the intricacy of neolithic rubble layers. the ba‘ja, basta ... · the intricacy of neolithic...

Neo-Lithics1/09

IntricacyofNeolithicRubbleLayers

33RubbleSlidesandRapidClimateChange

General Discussion

RubblelayersareacommonfeatureatmanyseventhmillenniumBCsiteslocatedonslopesintheJordanianHighlands. Three of these sites are discussed here:At ‘AinRahub, rubble layerswereobserved in1985in a deep trench transecting a slope; in Basta, theybecame subject of discussions in 1987, where theyform substantial feature of the Post-LPPNB slopestratigraphy;andinBa‘ja,theywerediscoveredquiteunexpectedly during the fifth season of excavations(2003) in the flattest area of the site which, andremarkably, here have no catchment for such hugeaccumulations.At all three sites, the origin of theseaccumulations of angular, fist-sized stones were anenigmaticissue,andindeedpartlystillare.

Significantly,rubblelayershavetheuniquepotentialto serve as an empirical source for furthering ourcomprehensionofabandonmentprocesses,subsistence

shifts,andclimaticchangeintheEasternMediterraneanduring the seventh millennium BC. Therefore, theserubblelayersmightevenprovideevidencefortheroleof natural causes in the decline ofLPPNB traditionsand lifeways (at around 6900 BC), the adoption ofpastoralismduringtheFPPNB/PPNC(firsthalfoftheseventhmillenniumBC),andthepotentialimpactsonagro-habitats in the PNA (Yarmoukian/ second halfof the seventh millennium)1. Although many localparametersareresponsiblefortheaccumulationofanindividualrubblelayer/slide(cf.below),theintenseandwidespreadappearanceofrubbledepositsbytheendoftheLPPNBandinthePotteryNeolithicmustberelatedtotheinfluenceofacommonagentwhichactuatedandcoordinatedvarious localparametersand ingredients:Periodsofheavierrainfallsand/ortopography-relatedflashfloodsandaquaticslopeerosionwouldappeartobethemainfactoroftheseaccumulationsorslides.Inthelightofevidenceforclimatechangeinthelateseventhmillennium BC (8200 calBP “Hudson Bay” event),inearliercontributionsrubblelayers/slideshavebeendiscussed solely in relation to the contemporaneousYarmoukian (Weninger et al. 2005, 2007). Recentconsiderations, however, seem to acknowledge thatrubble layers and slides are a much wider Neolithicphenomenon (Rollefson, this issue; Weninger, thisissue).Indeed,BastaandBa‘jahavelongattestedtoatleastthreerubblelayerepisodesduringtheLPPNBtoPNwhichonlypartlyandlocallyappearintheshapeofslides.

The stratigraphic, structural, and chronologicalintricacy of the Jordanian Neolithic rubble layersand slides warns against a mono-causal explanation.NotoneoftherubbledepositsfoundintheNeolithiccontexts discussed here is similar in terms ofarchaeological morpho-phenomenology and chrono-stratigraphy; rather, they represent locally restrictedand quite dominant depositions of fist-sized angularstone rubble, the origins of which appear quitepuzzlingatfirstglance.Asyet,theappearanceofsuchaccumulationsinnon-archaeological,i.e.naturalEarlyHolocene stratigraphies has remained uninvestigated.Although representing events highly influenced byindirect or direct anthropogenic influence, rubblelayers and slides could be an excellent chance toidentify impacts of Rapid Climatic Change (RCC;Weninger, this issue;Weninger et al. 2009) or otherenvironmental impacts which triggered the physicaldisplacementof rubble.However,undertaking rubblelayer/slide research is a very slippery terrain if itlacks consistent interdisciplinary approaches, and if

The Intricacy of Neolithic Rubble Layers. The Ba‘ja, Basta, and ‘Ain Rahub Evidence

Hans Georg K. Gebel InstituteofNearEasternArchaeology,FreeUniversityofBerlin [email protected]

Fig. 1 LocationsofLPPNBandPNsitesmentionedinthetext


Neo-Lithics1/0934

RubbleSlidesandRapidClimateChange

explanationsaredominatedbyabsolutechronological,pheno-stratigrapical, pedological (cf. Lucke inKafafiandLucke,thisissue)orgeomorphologicalarguments.Especially themere focus on supra-regional climaticchangemayleadusinwrongdirections,asmonogeneticexplanations may do in general. For example, adrainage regimemight have sorted and accumulatedrubblewithoutamajormoistphaseintheclimate,andjustbenefitedfromlocalcopiousslopehydrology.

Thispapersummarisesthearchaeologicalfeaturesoftherubblelayers;indepthgeomorphologicalstudiesmustfollow,thuspavingthewayforinterdisciplinaryresearch designed to approach one of the mostspectacularfeaturesoftheNearEasternNeolithic:thediscontinuities in settlement history and subsistencemodes during the seventh millennium BC in theSouthernLevant,andtheirrelationtorubbledepositsandpotentialclimatic/seismicimpacts.

The identification of the various interactinglocal parameters, causes, and forces thatmight havecontributed to the formation of any given rubbledepositisaprerequisiteforanydiscussionoftheroleplayed by RCC, or any other factor, in rubble layeraccumulation. Indeed, we have to accept that suchcomplex phenomena will not just provide evidencefor one singular cause: imagine, for example, theearthquakewhichtriggerstheflowofcolluvialrubble,heavily soaked by regional RCC rains, taking upfield stone clearing piles and cultural deposits on itsway,beforereachingitsfinalplaceofdeposition.Thedemand is that prior to an analysis of its conditionsand characteristics, a rubble layer per se should notbe takenas a signal for anything; in this respect, thefollowingfactors requirecarefulconsiderationbeforeanexplanationisoffered:

1. prevailingpalaeo-drainageregimesandpalaeo-topography

2. anthropogenicbarriersandimpacts(e.g. intra-site architectural barriers such as buildingterraces, agricultural barriers like valleyterraces, size-sorting and stone extraction bymanetc.)

3. evidenceofseismicimpacts4. origin of rubble components (e.g. natural vs.

anthropogenic, e.g. wall stone dressing, floorandwallcomponents,etc.)

5. intra-sitediversityofrubblewithinflow/depositand its sorting (in terms of its sedimentologyanddepositmorphology)

6. identification, chronology, and morphologyof rubble layers/slides in the Early Holocenelandscapessurroundingthesites

Additionally, it should be stressed that the samefactors(e.g.fluvial)thatmayhaveledtothedepositionof rubble layersmayalsohave caused their negativeevidence, i.e. their removal from the stratigraphicsequence.Arubbleslidedepositisonlyasnapshotofasite’ssedimentaryenvironment,andincreasedfluvial

surfaceenergycanalsomanifestitselfinthecompleteorpartialerosionoflayers,includingrubbleslides.

Bethisasitmay,therubblelayers/slidespreservedin settlements dating to the seventh millennium BCdemandexplanation,especiallysincetheiroccurrenceoftenappearsrelatedtodisruptionsinthehistory(intra-siteandregionally)ofoccupation.Indeed,theyrepresentawiderphenomenonin theEasternMediterraneaninthe seventh millennium BC (and in other, younger,time frames, too) at all placeswith an extantgravityregime (slope setting of sites). Indeed, many of oursitesdohavethislandslidepotential.Inafurtherstep,systematicsurveysneedtoinvestigatepotentialrubblelayers on pre-LPPNB and Late Pleistocene sites aswellasbelowandabovePost-Yarmoukianhabitations.Additionally,geomorphologicalsurveysneedtoclarifywhether rubble layers only occur in archaeologicalcontexts or have corresponding formations in thelandscape.

Itisverymuchacommonfeatureofthedebrisandmuddeposits/flows–orrubblelayers/slides–thattheyappearfartooextensiveforthesizeofthecatchmentfrom thematerial is thought tostem.From theBastaandBa‘jasitesitisclearthattheangularstonesoftherubblelayersmustderiveinalargepartfromflakingof the (dressed) wall stones, the fills of the double-faced walls, and (in Basta) the floor constructions.This means that the architectural debris from thesesitesproducedmostoftherubblefoundintherubblelayers.ForYarmoukian‘AinRahubsuchsourceshaveto remainunderdiscussion since the test trenchonlyrevealed in general (by the mudbrick fragments) anarchitecturalcontextoftherubblelayers,whileinsituarchitecturewasn’tcaught.

In our previous publications (e.g. Gebel 2004b,2006) we carefully spoke of rubble layers or rubbledeposits; the term rubble slide was promoted byWeninger et al. (2007) in their discussion of theYarmoukianlandslides.However,recent(spring2010)on-sitediscussionsinBa‘jaandBastawithChristophZielhofer (Leipzig University, geomorphology) andBernhardWeninger(CologneUniversity)regardingthediversityofrubblelayersledtotheconclusionthatweshould ratheruse themoreneutral term rubble eventinsteadofrubbleslide,sinceonlysomeoftherubblelayersshowmoraine-likefeatures.

