Paleorrenr.

Edhxards. P. F o ~ t h c o r n ~ nNatufian
Flanncry. K. V. 1972. The origin of the village as a settlc~ncnttype in hfesoamerica and thc Near

East: a comparative study. In Man, Settlernerzr rtr~tiUrOctrzisr~~

(cds P . Ucko. R . Tringharn and

G . Dimbleby). London: Duckworth.

Hauptmann. H . Forthcoming. PJeval~Tori. Paleorient.

Hoddcr. 1. 1988. Contextual archaeology: a n interpretation of catal Hiiyiik and a ciiscussion of thc

origins of agriculture. Bulletin (lithe lnstirltre ofArchaeology (Univcrsity of London) 24: -13-56.
Kozlow~ki.S . K . Forthcoming. Ncmrik 9. Paleorienr
Mcllaart. J . 1967. Ccctal Fliiyiik,

ci

neolithic toivn in Anatolicc. London: Thames 61- Hudson

Murray. N . 1086 Settlement and sedentism in the late cpi-palacolithic and early neolithic of tllc
Levant. Unpublished M A dissertation. Department of Archaeology, University of Edinburgh.
Nissen, H . J . . Muheisen. M. and Gebel. H . Forthcoming. Report on the first two seasons of
excavation at Basta (19861987). Palkorient.
Ohnuma. K. and Matsumoto. K . 1088. Lithic artifacts from Lcccl 6 of Tell Der I3all. Eski-Mosul
(Iraq): a preliminary report. Al-RriJician0: 73-89.
ozdogan, M. and Ozdogan, A . Forthcoming. qayiini~:a conspectus of recent work. Pale'of.ic.izr
Pcrrot, J . 1966. Le gisement natouficn dc Mallaha (Eynan), Israel. L'Anrhropologie 70: 157-81
Rollcfson. G . and Simmons, A . 1088. 'Ain Ghazal. In The Prehi.i.toi:yoJ'.loidan (eds A . Garrard anti
H . Gebel). Oxford: British Archaeological Reports, International Scrics.
ou l'origine cie.7 inkga1ijC.v. Paris: Sociktk d3Anthropologic.
Tcstart. A . 1982a. I,c.s cha.sseurs-c~ii1l~ur.s
'Tcstart, A . I982b. Thc significance of t'oocl storagc among huntcr-gathcrcrs: residence patterns.

population dcnsitics and social inequalities. Clcrrent Anllrropology 23: 523-37.

Valla, 1;. Forthcoming. Aspt:cts de sol de I'abri 131 dc Mallaha (Eynan). PalCoricnr

Watkins, T. and Haird. D . J . I087. QEYITZEZ

T ) ~ T E 1087. Univcrsity of Edinburgh, Department of

Archaeology, Project Paper no. 6 .

Watkins, T . , Baird. D. J . and Hctts, A . V. G . Forthcoming. Qermer Dere and the ncolithic o f TG.

Iraq. Pale'orient.

Watkins. T., Baird, D . J . . Bctts, A . V. C i , and Nesbitt. M. Forthcoming. Excavations at O c r r l - ~ e ~

Dcrc. Tcl Afar: an early aceramic ncolithic site in K.Iraq. Sltrnrr.
Woodburn. J. 1080. Hunters and gatherers today and reconstruction of the past. In S o ~ ~ i cititi
ej

M/e.sterrl Anthropology (cd. E. Gellner). London: Duckworth.

Woodburn. 3 . 1982. Egalitarian societies. .'Clan 17: 431-45.

The ongrnr of lzoure (in(/home?

347

Abstract

The origins of house and home?

Excavations in course at the site of Qcrmcz Dcrc in north Iraq have rcveaicd a stratified sccluence
dating from the end of the cpi-palacolithic period into the earliest centuries of the carly neolithic. A
sequence of subtcrrancan houses belonging in the latter part of that timc-span (825C7900 bc) is
remarkable for the elaborate care and cffort spent on their construction. maintcnancc and
de~nolitionireplacement.E ach house was kept ccry clean, and thcrc is a distinct lack of debris and
even the equipment of everyday life. Each housi was ccluippcd with one or more non-structural clay
pillars. At the end of the series six wcathcrcd hutnan crania were placed on the floor of the last house
as it was being obliterated. Thcsc very carly neolithic houses arc contrasted with those of the
preceding epi-palacolithic of the Levant. and it is suggested that they represent an important change
in the perception of the house as home and the focus of attitudes and activities of symbolic social
significance concerned with the conservation of property and the continuity and solidarity of the
owning family.