NOTES 715

lnn3.norr1eirl rlir~la.~crl:a~yii~i~olc~f~~owt~r.Su~ll~i~n~pton: I < m ~ c i ~ C. & H.1. I.ir:n%l. 1978. Die , l e s g r n l ~ e n ~ ~ ~ ~ dc i l in ix~srr i lp n l So~~lbnmplan. h'onigpfr,lz l n g t ~ l l ~ c i ~ s . 1909-1911. Rooo.

Pti.~~:UL:K. O.F.S. & V A . M:!XI:IIILU. 10514. 711r: liootnri iiilperirrl REINtZC~I.I~, hl. 1970. l-li.-lov 01 ~ x ~ r p l e CIS r, ~ 1 ~ ~ 1 ~ s syt~?t~ol ill porplly,). rltjurrir?~: GrOtd Dokllnn, Egypt, lillcri~n nrllarl. Arlliqllily RI~ISCIS. Cull. Lotun,us 115. I!l!14. Soutiii$mptnn sntl Exctor. R1r:llr:. P 1!175. Los roprGscnlstions r lu polnis clans les lsrtes

1!105. T?le Roirlnii i i i l ~ l c r i ~ , l p o r ~ ~ l l y ~ rjonriics: GcbelDoMrrm. l i t t l ' ia irr l du l l i l l lt mnynn Ago. Fmncilr 4: 151-71. Egvpl. Inlcrlrri Ruporl.. 1995. Southi~rnpton und Ernler Rlrlll'u. C;. 1992. Feirberggranile as o substiluto lorClnudiiinus

1'195. ??iu Roiit<tn hliilerinl unmlii~n.oonrri~s: Cclrel Dokl~riil nranilc, i n h.1. Witolkuns & N. l l u r r ict1.l. ,lr~cie,il slnric.~: , . . . . . . . Egypl. InIcriIn R P ~ o I ~ I . . 19!16. Snlllllilnlplocl itnd EICIUT ritii~n))illg, lnrrlc onclpmi,e,~n,tce: 1:ll-5. I.suvt?n.

1097. Siln.rv nrid asccii,rrllon ol bloil?; Closdirmor, IllN7- SCHI:IIEI(T. E. I9HZ. Dio Biat~gsscbicl~tt~ uri<l tl ie A~lgitbun dc:r 19!1:l 1: 'rim t o ~ ~ o g i o ~ i l i y rind rjoo,rir?;. Cuiru. l i tc~ior ircl iun Quullcn iihor riin ~or;ir~t~unge.n ~ u i s o r OIIUS

i'l:\rn~.D.P.S. & D.E Wtubws. 10VZ. In~ouiterlKami~n rni>rl,h lmrn I unrl reinor [;smnlliin Edith. Acln Ilirl, orl. Acorl. rci. Uncoln, Anc1r:iii l\lonu~ilcnfs h h m l a i y Rcpoii 36/92.

I'E,\I.III:~.U.~~S.. 0. Wll . l .~~\h l~-Tt~~n~' t : , R.S.TIIIIRI~ER A.G. T t ~ n t . ~ . 1094. Monr Clc~udi i~oor imcl the prubloro ul111u 'gmoilo drdforo': il guologici~l and guochomicnl nppmnch. ;lrlliq- uity 58: 209-30.

R:\tlT%. ]'.A. 19111. hlerlicvul milling. i n D.W. Crusrlcy lod.). h l e ~ l i e ~ ~ , ~ l inclustry: 1-15. L o r ~ c l ~ ~ r ~ : Cc~t~n<:i I lc>r IJrilis11 Archnnnlngy. Rcsnixrcll rupoi.1 40.

R:!tlTz. P.A, k R. MEESUS. I I l02. A,, Anglo-Soroll 11~~1I~iiriilIo1 Toritu.orrIt. London: Cuunci l lo r B r i l i r l i ~\rr : l~i~onlngy. Rcrullicli report R3.

enl. Hlirlg. 28: 229-35. srl$~-l-liil, E.R.A. 1060. Tllr! Alcririd of 21r~r1o Corrtnrrio.