Preliminary Definitions of Neolithic Rubble Layers and Rubble Slides

ThefollowingdefinitionsarebasedonobservationsofNeolithicrubblelayersatoursites(‘AinRahub,Basta,andBa‘ja),andareboundtotheoccurrenceof+/-fist-sizedangularstones:

A rubble layer consists of +/- fist-size angularstones,generally–butnotnecessarily–embeddedinafinermatrix; thismatrixmaycontainmaterial fromre-depositedcultural layers (charcoal, ash, smallflintartefacts and plaster fragments, etc.); occasionally

Neo-Lithics1/09



rolled/rounded +/- fist-sized stones occur amongthe angular stones. These stone accumulations canbe thin or they form thick and extensive horizontallayers following the inclination of an old surface orrepresenting restricted lenses or piles. Componentsof rubble layersmay not share a general orientation(althoughtheyoftendo),andthematerialcanevenbeofpurely anthropogenicorigin,e.g., from the typicalLPPNBdouble-facedwallsofwhichthedressedwallstonesweresortedoutandtheangularfist-sizedoftheinteriorwallfillremained.Ontheotherhand,rubblecanalsostemfrompurelynaturalsources,e.g.weatheredbedrock from the slope above a settlement. On theupperpartsofslopes,rubblelayershavethetendencytobemoreshallowandlinear,increasinginthicknessand taking on fan-like in-sediment morphologies inlowerlyingparts.Theyalsoevenedoutsurfacesbye.g.fillingsmallsurfacerunnels.Intheirmigrationontothesurfaceofasitetheyareoftenguidedbywallremainsstill exposedon theslopesurface. It shouldbenotedthat fist-sized rubble scatters on old surfaces are notdeemedrubblelayers.

Rubble slides are fluvially deposited rubblelayers, or a sequence of fluvially deposited rubblelayers,whichmaycontaininsituoccupationaltraces,ephemeral or solid installations (walls, burials,chipping floors, surfaces etc.; cf. Gebel et al. 1992).Sequencesoffluviallydepositedrubbleslidesmayalsocontainorbeinterruptedbylensesandlayersofotherwater-laid material (e.g. fine gravels) and/or aeoliansediments. Intra-site rubble slides potentially occurin all locationswhere a drainage or drainage regimeforcestheformation,movement,anddepositionoffist-sizedrubble.

Although the fist-sized stone rubble can containnatural colluvial material at some sites, it normallycomprises (re-deposited) cultural layers andarchitectural rubble; in-site rubblecontextsare rarelyfound sterile of artefacts. Rubble slides normallyassemble in their sedimentary environmentmaterialsfromanysourceslocatedhigheruptheslope,i.e.fromsettlementandfield/gardencontextsthatwereinhabitedor influencedbyhumansduringtheirdeposition.Ourdefinitionof rubbleslides includes thatsuchdepositsnotonlyareattestedonslopesurfaces,butalsofilleddrainages where they can appear – in case of laterincisions–inthesections.

Seventh Millennium BC Rubble Slide Evidence East of the Rift Valley

The preliminary list of Neolithic sites with rubblelayerseastoftheJordan/WadiArabaRiftValleyis(inalphabeticalorder,cf.alsoFig.1):‘AinGhazal,‘AinRahub,AbuSuwwan,AbuThuwwab,Ba‘ja,al-Baseet,Basta,Ghwair,andWadiShu‘aib.Potentialcandidatesfor the rubble layer discussion are ‘Ain Jammam,Beidha(Fig.5),KhirbetHammam,al-Shalaf,es-Sifiya,andUmmMeshratIandII(referencesformostofthese

Fig. 2 es-Sifiya:Sectionofsquarewithpossiblerubblelayer (photocourtesyofH.Mahasneh)

Fig. 3 Ghwair:Sectionofanupperslopeexcavationareawithrubble layersconsistingofrockdetritusandfine-grained(aeolian) material.ViewfromW(photobyauthorwithpermissionofM. Najjar)

Fig. 4 Ghwair:Sectionofanuppermostslopeexcavationareawith arubblelayersconsistingexclusivelyofrockdetritusinits lowerparts,joinedbyfine-grained(aeolian?)materialinits upperparts.ViewfromW(photobyauthorwithpermissionof M.Najjar)


Neo-Lithics1/0936


sitesaregivenbyRollefson,thisissue).OmryBarzilai,this issue,providedageneral reporton rubble layersfrommanyareaswestof theRiftValley,also for theNatufian - PPNA and Chalcolithic periods; he alsomentionsadditionalanthropogenicsourcesofangularstonematerialwhichwedonothaveinthethreesitesdiscussedhere.

As mentioned, we should be aware that themorphologies and phenomenologies of rubble layersand slides, featuring angular fist-sized stones andfound more or less compacted in lenses or layersaboveandinNeolithicruins,arenotallthesame.Thenatureofrubblelayersdependsonthecatchmentareafromwhichmaterials are taken up and re-deposited.For example, the purely natural “rubble layers” ontheupperslopeatthesiteofGhwair(Figs.3-4)havea very limited source and catchment: Here a desert-varnish bearing outcrop weathered its “thermal”detritusintotheLPPNBarchitectureintheshapeofarubblelayer,fullycoveredit,andisstillaccumulatingtoday.Theproximityofthissourceof“rubble”tothearchitectureagainstwhichithasaccumulatedhasmadeitalmostimpossibleforothertypesofrubble(e.g.re-depositingculturaldebris)tocontributetothe“rubblelayers” observed in this section. This situation maybe different further downslope where rubble layersare also expected to contain the fist-sized angularstones from the settlement (e.g. Fig. 3). InGhwair’suppermost slope, however, “rubble layers” are ratherthe result of aeolian/dune accumulations and a highshareofbedrockweatheringproducts(Fig.4)fromtheextreme differences between the daily temperatures’maximaandminima.

‘AinJammamisanexamplefor rubble layersnotnecessarilyaggregatingintheupperpartsoftheslope:Duetothesteepnessoftheslope,erosiontransportedallmaterialdownwards, including rock falls, culturaldebris and fist-sized stones, until a stable surfacedeveloped in which the ruined L/FPPNB and PNA

wall topsrest.Here, rubble layerswith theirshareofanthropogenicmaterialhavetobeexpectedinthemoreshallowmiddlepartsoftheslope,andareattestedquiteclearlyatthelowerfringesofthesite.

The date (or dates) and stratigraphic position ofthe horizontal gravel/ rubble layers resting betweenBeidha’s MPPNB architecture and the sandstoneformationtothenorth(Fig.5)needtobecomesubjectoffutureinvestigations;theycanrepresentatleastpartlyseventhmillenniumBCfills,butitisunlikelythatthecatchments reaching the area by the small gorges inthenorthernsandstone formationhavenotconstantlydelivered material onto the spot by the millennia.Nearby Siq al-Barid at least shows considerabledepositssinceNabateantimes.

For the (Post-) LPPNB es-Sifiya and al-Baseetslopes depositional conditions similar to Basta areexpectedwithrespecttotheirrubblelayers:Whilesuchwereobservedinasectionintheyear2000atal-Baseet,suchobservationsfores-Sifiyaneedtobeverified.

One further issue should be addressed here:Recent considerations by Zeidan Kafafi (cf. also thecontribution in this issue) tentatively claim that ameteorite impact in theEasternJordaniandesertmayhavecausedregionalclimaticchangeandmud/rubbleflows affecting seventh millennium BC settlementsin Jordan.This notion, however, has so far not beensubstantiatedbyanysoliddata,andshouldbeexcludedforthetimebeingfromtherubblelayerdiscussion.

Inthefollowing,therubblelayerdatafromthethreesitesdiscussedherearesummarised.

