A lost radiocarbon date for Shetland R C Barcham* CaldeT n 1949I S C r, excavate site sidf Stanydal dth eo W f Mainland eo e th n eo e th , largest island in the Shetland group. He later described this site as a Neolithic temple on the basis of some very broad morphological similarities with the Maltese temple of Mnaidra (Calder 1950,203-5) subsequens .Hi t identificatio excavatiod nan thif o s E dwellinstructura S f no m 0 g20 e knowd same an th ey nb nam ee architectura suggesteth m hi thud o temporade t an th sl l con- nexions between house and 'temple' and also prompted a fresh perception of the prehistoric landscape of Shetland. His hypotheses were tested favourably by further excavations at Gruting School - perhaps less ambiguously called Scutta Voe - and at Ness of Gruting (ibid, 343-56). * 24 Whitehaven Road, Horndean, Hants SHORTER3 NOTE50 | S Four house-sites and their congruent field systems were noted in an area of abandoned croftland at Ness of Grating. Two of these sites (NGR HU 277484 & 277845) were briefly examined with trial trenches lattee sincs ;th rha e beeburna s na reclassifietS moundO e th y .db The interior of a third site (HU 281483) was totally excavated. The walls of the structure were cleared and have been left standing. The excavated house measured 17-4 m by 11-6 m. Its walls were up to 4-7 m thick and attained a maximum height of 0-9 m. Within the core of the walling an enormous quantity of peat ash was found, sometimes supplemented by stones and brown soil, containing numerous sherds which were tentatively explained as the residue of a potter's work- shop. Sealed unde deepese rth t deposi werh as epurf f o o t12- eg 7 k carbonise d grain, identified by Hans Helbae hulles ka naked dan d barley, whic nexy hla halo t trougfa h quern pottere Th . y was assigned by Henshall (1956, 381-91) to the western Neolithic tradition although Beaker influence was suggested by some of the ceramic motifs and by the full and miniature battle-axes (Roe 1966, 222) and a sponge-finger stone (Smith & Simpson 1966, 149-51). Despite the affinities of the Ness of Gruting assemblage with artefacts from the mainland, the prehistor f Shetlanyo d still possessed onl e mosyth t rudimentary chronologies onle Th y. stratified find of metalwork and metallurgical debris was the product of the Irish smith at Jarlshof whose work was assigned to the Adabrock phase of the later Bronze Age (Hamilton 1956, 29; Coles 1960, 48). This provided the only certain terminus in the Shetland sequence. Persuasive argument culturae th r sfo l retardatio islande th f no s coul readile db y advanced (Henshall 1963a, 152) and were given obvious support by the evidence for lithic continuity. Identical stone imple- ments and trough querns were found in contexts which were possibly a millennium apart. For example, the stone tools from Ness of Gruting match those from a site at Wiltrow associated with pottery which is said to have been splashed with iron from smelting operations (Curie 1936, , althoug8) g 15fi connectioe 5& hth bees nha n vigorously challenged (Calder 1956, 379). n 1970I , however, Professor Stuart Piggott submitte sampleo dtw f carboniseo s d barley for radiocarbon determinations to the Laboratorium voor Algemene Natuurkunde at the Rijksuniversitei Groningenn i t sample froOn s . well-knowe emwa th n sit Itforf eo d Hil Sussen li x and yielded a date (GrN - 6167) of 1000 ±35 be (Holden 1972, 89). The other sample was registere- 616 gavd N 8 an resulesama e Gr s th df 176be a e5 o tn 5 I . year 0+ , Professor Piggott delivered one of a series of four lectures in honour of Professor Eric Birley; this was later published (Piggott 1974). In the paper, reference was made (ibid, 3) to field systems, barley cultivation f ardo n Shetlane si us e 18t th e hth footnota n di centur d n I . an eyBC (ibid, 10)e sourcth , e of this date was described as burnt grain from Stanydale. Yet Calder's reports make no mentio f burnno t grain having been foun t eitheda r sit f thaeo t name (pace Mega Simpsow& n 1979, 136). implicatione Th radiocarboa f so n dat either efo r Stanydale site were exciting. Since both sites were presumed to be related by reason of their proximity and the similarity of their con- struction and their assemblages, a date for one would indicate a comparable date for the other. samthe e At time distinca , t horizo largethe rn for Stanydal e site would hel furnispto h chrono- logical links with both the heel-shaped tombs of Shetland and the numerous house-sites (including