I-iilmmontlr,,.oitl,: I'ongoin. Vux C:IIIII\NSBN, n. & E, wi)l<Naa. 1876. Ro,,~i.sc:hc Stciiibrilclzc

riufdo!~ Frlrbcig on rler Brrgslrnsra iii hirlorirclie ttrtrl It~clinichnr Berleltung. Lli~rmrtildt.

IVIIITTI.OCK. U. 1955. Et~glirln l~isiaNcn1 docuiiteiilr. t:. .5,>o-l012. London: Eyru & Spottiswoodo.

I V i l . t . i ~ h l ~ - l ' t t ~ ~ ~ ~ ' ~ ~ 0. k R.S. THIINI~E. 1903. Mi~gt~oticsusccpti- b i l i ty user1 i n iloll-ric?slruclivc pruvcnilncing o f Rorniin griu>ilc colunlns. ;\r~:bnaoisr:rr). 35: 185-95.

Rock-art image in Fern Cave, Lava Beds National Monument, California:

not the AD 1054 (Crab Nebula) supernova

R.A. ARMITAGE, M. HYMAN, J. SOUTHON, C. BARAT & M.W. ROWE*

The visuol manifestation of the recent Hale-Bopp comet reminds us hen, telling are those rare objects rvhicl~ suddenlyflare in the sky. One can suppose ancient people living b,v

natural light were more compellingl,~ struck bv the sight of comets a n d supernovae, ond understandably researchers seek images of then1 in the sllapes of rock-art motiJs. An

absolute dating contradicts that supposition i n respect of a presunled image of the visible supernova o f r l ~ 1054.

On 4 July AD 1054 a supernova brighter than Venus appeared in the sky, remaining visible for approximately 23 days and 650 nights. I t was chronicled in five independent historic ac- counls, four from China and one from Japan (Dupvenduk 1942). Hubble [I9281 suggested that the Crab Nebula is a result of the AD 1054 evenl, a view generally accepted [Clark & Stephenson 1977). For at least 40 pears investigators have attribuletl certain distinctive rock painlings ant1

carvings in the western United Slates as record- ings of the A D 1054 supernova. More than 20 such depictions (circle or star-like synlbols and a crescent) have been lacated(Brandt & William- son 1979). In particular, two panels of rock paintings in Lava Beds National Monument, California (FIGURE 11, one at Fern Cave and one at Symbol Bridge, were lisled as recording the AD 1054 supernova. Brandt et 01. [1975: 52). noling that the orientation of the moon and su-

* Armil i lgo, Hyrnitn & Roiv,!. Dnp i l r l o~on l olCl iomislry. ' lbxns AXM Unisorsity. Collugc Sl i i l inr l T X 77843. USA. I : 8 1 1 s . . l ~ y m i ~ n ~ c h c m v x . l e m ~ ~ . n r l u r n w o @ c b c ! m a s . l i ~ ~ i ~ ~ ~ Suulhon, Canlcr l o r Acr:clci i~tor blilss Spcclrorn~:try. Lilwl.c!ncc! 1.iuel.lnrlre Nt l i r l r l i l l l.ilbon!lory. L i v o r ~ n o r c C A !14551. I ISi\ . E-mil i l : juhn.suutl~on@q~~ickmi~il.LLNL.gov Bitritt. Lilva Burls Ni~t ian ix l Monomon l . 'l'ulr Lakc! CIA 08134. I ISA . E-mil i l : li~bc.i~dminisln~tianQnps.gov

IZn<:<!i~v!d I 0 Drx:ctnI~r!r 1006, ;~c:~:~!plc~l 151 h,lttrc:l~ l ! l ! l i , rr!visocI 1!1 tvlny 1007,

, \Nr I~ l l r rY 71 [100 i ] : 715-10

710 NOTES

121°45' 121°30' Oregon 121°15'

42'00' Calbrnia

41'45'

PlCultE 1. Location oflhe Lovo Beds Nolionol Llonunlent, California.