The Basta Evidence

WhenthefirstevidenceofrubblelayersatBastawerediscussed with Hans Joachim Pachur, Ulrich Kamp,andMarkusNüsserby thesectionexposures in1987therewasmuchconjecture.However,evenatthistime,manyoftheideasexpressedalreadyhintedtowardsverycomplexprocesses,includingthetemporalexistenceofagriculturalfieldsandfieldclearingpilesonthePost-LPPNB Basta slopes. In addition, the rubble layerswere discussedwith an evenmore intriguing featureofBasta’s sedimentary environment inmind, the silt

Fig. 5 Beidha:Sectionwithhorizontallyembeddedgravel/rubble layersofunknowndate,locatedbetweenBeidha’sMPPNB architectureandthesandstoneformationsborderingthesitein theN.ViewfromSE

Fig. 6 Basta:ReconstructedLPPNBdrainagepatternonBasta’s present-dayslopes.ViewfromSSE(graphbyU.Kamp, publishedinGebel2004:Fig.1)

Neo-Lithics1/09



Fig. 7 Basta,SquareB70,Loci2-4:SurfaceofUpperRubbleLayers (URL).Notesub-topsoilFine-GrainedDeposits(FGD)inthe sections.ViewfromS

Fig. 8 Basta,SquareB68,CentralRoom/Courtyard1ofBuilding1 withroomsadjacenttotheNE:SectionwithUpperRubble Layers(URL)coveringthetopoftheLPPNBruin. ViewfromSW

Fig. 9 Basta,SquareB85,Room/Space3(foreground)ofBuilding VII:NotetherubbleflowoftheLowerRubbleLayers(LRL)in frontofwalls(Locus7,16,and8)andpassingthroughthewall opening.ViewfromE

Fig. 10 Basta,SquareB68:PartlyremovedUpperRubbleLayers(URL; cf.sectionswithURL)atthejunctionwiththeroomfills. UppermosttopsofLPPNBwallruins.ViewfromE

Fig. 11 Basta,SquareB83:TopofruinedLPPNBwall(Locus16) exposedunderneathandintheLowerRubbleLayers(LRL), locatedatthesameheightastheflimsyFPPNB/PPNCwall remains(Locus10),totheleft,intheLRL.Notetheinclination ofthemudflowstotheE(downslope).ViewfromS

Fig. 12 Basta,SquaresB86-87,SSection:SequenceofFine-Grained Deposits(FGD),UpperandLowerRubbleLayers(URL-LRL) abovethetopofruinedLPPNBwalls.Notethestone accumulationsdepositedaftertheURLformation,possibly representinganoldlandsurface(pavement)andtheremainsof fieldclearingpiles.ViewfromN


Neo-Lithics1/0938


depositsofAreaC(Kamp2004,Gebel2006).Onlyinthe1992campaign(Gebeletal.2004)thepost-LPPNBrubblelayersatBastareceivedmoredevotedattention(on account of a planned deep sounding). FlimsyPPNC-relatedoccupations(Fig.15)wereobservedinthelowestpartsoftheLowerRubbleLayer,andforthefirsttimeLowerandUpperRubbleLayers(LRL,URL;Figs.7-14)2weredistinguished; thesewereseparatedfrequentlyfromoneanotherbydeposits/layerswithahigherratiooffine-grainedsediment.Inthecampaignspriorto1992archaeologicalrubblelayerobservationswere hardly carried out, and ironically they werereferred to as the chebabian period, since the quickremovalof these thickdeposits required ahigh levelofman-power.TherubblelayersfindingatBastamightbecomparablewiththoseatWadiShu‘aib(Simmonsetal.2001;Rollefson,thisissue:Fig.5)wheretwosucheventsseemtoseparatethePPNBfromthePN.

TheLowerRubbleLayers(LRL,Table1)ofBastaArea B contain PPNC artefacts, curvilinear wallets(Fig. 15), chipping floor dumps and Tridacna beadworkshop remains,fireplaces, samagah installations/surfaces, stone robbing pits dug into the LPPNBarchitecture (related hoard of stone figurines, cf.HermanseninGebeletal.2004:94,101-102,Figs.15-16),apre-Yarmoukianarrowhead type,andveryfew

intrusive? sherds showing a relation to Yarmoukianpottery, etc. (Gebel et al. 2004: Table 1); of course,they also contained re-depositedF/LPPNBmaterials.TheLRLmusthavestartedtoaccumulateshortlyaftertheabandonmentofthesettlement,sincethewallsofstructureswerestillstandingtallandtherubblelayersmigrated into the ruin,evenpenetrating throughwallopenings(e.g.Fig.9).TheUpperRubbleLayers(URL,Table 1) contain all sorts of re-deposited materials,includingPalaeolithic to F/LPPNB andPN artefacts,re-depositedrubbleoftheLRL;insitufireplacesandsurfaces are less well preserved (compared with theLRL findings), and partially in situ finds of a PNA/YarmoukianchippedstoneindustryaswellasisolatedNeolithicpotterysherdswerefound(Gebeletal.2004:Table1).AsFigs.8-11 indicate, the ruinedwall topsof the LPPNB basements were still poking out ofthe surfaces at considerable heights during the URLdepositions.Thisissomewhatpuzzling,sinceitwouldmean–intermsofourcurrentabsolutechronologicalunderstandingof the rubble layers atBasta (Table1)–thatsomeruinedwalltopsoftheLPPNBbasementswerestillvisibleaftersome700years.Wouldn’t thisfindingnotindicatethattheURLofBastaaresomewhatolder,e.g.datingaroundthemidofthemillenniumBC?

The Lower and Upper Rubble Layers of Basta

cal BC Basta Periods Area A Area B Area C

Fine-Grained Deposits (FGD)? Fine-Grained Deposits (FGD) ?

predominantly

6000

Upper Rubble Layers (URL)/ Upper Rubble Layers (URL)

downslope sedimentation of with remains of clearing piles, debris

PNA cultural debris and mud flows

aeolian

Lower Rubble Layers (LRL) Architectural Phase B0 / Lower

PPNC- downslope sedimentation of Rubble Layers (LRL): curvilinear wall

related cultural debris fragments, fire places and surfaces

embedded in debris and mud flows

6900 sedimentation

FPPNB Architectural Phase AI: Architectural Phase BI: room fills, (silts)

rectangular rooms build of burial ground, rectangular and

undressed cobbles curvilinear rooms build of small slabs,

substructures C208 and

huge accumulations of C217:

workshop refuse (naviform Architectural Phases BII-III village fringe

chipped stones) room fills, large multi-roomed activities

LPPNB and rectangular architecture, (gardens,

Architectural Phases AII-III: substructures burials,

room fills, large multi-roomed chipping floors),

and rectangular architecture, no buildings?

substructures, “trash burials” Architectural Phases BIV ?

7500 in-room and in-channal burials

bedrock bedrock (C217)

Table 1 Basta:Chrono-stratigraphicalsummaryofthePost-LPPNB(afterGebeletal.2006:Table1).

Neo-Lithics1/09



seemnottorepresenttwomajorisolateddepositionalevents, rather they appear as two sets of locallyrestricted depositions. It seems that larger parts ofthe Basta slopes are covered by rubble layers, withonly smaller areas having escaped this depositions.Layer thicknesses are varied and have the tendencyto increase downslope; in general, the Basta rubblelayers tend to form restricted extensions, like largelenses,accumulationsonoldsurfaces,andevenpiles.ThethicknessoftheuppermostFine-GrainedDeposits(FGD;ChristophZielhofer:“Kolluvium”;Figs.12-14)increasesconsiderablydownslope.

Beforewediscuss thevarious scenariosof rubblelayer formation at Basta, the only radiocarbon datethat exists from the context of the rubble layers atthe site (from the earliestLRL, should be presented:KIA30847 (Basta47244)dates the remainsofafireplace (Square B83: Locus 8) contemporaneous withthe deposition of the rubble to cal BC 6749, 6721,6702 calBC (radiocarbon age: BP 7911 ± 56; P.M.Grootes, Leibniz Labor für Altersbestimmung, Kiel,pers.comm.)(Fig.17).Thedatereflectsperfectlythearchaeological PPNC-related evidencewe have fromtheLowerRubbleLayersinBasta(Gebeletal.2004,Gebel2006,Gebeletal.2006).

The understanding of the palaeo-topographicalslope settings are essential for the understanding ofBasta’srubblelayers(cf.Kamp2004:Figs.1-3;Gebel2004a:Fig.1C,2004b:Fig.1):ThetopographicalunitAreaA(Fig.6)representstheNEpartsoftheNeolithicvillageontheslopesbetweenasmallgully(apresent-dayvillagestreet)intheSWandthebedrockoutcropswith quartzite veins to theNE (Kamp2004: Fig. 1).ThelowerpartsoftheslopesareverysteepandborderthebottomofWadiBasta.TheupperpartsofAreaAareratherflatandpassoverintotheflattopographyoftheformerfieldsinAreaC.AreaB(Fig.6)islocatedinthecentral,steeperandspur-likepartoftheNeolithic

Fig. 13 Basta,SquaresB104-105,NSection:SequenceofFine-Grained Deposits(FGD),UpperandLowerRubbleLayers(URL-LRL) above(andbetween:LRL)thetopofruinedLPPNBwalls.Note thestoneaccumulationsdepositedafterURLformation, possiblyrepresentingtheremainsoffieldclearingpiles. ViewfromS

Fig. 14 Basta,SquaresB103-105,SSection:SequenceofFine-Grained Deposits(FGD),UpperandLowerRubbleLayers(URL-LRL) abovethetopofruinedLPPNBwalls.Notetheinclinationof therubbleflowstotheE.ViewfromWNW

Fig. 15 Basta,SquareB83,Locus5:FlimsycurvilinearPPNC/FPPNB wallfragmentembeddedintheLowerRubbleLayers(LRL). ViewfromS

Fig. 16 Basta,SquareA12,NWSection(bulldozercut):Hugesilty rubble/gravellayers(Loci61a-goftheNWSection),covering thetopoftheruinedwesternwall(Locus2)ofRoom16and adjacentareas.Theorientationoftherubble/gravelmixedwith someculturalmaterialsisorienteddownslope.Possibly representsthefillsofaseventhmillenniumBCrunnel.View fromE


Neo-Lithics1/0940


village.It isalsolocatedonWadiBasta’sNWslopesbetweentheaforementionedsmallgully(thepresent-day village street) in theNE and theflat slope areasin theWandSW.In theSE it reaches thebottomofWadiBastabyasteepinclination.IntheNWitmeetsaflatareawhichbelongstopographicallytoAreaC.Theoriginalspur-liketopographyofAreaBseemstobetheresultoftwoEarlyHolocenedrainagesintoWadiBastafromtheNW(Kamp2004).