Ness of Gruting) reported by Calder. Moreover, a site which has never comfortably accom- modate ordinarn da y domestic role (Henshall 1963a, 153; Thor Merritn& t 1978 ) coul57 ,e db firmly correlated, withi limitatione nth singla f o s e date, with sitesimilaa f o s r (though more insubstantial) nature furthe furtheA . S r r correlatio radiocarbof o n- n datBeaked ean r sherds (from Stanydale: Clarke 1970, 521) - would rebuff those who believed in the backwardness of the islands during the 2nd millennium BC. Neithe Groningee th f ro n (Ja t assayye n s 1980sha ) been publishe Radiocarbon;n di however, 504 | PROCEEDINGS OF THE SOCIETY, 1978-80 secona d Shetland radiocarbon datpublishes ewa than di t journa 197n i l 1 (Barker, Burleigh& Meeks 1971, 177): this was the product of a radiocarbon assay on burnt grain from Ness of Gruting. The sample had been submitted by R B K Stevenson nearly 20 years after its collection from its place of storage in the National Museum of Antiquities. The resultant date was 1564± 120 be (BM-441). It is unfortunate that Dr W G Mook, the Director of the Groningen laboratory, was unabl informatiofineto dany Stanydalthe non e sample responsin , quera eto y fro writermthe . Professor Piggott helped to bridge this strange hiatus when he wrote (in lit, 17 2 79) that the second sample whic beed hha n submitte Groningeo dt n (GrN - 6168 alsod ha ) been froe mth grain cach t Nesea f Grutingso t somA . e pointbeed ha n t i ,misattribute Stanydale th o dt e sites, a mere 2 km away. Thus, there are two radiocarbon dates for Ness of Gruting. The better known date (BM-441) untid ha l recently only seldom been discussed (Henshall 1974, 162; Cole Hardins& g 1979, 251; Whittle 1979, 170) discrepance .Th y betwee dateo tw unremarkable s i e nth ; the statisticalle yar y indistinguishable (vide War Wilsod& n 1978, 20-1 should )an appraisee db e lighe th th f o tn d i suggestion that a realistic standard deviation of at least 100 years should be applied to all routine radiocarbon measurements (Pilcher & Baillie 1978, 220-1). In addition, the error potential in radiocarbon dating of short-lived samples (in particular, carbonised grain) has recently been noted (Campbell, Baxter & Alcock 1979, 35). It is also worthwhile to stress again that the grain had been in storage for two decades. e importancTh e Nes th f Grutin o f s o e g date s thai s t they plac e assemblageeth s from Calder's excavations unequivocably withi nBeaker/earliea r Bronz contexte eAg . They compare well wit datee hth s derived fro intermediatn ma e phas occupatiof eo house-sita t na t Sumburgea h (Table 1), 1 km from Jarlshof (Department of the Environment 1975, 126). In addition, chrono- logical dept bees hha n prehistore giveth o nt Shetlanf yo seriea firse f th o radiocarbof t sy o d b n dates from the settlement of Scord of Brouster (Calder 1956, 370-1; Whittle 1979, 169). Some addeoutline detaie lateb e th n th do rt f lca e Bronzo Shetlann i hypothesi n e a f eo Ag dd witai se hth that some of the Shetland burnt mounds should have dates comparable with those noted for such site Orknen si y (Hedges 1975 23)generae g fi ,Th . l similarit pottere th f yo y fro sitee mth s excavate Caldey db r (and Curie) indicate distincsa t chronological horizon, althougR W hA Whittle (pers comm s suggesteha ) dlikenesa s between pottere th y from Nes f Grutino s d gan Scor f Broustero d ; indeed , e pottermucth f ho y once though extraneoue b o t e mait th no t s cultural sequence and proof of casual occupation (Stevenson 1960, 1) has been reassigned to earliee th r tradition (Henshall 1963b, t 68)discussioYe . materiae th f no l cultur f prehistorieo c Shetland remains ill-informed. Tacitus records one cursory inspection of Shetland (De vita Agricolae, 10, 4); there have been many others. The failures of perception which have been articulate r Orknedfo y (Clarke 1976, 240-5 distinctle )ar y pertinen Shetlano t wells da . TABLE 1 Archaeological radiocarbon dates for Shetland (to 31 xii 79) R NG RadiocarboSitne date (years be) Reference 1 Hil Shurtof o l n HU-441403 2850± ? (UB-2122) Whittle 19790 17 , 2 Scor Broustef do r HU-257516 222 (HAR-24130 0±8 ) Whittle 19799 16 , 3 Nes Grutinf so g HU-281483 1760±55 (GrN-6168) Piggott 19740 1 , 4 Ness of Gruting HU-281483 1564 ±120 (BM-441) Henshall 1974, 164 5 Sumburgh HU-392107 1679 + 53 (GU-1006) Whittle 1979, 170 6 Sumburgh HU-392107 1550±153 (GU-1015) Whittle 1979, 170 Dat fros i e1 m charcoal beneat hfield-dykea froe ; ar datem 6 house-sitess2- .