pernova are not correct in the Fern Cave rock Experimental procedure painting, concluded that 'errors in recording We took charcoal pigment samples from three the orientation of the crescent moon are com- figures in proximity at Fern Cave: a crescent mon'. The only direct means of assessing the pointing donrnward and two near-by circles. likelihood that a 'supernova' representation one above and one below the crescent (FIGURE records the AD 1054 event is to date the rock 2). Small amounts ofcharcoal were scraped from painting or carving. In our laboratory at Texas the crescent and two circles individually. Rubber A&M University, we developed a plasma-chemi- gloves were worn to avoid contamination dur- cal extraction technique that permits analysis ing sampling and all subsequent handling. Each of liC in rock paintings, whether the pigments charcoal sample was placed on aluminium foil. used were charcoal or inorganic iron- and man- wrapped, and sealed in a plastic bag. The motif ganese- oxides and hydroxides with organic was photographed beIore ancl after sample col- binder/vehicles (Ilger el 01. 1996). This paper lection. Damage incurred to the three paintings presents direct "Cage estimates on rock paint- was sosmall that i t was difficult to determine by ings that have been suggested to represent the visual inspection where thesample wasremoved. AD 1054 supernova. The AMS 'IC analysis on The samples were treated with 1 M NaOll each sample using our plasma-chemical extrac- ant1 sonicated at 50°C, a standard procedure tion technique shows that these images (lo not used to remove possible humic ancl fulvic ac- represent the AD 1054 supernova. ids that might contaminate the charcoal to be

approximate boundary Lava Beds - National Monument

Mt Dome Quadrangle

Medicine Lake Quadrangle

-

Tulelake Quadrangle

8 O U

km

limber Mt Quadrangle

NOTES 717

FIGLIKE 2. Tile rock art prmel showing 1u.o circles to the left oJu crescent. on irnnge- ll~ut hos been suggested us o recording oJ the AD

1054 supernol,n e.~plo,sian. The scale is 10 cm long.

FIGURE 3. Rodio- carbon colibrotion c~lrve in the vicinity 3 850 ofAn 105-1. The "C , - dote corresponding most closely n.ith AD 2 I 10.54 1938 ' j ~ j , eo r s b.p.1 lies wifhin the i l u uncertainty bond of the colibrotion curve for colibroted ages from AD 1030- 950 1000 1050 1100 1150 1200 1250 1160. (Stuiver 6 Reinrer 1993). calendar year (AD)

analysed for 'IC. Humic and fulvicacids are brom- ish-yellow in NaOH. It took 8-10 one-hour treat- ments with NaOH before the solutions appeared clear: we then did three additional extractions to ensure complete removal of the humic and hlvic acid components. Neutralization of the NaOH solutions with 1 M HC1 produced no hu- mic acid precipitate; thus the brownish colour

was likely due to fulvic acids. The NaOII-treated charcoal samples were then rinsed with doubly distilled, de-ionised water, filtered and dried; they were then ready for plasma-chemical extraction of the organic carbon for '"C analysis.

In preparation for the extraction, we use ra- dio-frequency generated, low-temperature [<175") oxygen plasmas to remove organic

material as COZ from the empty reaction cham- ber. Argon plasmas are used on the sample af- ter its insertion into the chamber to remove adsorbed C02 from the system. Finally, oxy- gen plasmas are utilized on the paint sample to convert the organic carbon to CO.,, leaving the substrate rock and accretion carbonates and oxalates intact. This organic carbon is then analysadby AMS. Experimental details, reported in our previous paper(1lger et ol.1996), are not repeated here. Since the introduction of our plasma-chemical technique in 1990, we have demonstrated its validity on numerous sam- ples of known 'IC content: charcoal (two daled previously by Beta Analytic, Inc. and one dated previously at the University of Texas Radio- carbon Laboratory), Third International Radio- carbon Intercomparison wood and African Ostrich shell (dated at the University of Ari- zona). Satisfactory agreement was observed in all cases. Our analyses of '*C-free samples - Albertite, IAEA wood and Axel Heiherg wood - demonstrated that our technique does not add significantly to the modern carbon back- ground of the AMS. The following have also been studied and do not affect our ability to estimate the age orrock paintings by 'C analy- sis: argon and oxygen sources; mass fractio- nation; calcium carbonate, magnesium carbonate, limestone, and calcium oxalate de- compositions. The "C determinations we oh- tained on rock paintings from France. Montana. Texas and Utah are consistent with the age ranges expected from arcliaeological inference.