The post-LPPNB sedimentary stratigraphies ofBastaarea sequenceofdepositional, re-depositional,and extraction events which modified the reliefover the seventh millennium BC. While the naturalimpactsonthesedimentaryenvironmentsoftheslopesat Basta were reduced or controlled by F/LPPNB

humanoccupationduringthesecondhalfoftheeigthmillennium BC, natural causes and materials againgainedtheupperhandduringtheseventhmillenniumBC, i.e. following thecloseofpermanentoccupationatthesite.DuringtheF/LPPNBoccupationsatBastaacombinationofdomesticbehaviourontheonehand,andnatural alterations on the other (e.g. by drainageregimes, colluvial materials, heavy rain/snowfall,extreme temperature maxima/minima and otherclimatic parameters) impacted upon this particularlandscape. Against this background, we have toexpect (Kamp 2004,Gebel 2006,Gebel et al. 2006)the existence of protective structural measures, suchas (terrace and barrier) walls and ditches, designedto offer some protection against both aquatic slope

Fig. 17 CalibrateddatefromYarmoukian-relatedrubblelayersin‘AinRahub(top)andfromtheLowerRubbleLayersinBasta(bottom). GraphpreparedbyB.Weninger

Fig. 18 Ba‘ja:TopographyandidentifiedlocationsofRubbleFlows/Fine-GrainedGravelLenses(RF/FGL)andFine-GrainedLayerscharacteristicforthe sub-topsoillayers(FGM),incontrasttothepresent-daysurfacedrainageregimeofthesite

Neo-Lithics1/09



erosion and colluvial accumulation inLPPNBBasta;indeed,naturalslopedrainagerepresentedapermanentthreattoF/LPPNBvillagesontheslopesintheregion.Intra-site rainfall and snow management, debrisflow management, slope pressure management: Allthese factors should be reflected in the architecture,the architectural planning, and the stratigraphies atBasta.However,itisofnotethatthebestevidenceforprotectivewalls,barriers,andditchesshouldbefoundonthefringesofthesettlements.

In the lowest LRL at Basta, dated to the PPNC,rubble deposits were apparently used by squatters,stillresiding,manufacturingbeadsandchippingflintsbetweentheerodingruinsofF/LPPNBwalls,etc.Thedecaying F/LPPNB structures produced thick, ratherhomogeneous, and consolidated fine accumulationsand patches of disintegrating plaster and roof/ceilingmaterials, a characteristic of the LRL. In the upperLRL, fireplaceswere still operated, and other in situtracesofhumanactivitiesareinevidence;infact,theselayers witness someYarmoukian-related features. Incontrast, theURLshowfar fewerhabitational traces;rather, they display locally restricted sequences ofdownslopesedimentation,possiblyinterruptedby(re-deposited) remains of field clearing piles. The URLmay have become deposited around 6200BC, if notearlier(cf.above).

The understanding of the huge silty rubble/gravellayers inAreaA (NW Section, Loci 61a-g; Fig. 16;Gebel2006;69,Fig.2.A),reachingthicknessesof2-3m and covering the ruined LPPNBwall tops is stillpremature.Theydonot containmuch cultural debrisat this spot, as opposed the section layers to theNEand the NE Section (Gebel 2004b). Their formationmusthaveinvolvedsiltymaterialsofAreaC:itappearsthat they represent the fill of a seventh millennium

drainage/runnelinatthisspot(cf.therunnel’ssection/inclinationsinFig.2AofGebel2006).

Asidefromtheanthropogenicrubbleofthevillage,physicalweatheringproducts(angularrockdetritusfromblocksizetosand/silt)andaeoliandepositionwereallimportantcomponentstohavecontributedtothemassofmaterialthatdevelopedinthecatchmentsofBastaandpenetratedintothesettlementarea(Kamp2004).Today, the areawitnesses torrential rainfall episodes,andwehaveeveryreasontoassumethatthiswasalsothe case in the seventh millennium BC; therefore,wemust expect such events to still be visible in oursquaresandsections,too.Theoriginandimportantroleofaeoliansiltinthesedimentaryenvironmentsofthesiteisstillpoorlyknown(e.g.ashareofmorethan30%wasfoundinAreaC,cf.Kamp2004):duststormsmaybetheoriginofthesesiltswhichaccumulatedfor2-3mduringandaftertheLPPNBinAreaC,whereevenindividualaeolianeventscouldbetraced.Thesitewassubject to aeolian erosion in the seventhmillenniumBC,too(cf.alsoabovetheAreaAsiltevidence),butwedohaveyetaclueontheaeolianmaterials’shareinArea’sBrubblelayers(Kamp2004).InAreaBtheaeoliancomponentsseemtobeoflesserimportance.

The Ba‘ja Evidence

Under the heading: Huge Rubble and Fine GravelFlows, Wall Rubble and Air Hollows we openeddiscussionsfocusingontheextraordinaryevidenceforhigh-energyeventstohaveoccurredatBa‘jaandwhichwerefollowedbyLPPNBarchitecturalreoccupationinAreasCandB-Southofthesite(GebelandKinzel2007):Hugerubbledepositsandotherfeaturescharacteristicof earthquake destruction were noted (Fig. 18). Inadditiontothis,the–fluvialorseismic/fluvialrelated– destruction of the eastern part ofArea C by slopesubsidence is also attested (Gebel and Bienert et al.1997:Fig.6),thoughitisstillunclearastotheprecisenature of the accountable high-energy event. Theearlierearthquake inAreaB-South(Figs.21-22)wasfollowedbythickdepositionsofstonerubble(RF,upto1.5minheight;Figs.22-23)2withembeddedwater-

Fig. 19 Ba‘ja,AreaB-South:EvidenceofRubble/GravelFlowswith Fine-GravelLenses(RF/FGL)exposedinSquareB64-South andtumbledwalls(earthquakedamage?)inSquaresB84-85

Fig. 20 Ba‘ja,AreaB-South,SquaresB84-85:Tumbled LPPNBwalls(earthquakedamage?)


Neo-Lithics1/0942


deposited fine gravel accumulations (FGL) that restagainstthetallstandingwall(Locus4)inSquaresB64-SouthandB74(Fig.19)orwerefoundunderthelaterarchitecturalre-occupation(Locus5)inC-10/10(Fig.23);thewater-depositedfine-graveldeposits/lensesareastrong indicationofanaquaticslopeerosionwhichtookupandsortedfloor/ceilingcomponents.Severalspots provide indications for some deconstructionprior to the start of the latest architectural phase.Remarkably, there exists no catchment for a naturalsourceoftheseRF/FGLmaterialsatBa‘ja:Therefore,it must be concluded that they are of anthropogenicorigin(contrarytoanassumptioninGebelandKinzel2007, cf. below).Thewall rubble resulting from theearliest earthquake (and fromsubsequent instabilitiesofhousewalls)wereburiedbythesecomplexrubbleand fine-gravel sequences in SquaresB64-South andB74andC-20/20.Afurtherearthquakeappears tobeattestedbythetwistedwallsinupperB84-85(Fig.20).Earlier considerations (Gebel and Kinzel 2007) thattheRF/FGLflowsresultfromflashfloodsreachingthecentralupperpartsofthesettlementfromthegorge(Siqal-Ba‘ja),andthatthefloorofthesiqwasmuchhigherthan today, require revision following new insightsgained from recent fieldwork at the site (ChristophZielhofer,pers.comm;spring2010).

In the following we present and discuss theindividualpiecesofevidencefortherubblelayersand

relatedhigh-energyevents.Foramoredetailedaccountofthesefindings,seeGebelandKinzel2007.

AreaB-South(Figs.18-22):Theexcavationinthesouthern half of Square B64 has provided insightsintohugeintra-siterubbleandgravelflows(RF/FGL)restingagainst theaforementionedhighwallLocus4inB64-SouthandB74(Figs.19,22)andoverthewalls(Loci13,29,25-26),andthewallrubbleaccumulationswithairpockets(Loci16,21,and24);similarfeaturesare reported fromAreaC–cf. below–at adistanceofsome20-30m).Thewall rubble–sometimesstilldeposited in a fallen-domino arrangement – with airpocketswasfoundtobemixedwithahigheramountofloose, re-depositedmaterial, includingmortar/plaster/ceiling debris, containing charcoal, and appeared tohavebeen,atleastpartially,intentionallyburied.Wall13 seems to have been reduced in height, probablyduring the erection of the upper phase ofWall 4 (=coarse-facedupperpartofWall4).OntopofWall13rests themoraine-typeflowoffist-sized stone rubblewithembeddedfinegravellenses(RF/FGL)thatisalsoattestedintheeastsectionsofB64andB74andreachesheights of 1.5m. (Figs. 19, 22). In B74, fine graveldepositsmigrated inside the “gate” ofwall Locus 4,whichwasblockedduringtheRF/FGLevents.WithintheseRF/FGLdeposits, fireplaces and surfaces exist,proving that deposition happened in short episodeswhile the inhabitants were using the (temporary)surfaces.Thewholeaccumulation,however, is ratherhomogeneous, contains aside the angular rubbleoccasionallyfist-sizedlimestonegravel,andgivestheimpressionoffastdepositioninasrestrictedtime.Thethirdhigh-energyeventinAreaB-SouthisrepresentedbythetwistedwallsinupperB83andB84:Theenergytowhichthewallsweresubjectedcausingthemtoleaninalldirections,andthereforenotabidingtoaspecificvector or pattern; this latter feature also leads us toconcludethatthisresultedfromanearthquake.

Fig. 21 Ba‘ja,AreaB-South,SquareB64-South,Loci21and 23:LPPNBwallrubblefoundwithairpockets(earthquake damage?)

Fig. 22 Ba‘ja,AreaB-South,SquareB64-South:Excavatedearliest architecturalremains(LPPNB)withpartlyremovedwallrubble loci(earthquakelociwithairpockets,cf.Fig.17)anddeposits ofRubble/GravelFlowsandFine-GravelLenses(RF/FGL) above.ViewfromS

Neo-Lithics1/09



AreaC,SquareC-10,BaulksC-20/20andC-10/10(Figs. 18, 23):Here, a stairwell inC-20/20 connectsthetwoolderoccupationallevelsinC-10/10/-20/20/21,andalateroccupation/buildingphaserestsonthefist-sized stone rubble flow with embedded fine gravellenses (RF/FGL). Similar to Area B-South, westernArea C witnessed three major impact events: anextensive earlierwall rubble pilewith air pockets inC20 (incompletely excavated) in a rather large openspace, a huge rubble and gravel flow resting againsthigh standing walls, and the reorganization of spaceand architecture during an upper architectural phase.During the latter, also the impressive stairwell inC21 (BienertandGebel2004:Pl.5)musthavebeenerected. For the first time, we were able to isolatelocallyadistinctlaterarchitecturalphaseinBa‘jafroman earlier occupation which represents a disruptionof thesite’sarchitecturalmorphodynamiccomplexityof succeeding modifications that normally preventthe identification of clear sub-phases. Together withbuttressLocus114ofC10andWall6ofC-10,thisE-Wrunningwall Locus 5 denotes the latest architecturalphase inwesternAreaC (Fig. 23). It is erectedon aRF/FGLrubbleflowwithlayersofsmallfluvialsortedand deposited gravel (8-15mm); this is also the casefor buttress Locus 114, Wall 6, and buttress Locus26.Thesewater-laidfinegravelsarealsofoundinthe

northsectionofC20,wheretheyaccumulatedagainsttheE face ofWall 10 (formerBaulkC-20/20).Herethese fine gravels appear as lenses and layers insidethe upper parts of a rubble flow, consisting of fist-sizedstones.Alltheaforementionedwallremainsandlayerswerecoveredbythelightbrownishfine-grainedmaterial (FGM) formingalso the sub-topsoil layer inallAreaB;itsthicknessesreaches60cm.TheRF/FGLrubble/ gravel seems to have terminated the earlierarchitecturaloccupationinwesternAreaC,causingthereorganizationof its space.ThepartialdestructionofthisphaseappearstobeevidencedbythedepositionofthehugewallrubbleintheopenspaceofC20andinthespacebetweentheWalls120inC20and5,26and8inC-10(wheremanylintelstoneswerealsofound).Theorientationsofthiswallrubblearevarious;thedepositsfeature a high frequency of air hollows, revealing arapidandprobablyintentionalfillingofthespace.Itisassumedthatthisactionrelatestothedeconstructionofwallsintheareafollowinganearthquake.ThismustalsohavetwistedthecompletestairwellLocus129inC20,simultaneously leaning it down by the height of onestep:Theearthquake,thesubsequentdeconstructionofarchitecture, and intramural filling of the large spaceinArea C20 preceded the migration of rubble/smallgravelflows(RF/FGL)intothearea.WaterappearstobetheagentoftransportandmovementoftheRF/FGLbefore the walls of the latest occupation in westernAreaCwereerected.

Seismicity has so far been a rather neglected topicwhendiscussingNeolithicrubbleslidesortheinterruption/abandonmentofsettlements.Ifwetakeasameasurethefrequencyofearthquakeeventstohaveaffectedtheareain the last 2000 years, we can assume that every 200yearsamedium-majorearthquakeshouldbeexpected;forexample,in551A.D.Petrawasalmosttotallyabandonedafter an earthquake destroyed its buildings, andAqabawas twice destroyed in 363 and 1068A.D. (Migowskietal.2004,KorjenkovandSchmidt2009).TheLPPNBmega-sitesarelocatedalongtheDeadSeaRifttectonics,andwerethereforealsovulnerabletodestructionbysuchcatastrophic events; however, and quite remarkably, ourdiscussion of the descent of themega-site phenomenonhasuntilnowfailedtoconsidertheroleofseismicityinthe relatedprocesses.SinceLPPNBbuildingunitsweremostly erected upon terraces or built on or into slopes,any leaning walls were simply explained away as theresult of slope pressures, e.g. as was the case with thelongwall inBa‘ja’sAreaD;also,pronouncedcracks inwalls/pillars,e.g. inBastaB68: 18,were also subjectedtothisinterpretation.Certainly,andwithoutadoubt,thisagglomeration of evidence calls for increased in-depthresearch intoseismicityand its impactonourmega-sitearchitectures.

In conclusion, the clearest evidence for LPPNBearthquakes affectingLPPNB sites stems fromBa‘jaAreaB-SouthandwesternAreaC.Here,itshouldalsobe noted thatArea B-South lies between a southernsandstone outcrop and the northern sandstone ridge(Fig. 18), which are at a distance of some 15-20 m

Fig. 23 Ba‘ja,SquaresC10andC20:SequenceoftumbledLPPNB stairwell,openspacewithLPPNBwallrubbleshowingair pockets(earthquakedamage),Rubble/GravelFlowsandFine- GravelLenses(RF/FGL),latestLPPNBarchitecturalphase remains,Fine-GrainedLayers(FGM)deposits.ViewfromS


Neo-Lithics1/0944


(BienertandGebel2004:planbetweenPages122and123),andthelowermostarchitectureintheseareasisprobablyindirectcontactwiththeunderlyingbedrock.Itfollowsthatduringanearthquake,shockwaveswouldhavebeentransferredheredirectlyintothewallsofthestructures. To summarise, the following earthquakefeaturesareattestedinBa‘ja:-wallspushedbyperpendicularwalls(tiltingwallsinvariousdirections)-wallrubbleinfallen-dominoarrangements;airpocketsintheirrubble-lateraldeflectionandwallsplitting

Blocked/inserted doorways/wall openings andwallreinforcementsbyaddingparallelwalls(e.g. theblockingandclosurewallofthe“gate”inB74)couldverywellbesecondaryearthquakeevidence,meaningtheresultofspacereorganizationafteranearthquake.TheremightthenbeachanceinAreaB-Southtofindtheskeletalremainsofearthquakevictims.

The ‘Ain Rahub Evidence

At the Late Epipalaeolithic/Early Pottery Neolithicsite of ‘AinRahub (13 kmNWof Irbid and 4,5 kmNNEofSal;GebelandMuheisen1985)Yarmoukianfindswereencounteredinrubblelayerssealedwithinthestratigraphyofa terracespur; the terraceremainsbelong to the lowest terrace inWadiRahub (Hannss’T1 Terrace, cf. Muheisen et al. 1988: 475ff.); thegeomorphological setting of ‘Ain Rahub (420 ma.s.l.) was studied by Christian Hannss in 1985 bystereoscopic analysis of aerial photographs. In thefollowingyears,muchof the topographyof thearea,includingthevicinityofthespring,wasalteredbystreetbuilding,bulldozing,androckblastingfromthenearbylicensed excavation of graves (Siegfried Mittmann,pers.comm.),finallyhinderingfurtherexcavations.

Physiographically, the location represents aterrace spur (Fig. 24: dotted area) between Wadi

Fig. 24 ‘AinRahub(sitearea dotted):Aerialview.