We used oxygen plasmas coupled with high vacuum techniques to remove o~ganic carbon from the charcoal paint of each Fern Cave sample: the two circles and a crescent shown in FIGIJRI: 2. 'l'lie CO prorl~~vud was rullccted IIV rrrazing in i~

liqulrl-ni~rogen cooled gli~ss-fingrr. Rat~ior:;~rbon contents of the samples were measured at the Center for Accelerator Mass Spectrometry of the Lawrence Livermore National Laboratory.

Results and discussion Radiocarbon determinations and calibratetl calendar dates are shown in TABLE 1 (Stuiver & Reinler 1993). Attempts to verify the valid- ity of rock paintings as representations of the AD 1054 supernova are limited by the nature and uncertainty of the radiocarbon calibration curve in the critical area. The "C determina- tion that most closelp corresponcls to the cal-

snmple weight of lnboratory uncalibratedcalihrated AD

carbon, pg numher determination, b.p. +la uncertainty

+1c uncertainty

Lower circlo 250 CAMS-27229 840+70 1020-1290 Upper circle 185 CAMS-27860 230C70 1490-1955* Crescent 230 CAMS-278fil 330+50 1440-1670

1955 dcnolos the influoncu of bomb "C.

TABLE 1. Radiocarbon detsrnlinntions and calibrated calendar doles (Stuiver 6 Rein~er 19931 for the thee Fern Caves rock art images: ttvo circles and a crescent.

endar date AD 1054 is 938 'IC years before present (b.p.1. FIGURE: 3 illustrates that 938 years b.p. lies within the i l o uncertainty band of the calibration curve lor calibrated ages from AD 1090-1160. Even if one could measure the 'IC content with perfect accuracy and infinite pre- cision, it would not be possible to limit the calendar age of a rock painting painted in AD 1054 to better than this 130-year range: there is no way to determine which is the 'Lrue' date. Nonetheless, analysing "C from the rockpaint- ings directly is the only way to affirm or deny that they are consistent with their assi, onrnent as representing the AD 1054 supernova. A "C content consistent with AD 1054 does not prove that a depiction is of the supernova: "C determi- nations not consistent with AD 1054 (within experimental uncertainty] effectively rule out an image as recording the supernova.

For the Fern Cave samples, the lower cir- cle, calibratecl at AD 1020-1290 (+2o range) is distinctly older than both tha upper circle, at AD 1490-1955, and the crescent, at AD 1440- 1670. Radiocarbon determinations on the up- per circle and crescent are not significantly different from one another: they may have been painted contemporaneously. Both features date from several centuries after the AD 1054 super- nova. There were no discernible differences in terms ofcolour or other visual features between the three figures studied here. The '+C contents of L11e three figures of the painted panel dem- onstrate conclusively that it does not represent the AD 1054 supernova.

NOTES 719

Acknocvledgernen1.s. Partial support for this work was sup- National Nood from lhn l lS Dcpnrtrnonl rrf Erlucntian. Wc pliotl by the Chnrlcs A. and Annc Morrow Lindbcrgh Foun- upprcciotc tho caopcration o f the following ill tile L a w Bods dillion and the National Ccnlcr far Frosomotion Tcchnolagy National Monurnooi: Crsig W. Darrnan: Cnry i-luthuwuy: a n d Training. Discussion with Dr Puulil Rcirncr concorn- Michcllc Mooro: Chris Raundtroo: and Kim Strussburg. ing tho colibrotion of rodiocnrhon dotes was holphll. One Lawrence Livormorc Nntionsl Laborntory is funded by tho of os (RAA) holds o Grndualn Assistnnlship in Aroits of US Dcpailmcnl ofEnorgy undcr Contract W-7405-Eng-48.

Raforonccs BKANDT.J.C.. S.P. MI\RI\N.R.A. W I L L I ~ ~ ~ I S O N , R.S. IIAl<HINCTON, A.D. Part I. Tho ancient oricntill chronicin, Poblicnlions

C. COCHRAN. M.M. KENNEDY, W.J. KENNEDY & VON DEL of the rlrlrononricnl Sacicty of tllc Pncific 54: t11-4. C H ~ ~ ~ I ~ I E I < L I \ I N . 1875. P o ~ ~ i b l ~ . r o ~ k OII records oft110 Cmh Illlilll1.E. E. 10ZU. Novric or temporary sl&ar. Arfrononticnl So- Nebeln supornova in tho woetorn Unitcd Stntos, in A.F. cictjr ofthe Pacific L#nflrt 14: 55-8. Avoni lod.l,Amhoeanstmnon~~vlnPic-Calumbinn Ametico: II.GER. W., M. HYhlAN. I.R. SOLrrliON & M.W. RDIYC. 1886. Rn- 45-58. Austin (TXJ: Univorsily o f l b ~ o s PIBSS. diocnrban DotingalAnciont Rock I'nintiogs, i n k1.V. Oms

BRANm. J.C. & R.A. WILLI~ \~~SON. 1979. The 1054 S L L ~ C ~ O Y V onri lod.1 ~Irci~oeologicul Cltrmirtr).: O i p n i c , Inorganic. atid nstivo Atnotican mck art. Archneooslmnonty 1: 51-37, BiocltcniicnlAnoiysij:401-14. Wnsl>ington [OC): hmsricnn

CLAI~K, 0.H. & F.R. STCPIIENSON. 1977. Tile filislon'cal Super- Chernicol Socicty novoe: 140-60. Nuw York [NYI: Pcrgninon Prorr. srutvnn, M. & P.J. REIMEH. 1990. Exlendcd "C dills ixlst! innd

DuYvENDue, J.J.L. 1942. Furtiter dvti~ bcuring an tlla idcntifi- icuisod CALIB 3.0 "C age culibri~tion pmgnlro. Radio- cotion of tho Crab Nninlln with tho suparnovn of 1054 carbon 35: 216-00.

Spinning or sailing?: the boat models kom Eridu

More on whether the prehistoric pottery vessels from Eridu, in kfesopotomio, ore models of precocious boats, o r 0s wos orgued in o recent ANTIQUITJ: spinning bowls.

A sceptical view of received wisdom is much to be encouraged, and in that sense we wel- come Thomas Strasser's reinterpretation, in the December A N T I q u r r Y , of the so-called boat models from Eridu. There are, however, strong arguments for rejecting his spinning bowl hy- pothesis. The h o w n spinning bowls from Egypt and Palestine are massive and heavy, in order to provide tension against which to pull and ply or twist the fibres being prepared. (Techni- cally these bowls are not for spinning per se: Barber 1992: 72.) They must be sufficiently stable to stay in place and not overturn. Barber sug- gests that they are also used for 'wetting' and that these bowls were associated specifically with the production of linen (Barber 1992: 72). The eastern Mediterranean examples have heovy handle-like loops within the base which show thread wear on the undersides. The Eridu ves- sels are far too fragile for such usage and the Eridu 'thwart' is not sufficiently heavy to have survived the necessary tension: nor would the

boat-shape itself have been stable [see illustra- tion in Safar et 01.1981: 227). Strasser also sug- gests - on the model of Aztec spinning bowls -that the socket, preserved in one of the boats and previously assumed to have held a mast. in fact held a rotating spindle from which the thread was drawn. However, the surviving socket is off-centre, which would have produced an unreliable wobble, as would the shape of the vessel itself.

The Eridu 'boats' dale from the early 5th millennium BC, a time when no spinning bowls are attested anywhere. They are found, more- over, within a culture in which yarn and thread are traditionally produced by the use of hand- held spindles (widespread by the time of the Neolithic villages of the 7th millennium BC and more suitable for wool). Unlike Egypt, no spin- ning bowls are illustrated even in later peri- ods, nor have they been found, whereas the use of hand-held spindles continues [for example, Lhe 3rd-millennium BC spinning ladies of Mari:

* M c D n n ~ ~ l d lnstilulc far Archocologici~l Rcscurch, Downing Strect. Cilrnbridgo cn2 :~EI<. 6ngl;lnrl

Rcceivcd 15 Fcbruary 1907. nuccptcd 20 Murcll 1807. ANI'IQIII'I'Y 71 (1087): 718-21