Neo-Lithics1/09



Rahub and a tributary drainage. Its stratigraphycomprisesalluvial,colluvialandculturallayers.In1981, a final Natufian settlement (extending ontothe spur)was exposedduringbulldozingactivitiesat the eastern foot of the spur.A test unit of 3x1m cutting into the slope, carried out by ReinderNeef,wasoriginallyintendedtoreachtheNatufianlayers in order to determine the overall extensionof the site in the spur’s slope.Toour surprise, theTest Unit (Fig. 25-26) revealed concentrations ofmud brick debris, grinding tools, andYarmoukianpottery (Kafafi1989)atdepthsbetween59.70and58.80 m (excavation-internal height) (some evenoccuring at depths of 58.20m; cf. Fig. 25)whichare partly embedded in the rubble layers between59.90 and 58.90 m. At that time, these findsrepresented the second Yarmoukian site east ofthe JordanRiver, and still the nature of the site isnotclearasallof its layersweresealedwithinthe

spur.The rubble layers resulted frompossibly twoimmediately succeeding events and were formedbydenselypackedstones(Figs.25-26),apparentlyrepresentingmudflowswhichtookupYarmoukianculturalmaterialsincludingYarmoukianpotteryandbrick fragments on their way to deposition. FinalNatufian (12000-10200calBC) findsoccurredc. 1mbelowthelowermostrubblelayer,concentratedatdepthsaround57.70m.

Christian Hannss commented (Muheisen etal. 1988: 479) that the „sediments and limestonedebris“oftherubblelayers„mostlikelyarenotofdirect colluvial origin but were deposited as wadiaccumulations.Major colluvial deposits cannot beexpectedhere,astherearenoextensiveslopesabovethe lower terrace of ‘AinRahub.“While the goodpreservation of theYarmoukian sherds contradictsthe interpretation of wadi accumulations, Hannss’understanding thatnodirectcolluvialoriginof therubblelayersshouldbeassumedappearsplausible.Mostlikely,the‘AinRahubevidencerepresentsoneor twointenserubbleslidesmovingonto theFinalNatufian/Post-FinalNatufianslopesurfacesfromtheimmediate slopes to thenorth.Here aYarmoukiansettlementmusthaveexisted,thematerialofwhichbecameacomponentoftherubbleslides.

TheYarmoukianrubbleslideat ‘AinRahub isdatedbyoneQuercussp.charcoaldate(GrN-14539:7480+/-90BP;W.G.Mook,CentrumvoorIsotopenOnderzoek,GroningenandR.Neef,pers.comm.1987).This14C-age(Fig.17), equivalent toa calibratedageof6490 -6170calBC(95%), is ingoodagreementwithotherdatesforthe Yarmoukian (Weninger, this issue of Neo-Lithics).However, whether this date represents the date of therubbleslideitself(e.g.remainsofafireplaceduringthedeposition of the rubble), or not simply the (potentiallyearlier)dateoftransportedcharcoalfromtheYarmoukiansettlement,remainstobediscerned.Thisinterpretationalproblem applies tomany of the available 14C-ages forthe“Yarmoukian”rubbleslides,andcan–atthepresentstateofresearch–onlyberesolvedbyapplicationofdirect(exposure)datingmethods,e.g.OSLand10Be,orbytheradiocarbondatingofwell-observedinsitufeaturesfromwithinarubblelayersequence.

Rubble Layer Archives: Research Perspectives

TheintricacyofseventhmillenniumBCrubblelayersatNeolithicsitesinJordanresultsfromthepolygeneticand polycausal elements that were involved in theirformation. This should not make us ignoring theirpotential as an important source of information onclimatic change. By this, we do not mean that theorigin of rubble (be it anthropogenic or natural) isirrelevanttodiscussions,butwedosuggestthatevenlocally transported anthropogenic rubble may reflecta changed or changing climate regime. Due to thecomplexityof rubble layers, future analysisdemandsa multidisciplinary (e.g. prehistory, geomorphology,

Fig. 25 ‘AinRahub,TestUnit,NWSection.(sectionwidth:1,00m)


Neo-Lithics1/0946


pedology, radiocarbon and other dating methods)approach, and any interpretationbasedupona singlegenesis must be disregarded. Future rubble layerresearchmustnotonlyconcentrateoncollectingmorerubble slide archives from more sites in differentphysiographic locations and from different EarlyHolocene contexts, but rubble layers also need tobe properly excavated and observed in terms ofmicrostratigraphy, depositional events, and in situfeatures.Theexcavationofrubblelayersatprehistoricsitesmusttaketheformofajointinfieldcooperationbetween archaeologists and geomorphologists. Itfollows that the analysis of rubble layers withoutdirect correlation with the surrounding naturalsedimentary environments amounts to an incompleteandfragmentaryenterprise.

Rubble layers occurring during and immediatelyafter prehistoric occupations are a most importantsourceforimprovingourcomprehensionofaregion’ssettlement history. Further, they not only provide uswithanexplicitunderstandingoftheoccupationalfateof a given site, but they also help to identify (intra-site) areaswithin a site thatwere deserted during anotherwisepermanentpresenceofhumansat thesamelocation.Inthesite’snaturalenvironmenttheyareoneof the important sources of information on land use,vulnerability of biotic resources, and anthropogeniceco-impacts.

Themainproblemlieswiththeabsolutedatingofrubble layers.Raisedawareness isneeded to identifypotentiallyundisturbedinsitutracesofoccupationandsurfacesinthedepositionalsuccessionofrubblelayers;indeed,thistaskshouldnotposetoogreataproblem.Otherwise,thedatingofrubblelayersissubjecttothehigh risk of datingmuch older re-depositedmaterialtaken up from transformed cultural phases furtherupslope.

Rubblelayerawarenessisrequiredinallrespects.

Acknowledgements:IwouldliketothankBernhard“Bernie”Weninger for initiating the recent rubbleslide and subsequent RCC discussion, and for re-awakeningourconcernofthesephenomenawhich,sinceour1987seasoninBasta,wesimplyreferredtoasrubblelayers.B.Weningerinitiatedaninfieldprojecton the rubbleslidesof ‘AinGhazal,Basta,andBa‘ja(financedbytheSonderforschungsbereich806atCologneUniversity),startingsectionsamplingworkat‘AinGhazalin2009.InSpring2010,andincollaborationwithus,additionalgeomorphologicalsamples were taken by Christoph Zielhofer inBa‘ja andBasta, aswell as 3DLaser Scanning ofthe rubble layers and related architectures by D.Hoffmeister, Cologne University. Since 1987, Ihavediscussedthephenomenologyandmorphologyof our rubble layerswithmany colleagues:UlrichKamp, Hans-Joachim Pachur, Christian Hannss,Bo Dahl Hermansen, Hans J. Nissen, HenningFahlbusch, and others, but only the recent rubbleslide/RCCdebatebrought a new impetus into thediscussionofthishithertomysteriousphenomenon.Sincethe1980s,GaryRollefsonandZeidanKafafidiscussed internally the rubble layers for theirsites, and Gary Rollefson was among the firstwho understood the potential and implications ofthis topic. Recently, discussions were joined byChristoph Zielhofer, Leipzig University, throughhisinfieldgeomorphologicalandgeoarchaeologicalcooperation in Spring 2010, which are awaitingsupplementation by his laboratory results. Thephotos of Basta were taken by Margreth Nissenand Gerald Sperling; all others are by H.G.K.Gebel. Graphical assistancewas provided by JanaPokrandt, and languageeditingwascarriedoutbyLeeClare,CologneUniversity.IthankHamzehM.Mahasneh,MuhammadNajjar, andAlan Simmonsfor the permission to use the photos of Figures23-25.All our research on the Neolithic of SouthJordan would not be possible without the supportof theDepartmentofAntiquitiesof theHashemiteKingdom of Jordan whose constant support wegratefullyacknowledgehere.

Fig. 26 ‘AinRahub,TestUnit:Excavationofthelowerpartsofthe rubblelayers.ViewfromSE.

Neo-Lithics1/09



Notes1AllabsolutechronologyinthiscontributionreferstocalibratedradiocarbondatesBC.Thechronologicalabbreviationsusedhereandtheircurrentabsolutechronologicalequivalentsare:LPPNBLatePre-PotteryNeolithicB(c.7500-7000/6900BC)FPPNBFinalPre-PotteryNeolithicB(c.7000-6800?BC)PPNCPre-PotteryNeolithicC(c.6800?-6500?BC)PNA(Yarmoukian)PotteryNeolithicA(c.6500?-6200?BC)The contents of this contribution became the basis for infielddiscussionsinthefieldinthecourseofB.Weninger’sproject(cf.theAcknowledgements)inSpring2010.ThedelayedNeo-Lithics1/09issueallowedsomereferencetobemadetothisproject,butcouldnotconsiderfullyitsresults.