720 NOTES

Parrot 1962: plntc 11) . The only 3.lrsopolami;1n sire of Dosariv;tl~, so i t is the Ericlu reslrlls rhnr bn\vls \\,ill1 1nrtel.1iandles' are o sn~a l . nllntoer. slioulcl be ignored IOares el r i l . 1977,226-7 and largely from the Early Dynastic period, witli interior clay 'flaps', originating at tlie rim and usually isolating a central area in the bowl. Their purpose is not clear; the three- and four-flapped varieties have been interpreted as pot-stands, possibly for collecting seepage from the sup- ported vessel (cl. .. inter olio, Abu Salabikh: Moon 1987: 43; the Nuzi 'goose neck potstand': Starr 1937: plate 95B; Tepe Gawra Level IV: Speiser 1935: plate 29b; and the massive single-strap example from Habuba Kahira Siid: Strommenger 1970: figure 24). These show no sign of having served as spinning bowls, nor would their de- sign have been efficient for such a purpose (see now Allen in press). In the eastern Mecliterra- nean the earliest spinning bowl would appear to he that from Myrtos, illustrated by Strasser (Early Bronze Age): such vessels are not attested in Egypt before the Middle Bronze Age (12th Dynasty, witli a single model dating to the 11th Dynasty), and appear later still in Palestine. Strasser suggests that theboat interpretation is 'precocious' and 'incongnlously early'; the same can be said of the spinning bowl hypothesis.

Nor has Strasser read carefully the Roaf & Galbraitli article (1994) which is said to 'cast doubt on' the neutron activation analyses of 'Ubaid pottery from sites along the Arabian Gulf with which the Eridu boats have been coupled. Roaf & Galbraitli suggest more sophisticated statistical techniques and point out several re- grettable errors, the most serious of which in- volves tlie attribution of the same data to two sites. The original computer data are now lost, but the first print-out, at which time only the Arabian slierds had been analysed, shows these data to be correctly attributed to the Arabian

Thomas E Strasser* comments: Bourriau & Oates present several interesting observations that warrant a brief response.

First, it is not entirely clear to me why the Eridu vessels are not 'massive and heavy' enough to function as small domestic spinning bowls. Dothan (1963: figure 1) published several ce- ramic spinning bowls of the same size and even smaller. Second, not all spinning bowls had flat bottoms, so wobble must have been some-

figure 2; ~ i a f & ~ a l b r a i t h 1994: 773). This un- fortunate error in oo way invalidates the over- all results which are further supported by Kamilli's thin-section data. Indeed it serves to emphasize the similarity of 'Ubaid pottery from the Gulf and from southern Mesopotamia. Moreover, Roaf & Galhraith conclude tliat their study 'supports the two main conclusions of SMU [Oates et 01.19771: some (or indeed most) of the Ubaid pottery found in the Gulf could have been imported from southern Mesopota- mia; tlie coarse red ware . . . is very different and may have been produced locally' (Roaf & Galbraith 1994: 778). Recent archaeological work in the Gulf continues to demonstrate the mari- time distribution oftlie 'Uhaid pottery (see, most recently, Uerplnann Yr Uerpniann 1996, which publishes 'Ubaid-related sites as far south as Umm a1 Qaiwain). The fact that boat models from Mashnaqa on the River Khahur are of an- other, and better-known, design is irrelevanl to the argument since such canoe-like vessels would have been suitable neither lor carrying cargo nor for negotiating the vast stretches of open water in the Gulf, where access to drinking water is seriously limited Technologically, the broad beam of the Eridu 'boats' would have provided not only space for cargo but also greater stability.

We are not insisting that the Eridu vessels must be sailing boats, though we see no strono

9 argument to the contrary and the 'seafaring activities of the 'Ubaid period remain securely attested, only that on closer examination the spin- ning howl hypothesis would seem to be even less plausible. We are grateful to Strasser, however, for leading us to look more closely at the all-too- limited evidence for spinning techniques.

how managed, perhaps with a wedge [a piece of cloth?). Third, despite the definite evidence for spinning in the 'Ubaid levels at Eridu, tlie idea that contemporary spinning howls would be just as precocious and incongruous as sail- ing boats is a valid criticism tliat reveals a point I failed to make in ANTIQUITY. There are more misinterpreted objects i n the Neolithic and Bronze Age archaeological record that will neecl to be reinterpreted as accoutrements of textile

' Dcpaitmt:nis uf Humnnilios &Rcligious Studios, and History, Cillifol.niu Statc University ISscrumcntol, Sucrurnonlo CA 95819. USi\.