2 The following abbreviations were used for the characteristicstratigraphicunitsoftheBasta/Ba‘jasedimentaryenvironments.Sincetheoriginandcompositionofthesedimentaryfeaturesarenotexactlysimilar,foreachsitedifferentabbreviationsareused.FGM Fine-Grained Layers, characteristic for the sub-topsoillayers(Ba‘ja)RFRubble/Gravelflow(Ba‘ja)FGLFine-GravelLenses(Ba‘ja)URLUpperRubbleLayers(Basta)LRLLowerRubbleLayers(Basta)FGD Fine-Grained Deposits, characteristic for the sub-topsoil(Basta)

References

BienertH.D.andGebelH.G.K.2004 SummaryonBa‘ja1997,andInsightsFromLater Seasons.In:H.D.Bienert,H.G.K.GebelandR.Neef (eds.),CentralSettlementsinNeolithicJordan.Studies inEarlyNearEasternProduction,Subsistence,and Environment5:119-144.Berlin:exoriente.

GebelH.G.andMuheisenM.1985 Noteon‘AinRahub,aNewLateNatufianSiteNear Irbid,Jordan.Paléorient11.1:107-111.

GebelH.G.K.andBienertH.D.withcontributionsofKrämerT.,NeefR.,Müller-NeuhofB.,TimmJ.,andWrightK.I.1997 Ba‘jaHiddeninthePetraMountains.Preliminary Resultsofthe1997Investigations.In:H.G.K.Gebel, Z.KafafiandG.O.Rollefson(eds.),ThePrehistoryof Jordan,II.Perspectivesfrom1997.StudiesinEarly NearEasternProduction,Subsistence,and Environment4:221-262.Berlin:exoriente.

GebelH.G.K.2004a Present-daySiteSettingandPhysiographicUnits.In: H.J.Nissen,M.MuheisenandH.G.K.Gebel(eds.), BastaI.TheHumanEcology:35-52.Berlin:exoriente.2004b SitePreservationandSiteFormationProcesses.In: H.J.Nissen,M.MuheisenandH.G.K.Gebel(eds.), BastaI.TheHumanEcology:95-116.Berlin:exoriente.

GebelH.G.K.andHermansenB.Dahl2004 LPPNBBa‘ja2003.Neo-Lithics2/04:15-17.Berlin:ex oriente.

Gebel H.G.K., Muheisen M., Nissen H.J., and Qadi N. withcontributionsbyBeckerC.,HermansenB.Dahl,NeefR.,ShafiqR.M.andSchultzM.2004 LatePPNBBasta:Resultsof1992.In:H.D.Bienert, H.G.K.GebelandR.Neef(eds.),CentralSettlements inNeolithicJordan.StudiesinEarlyNearEastern Production,Subsistence,andEnvironment5:71-104. Berlin:exoriente.

GebelH.G.K.2006 TheStratigraphyandLocusData.In:H.G.K.Gebel, H.J.NissenandZ.Zaid(eds.),BastaII.TheArchitecture andStratigraphy:11-132.Berlin:exoriente.

GebelH.G.K.,KinzelM.,NissenH.J.,andZaidZ.2006 Summaryandconclusions.InH.G.K.Gebel, H.J.NissenandZ.Zaid(eds.),BastaII.The ArchitectureandStratigraphy:203-224.Berlin:ex oriente.

GebelH.G.K.andKinzelM.2007 Ba‘ja2007:CrawlSpaces,RichRoomDumps,and HighEnergyEvents.ResultsoftheSeventhSeasonof Excavations.Neo-Lithics1/07:24-33

KafafiZ.1989 LateNeolithic1Potteryfrom‘AinRahub,Jordan. ZeitschriftdesDeutschenPalaestina-Vereins105:1-17.

KampU.2004 GeomorphologicalSiteSettingandGeochemical Results.InH.J.Nissen,M.Muheisenand H.G.K.Gebel(eds.),BastaI.TheHumanEcology:53- 94.Berlin:exoriente.

KorjenkovA.M.andSchmidtK.2009 AnArchaeoseismologicalStudyatTallHujayrat al-Ghuzlan:SeismicDestructionandChalcolithic andEarlyBronzeAgestructures.InLutfiKhalil andKlausSchmidt(eds.),Prehistoric‘AqabaI:79-98. Orient-Archäologie23.Rahden,MarieLeidorf.

MigowskiC.,AgnonA.,BookmanR.,NegendankJ.,andSteinM.2004 RecurrencePatternofHoloceneEarthquakes AlongtheDeadSeaTransformRevealedbyVarve- CountingandRadiocarbonDatingofLacustrine Sediments.EarthandPlanetaryScienceLetters222: 301-14.

MuheisenM.,GebelH.G.,HannssC.,andNeefR.1988 ‘AinRahub,aNewFinalNatufianandYarmoukian SiteNearIrbid.InA.N.GarrardandH.G.Gebel(eds.), ThePrehistoryofJordan.TheStateofResearchin 1986.BritishArchaeologicalReports-International Series396.2:472-502.Oxford,B.A.R.

SimmonsA.H.,RollefsonG.O.,KafafiZ.,MandelR.D.,an-NaharM.,CooperJ.,Köhler-RollefsonI.,andRolerDurandK.2001 WadiShu‘eib,aLargeNeolithicCommunityinCentral Jordan:FinalReportofTestExcavations.Bulletinof theAmericanSchoolsofOrientalResearch321:1-39.


Neo-Lithics1/0948


WeningerB.,Alram-SternE.,BauerE.,ClareL.,DanzeglockeU.,JörisO.,KubatzkiC.,RollefsonG.,andTodorovaH.2005 NeolithisierungvonSüdosteuropaalsFolgedes abruptenKlimawandelsum8200calBP. InD.Gronenborn(ed.),Klimaveränderungund KulturwandelinneolithischenGesellschaften Mitteleuropas,6700-2200v.Chr.RGZMTagungen1: 75-117.

Weninger B.,Alram-Stern E., Bauer E., Clare L., DanzeglockeU.,JörisO.,KubatzkiC.,RollefsonG.,TodorovaH.,vanAndelT.2007 AbruptClimateForcingBbservedatEarlyNeolithic sitesinSouth-EastEuropeandtheNearEast. InH.Todorova,M.Stefanovich,G.Ivaniv(eds.), IntheStepsofJamesHarveyGaul,2.The Struma/StrymonRiverValleyinPrehistory. ProceedingsoftheInternationalSymposium ‘StrymonPraehistoricus’,Kjustendil-Blagoevgrad (Bulgaria)andSeres-Amphipolis(Greece),27.09- 01.10.2004,Sofia:7-28.

Weninger B., Clare L., Rohling E.J., Bar-Yosef O., Böhner U.,BudjaM.,BundschuhM.,FeurdeanA.,GebelH.G.K., JörisO.,Linstädter J., Mayewski P., Mühlenbruch T., Reingruber A.,RollefsonG.,SchyleD.,ThissenL.,TodorovaH.,andZielhoferC.2009 TheImpactofRapidClimateChangeonPrehistoric SocietiesDuringtheHoloceneintheEastern Mediterranean.DocumentaPraehistorica26:1-53.

Commentary

Neo-Lithics1/0949


Weninger’smasterfulcompilationofpaleoclimaticdataclearlyshowstheoccurrenceofanumberofepisodesof rapid climate change (RCC), and it is expectablethat suchphenomena shouldhavehadenvironmentalconsequences across the globe. One RCC coincidedwith a phenomenon (the rubble layers) on a numberof essentially contemporaneous archaeological sitesinJordan,anditwasbothtemptingandobligatorytodetermine if the correlation offered some aspects ofcauseandeffect.

ThecontributionbyBarzilaicovered thepresenceof “stone surfaces” in archaeological sites thatmighthaveoriginatedfromagenciesotherthannaturalones,and thiscontention iscertainlyacceptable.OnepointthatshouldbemadeisthatwhenwehavetalkedaboutrubblelayersintheLateNeolithic/Yarmoukianperiod,we were not concerned simply with pavements, butwithvastaccumulations.ThesituationatArdel-Samraappearstoconformtosuchmassiveaccruals,butitisnotclearfromBarzilai’sarticleifthemoundsofstoneswerebeneath,above,orinterspersedwithYarmoukianculturaldebris;inotherwords,couldthismovementofangulardebrisbeduetoflashfloodinganddepositionof materials from the wadis and gentle hill slopesdetectableinhisFig.1?

Barzilai’s description of anthropogenic sources ofangular rock debris covers conditions that are wellrecognized throughout the Levant. At ‘Ain Ghazal,forexample, thereare largeanddense lensesoffire-cracked rock (FCR) – which is almost always fire-crackedflint–duringtheMPPNBintheCentralField,butsuchoccurrencesareofaverydifferentcharacterfromthesituationintheYarmoukianperiod(aswellasintheLPPNBandPPNC).IntheYarmoukianlayers,the rubble is dense, deep, and virtually continuouslydistributedacrosstheentiresite,bothwithinbuildings(probablyabandonedbeforethedeposition)andinthebroadspacesbetweenthesparselybuilt-upYarmoukianvillagearea.In theMPPNB,FCRoccursdenselybutonlysporadicallyintightlydefinedclustersofdebris,andalwaysassociatedwithashydeposits.

ThesuggestionsthatFCRisassociatedwithcookingis probably not the case, or at least not a completeaccountingforthepresenceofthecrackledflint.WhilemanyhearthsincludeFCRinandaroundthem,thereareotherhearths(particularlythoseinsidetheMPPNBhouses) where FCR is absent or only intermittentlypresent.MostoftheFCRconcentrationsareinoutdoorlocations,sotheassociationofFCRislikelyconcernedwithsomeformofprocessingofmaterialsotherthanfood,butjustwhatprocessingremainselusive.

Gebel also considers the likelihood that not allrubbledepositsareduetoclimaticconditions,andthat

“primemovers”asexplanatorydevicesareveryoftensuspicious if not outright misleading and erroneous.EarthquakeevidenceatBa’jaisparticularlyimpressive,andmuchoftherubblethatendsupinrubblelayersmayowetheirultimateoriginsnottonaturalcauses,buttoanthropogenicpracticesaswell.Nevertheless,henotesthat usually there are indicators that water transportwasresponsibleatleastinparttotheaccumulations.

What is importantabout theevidence fromBasta,I think, is that the rubble layers occurred in layersequivalenttothefinalpre-ceramicperiod,thusantedatingaYarmoukianage1.Thisfollowsarefinementoftheso-called “8.6-8.0 k.y.a. event” to indicate that itwas aperiod of time that,while geologically speakingwasa“sudden”development,actuallyspannedarelativelylong time at its onset (see Weninger, this volume).Thisisalsoastrongpieceofevidencethatthe“rubbleevent” actually consisted of several climatic pulses,andthesepulseswerenotnecessarilycontemporaneousacross theNear East but instead varied according togeomorphic andgeographic elements affecting stormtracks. The suggestionGebelmakes, that the end ofthelargeLPPNBoccupationofBastabythebeginningof the 7th millennium BC, is also an excellent caseforarguingthattheLPPNBeverywherewasasmuchaffectedbyclimaticdeteriorationasbyculturalfactors(e.g.,RollefsonandPine2007),althoughsuchculturaldegradation certainly had a coeval impact of theenvironment.

Evenso,theeffectsonthelocalenvironmentattheendoftheLPPNB/FPPNB/PPNCatBastawereclearlymore powerful than in the north at ‘AinGhazal andWadiShu’eib.Thismight relate to thedifferences inannual precipitation: the area around ‘Ain Ghazalreceivesc.250mmrainfalleachyear,whilethemodernsituation atBasta is only160mm (Neef 2004: 188).ThePPNCoccupationat‘AinGhazalcontinued,albeitacrossamuchmorereducedareaofthesite(lessthanthree-fourths of theLPPNB site area and far belowthe density of residential structures and projectedpopulation levels)2. The population density declinedeven more at ‘Ain Ghazal during the Yarmoukianperiod,althoughtherewasstillasubstantialpopulation,perhapsasmuchas300-400people.

As was the case at Basta and Ba’ja, populationsexplodedduringtheearlierpartoftheLPPNB,andlikethe situation at the southern sites, there is a possible“sudden” impact on the site’s people. While thepopulation atMPPNB ‘Ain Ghazal was modest andspread across theZarqaRiver to the eastern bank toonlyamoderatedegree, thesuddeninfluxofLPPNBimmigrantsturnedtheEastFieldintoamajor“suburb”of the main site. But this eruption of settlement

Commentary

Gary O. Rollefson WhitmanCollege,WallaWalla [email protected]

Neo-Lithics1/09

Commentary


expansion may have been necessary; it is possiblethat the surge in population began to exhaust localresources, especially farmland and pasturage as wellaswoodresourcesforfuelfordomesticuse(RollefsonandPine2007).TheextensionofdomesticbuildingseastwardacrosstheZarqaRiverwaswell-established,butbefore7203±95cal.B.C.alargeritualstructurewas built, cutting intowhat appears to have been anessentially abandoned zone of ‘Ain Ghazal by thattime (Rollefson1998:51-54,but especiallyFootnote24).Theconstructionofthisbuilding,whichrequireda major communal effort) may reflect deterioratingclimaticconditionsalreadybeforethebeginningofthePPNCperiod,aswasseenatBasta.

The situation that Gebel describes concerningearthquakesasapossiblecontributortorubbledepositsbeforethePotteryNeolithicperiodmightalsohaveaparallelat‘AinGhazal,althoughevidenceremainsweakatthemoment.ThefinalstageofthecircularLPPNB“shrine”intheNorthFieldappearstohavesufferedanarchitectural disruption that included severe damageand partial disintegration of the floor; a replacementof thecircularbuildingwasrapidlyundertakenabout5mtothesouth,butthereplacementappearstohavebeen used for a brief time (Rollefson 1998: 47-48).Thefloordamage in theearlierbuildingsuggests thepossibility of earthquake damage, although unrelatedslope subsidence instead can’t be dismissed at themoment.Anotherbitofevidence thatmight relate toearthquake damage at ‘AinGhazal contemporaneouswith the situation at Ba’ja comes from a two-storybuildinginthesouthFieldthatdatestotheLPPNB.Inthiscase,thesectionexposedbybulldozerworkshowsan upper painted plaster floor that collapsed into theconfinesofalowerroom.SuchacollapsewasseenintheNorthFieldat‘AinGhazal,butthiswascertainlyduetoafirethatburnedroofsupports(RollefsonandKafafi1996:13-14)andhadlittleevidentrelationshiptoseismicactivity.

The contribution by Kafafi, Lucke, and Bäumlerleavesonesomewhatnonplussed.Muchofthearticleaddresses architecture both prior to and within theperiodunderconsideration(the“8.6-8.0k.y.a.event”).Twostandinggeological/archaeologicalsectionsweresampled (the eastern South Field and the westernCentralField).Considerableeffortismadetodescribethe composition and development of five very largeandundefinedarchaeological layers (rarely identifiedastoarchaeologicalage),butnoneofwhichdealwiththe layers that are evidently (from their illustrations)Yarmoukian in age.Much of the geological analysisrelatestoterrarossadevelopmentat‘AinGhazal,andthissurelyhaslittletodowiththegray,rockysedimentsthatcharacterizeYarmoukianlayers.Thestatementthat“...itisnotclearwhetherthe‘Yarmoukianlandslides’wereindeedlandslidesorwhethertheywereconnectedwithheavyrainsorearthquakes”isperplexingsinceitseemstobespeculationthattheresearchshouldhaveaddressedinthefirstplace.Itispossible,ofcourse,thattheresearchprojectisongoingandthatthisissuewill

beaddressedinthefuture.In summary, the discussion of the relationship of

rubblelayerswithanthropogenicandnaturalagencieshas shown that both could be responsible, and bothkindsofactivitycouldhavebeeninplaysimultaneouslyat some sites, depending on topographical situation.And it is also possible, even probable, that the rarecloudburst thatdumpedenormousquantitiesofwaterondegradedslopes(eithernaturally,duetodrought,ordue tohumanactivitydue todeforestationandbrushremoval, orboth)did, in fact, result inmovementofmassesofdebrisdownthehillsides.The8.6-8.0“event”witnessedalongperiodofdepressedtemperaturesandreducedrainfall,andoverthose600yearsitislikelythatdifferentcombinationsofnaturalandhumanagenciescontributedtorubblelayersinthehillyregionsofthesouthernLevant.

Notes1 It is intriguing thatGebel insertsacoupleofdistinctions intotheendofthelatepreceramicNeolithicperiod,usingLatePPNB,FinalPPNB,andPPNCsubdivisions.This topic isdeservingofmorediscussioninalaterissueofNeo-Lithics.

2Intheirarticle,Kafafietal.claimthat‘AinGhazal’smaximumareawasc.10hectares;thisisthecasefortheLPPNBsettlementonthewesternsideoftheZarqaRiver.TherewasalsoanLPPNBenclave of 3-4 hectares across the Zarqa River from the mainsettlementduringthisperiod).

References

NeefR.2004 VegetationandPlantHusbandry.InH.J.Nissen,M. MuheisenandH.G.K.Gebel(eds.),BastaI.TheHuman Ecology:187-218.Berlin,exoriente.

RollefsonG.1998 ‘AinGhazal(Jordan):RitualandCeremonyIII. Paléorient24/1:43-58.

RollefsonG.andKafafiZ.1996 The1995Seasonat‘AynGhazal:PreliminaryReport. AnnualoftheDepartmentofAntiquitiesofJordan 40:11-28.

RollefsonG.andPineK.2007 MeasuringtheImpactofLPPNBImmigrationinto HighlandJordan.StudiesintheHistoryand ArchaeologyofJordan10:473-481.

the intricacy of neolithic rubble layers. the ba‘ja, basta ... · the intricacy of neolithic...

Documents