Proc Soc Antiq Scot, 128 (1998), 95-104

A copper flat axehead from Wester Cartmore Farm, Lochgelly, : a cautionary tale Graeme Whittington*, Trevor Cowiej & Peter Yeoman J

ABSTRACT In 1995 a Late Neolithic/Early Bronze Age copper flat axehead was discovered by a metal detectorist in a field near Lochgelly, Fife. Further investigation revealed that the find-spot was located in a glacially deepened depression containing organic sediments. Although the precise stratigraphic location of the axehead could not be ascertained, the find circumstances appeared superficially to be in keeping with deposition ofmetalwork in a 'wetland' setting. In the event, pollen analysis showed this to be untenable. The results provide a salutary reminder of the need for caution when interpreting the possible significance of discoveries of metalwork from peat deposits, in the absence of a secure context and independent dating evidence.

Y DISCOVER F O S CIRCUMSTANCE During the late summer of 1995 Derek Adie of Lochgelly, Fife, found a Late Neolithic/Early Bronze Age copper axehead while metal detecting at Wester Cartmore Farm on the western outskirts of the town (illus 1) (NGR: NT 1745 9412; Fife Sites & Monuments Record NT19SE 071). Mr Adie initially took the axehead to Museum and shortly afterwards the discovery was reported to the National Museums of . Having been claimed as Treasure Trove, the axehead was subsequently allocated to Museum, within whose collecting area the find-spot actually lay. At the earliest opportunity, two of the writers (PY & TC) accompanied Mr Adie to assess the find circumstances. The find-spot lay in poorly drained rough pasture to the west of the farm steading. A trench, approximately 1 m square, was excavated at the location indicated by the finder (although it was no longer possible to discern the small backfilled hole originally dug to retrieve the axehead). No archaeological features were found, but the trench did reveal that at the depth of approximately 0.3 m, at which the axehead was said to have been found, there was a marked change to apparently undisturbed organic sediments. A ranging rod was used to probe them and they were found to be at least 1 m in depth. Further metal-detecting was also carried out in the immediate area but with no positive result.

* School of Geography and Geosciences, The University, St Andrews, Fife KY16 9ST EHh F 11J Edinburg , Street s Chamber , Scotland f o s Museum l Nationa , Archaeology f o t |Departmen , GlenrothesHouse KYe e T Fif Fif ,7 5L , Council e Fif , Unit y Archaeolog e Servic g {Plannin 96 I SOCIETY OF ANTIQUARIES OF SCOTLAND, 1998

• AXEHEAD FINDS (ABOVE) N BRITAI H NORT N I

Y ONL E PROVENANC Y COUNT O

FINDSPOT OF AXEHEAD

PREHISTORIC SITES (RIGHT) LOCHGELLY A ARE Y LOCHGELL E TH N I

2Km /LUMPHINNANS . text the n i o t d referre s site l archaeologica l loca axehea e e th th f o d t dan find-spo e th g showin p ma n ILLULocatio S1 The inset indicates the distribution of copper axeheads from northern Britain after Schmidt & Burgess (1981), s addition h wit

Y VICINIT E IS NTH SITE C PREHISTORI The find-spot of the axehead is located within a fairly rich landscape of prehistoric sites, many of 19te d th h an n i a are y Lochgell e th f o n industrializatio e intensiv e th despite d survive e hav h whic 20th centuries. t north-eas m k 5 2. d locate e ar ) 002 E (NT19N n Cair w Harela e th f o s remain d mutilate e Th of the find-spot; the cairn was investigated in the 1890s, when three cists were found, two of which s thi r earlie d recorde e wer y nearb e circl e ston a d an n cair r Anothe . Vessel d Foo a d containe h eac n o , area e sam s thi n I . 033) & 7 03 E (NT19N d destroye n bee y subsequentl e hav h bot t bu , century the south side of Loch Ore and 1 km to the north of the find-spot, extensive traces of prehistoric settlement are known. Earlier this century, the loch was seriously affected by drainage works related to the coal industry and they led to a number of finds being made, including a 'Roman' spear (NT19NE 039), a possible crannog (NT19NE 035) and a dugout canoe (NT19NE 036). E CARTMOR R WESTE M FRO D AXEHEA T WHITTINGTONFLA R COPPE : , COWIYEOMAN E& 97 y promontor d defende , enclosed n a o t t adjacen y immediatel d foun s wa , destroyed w no , latter e Th at Clune Craig (NT19NE 023), which stood above another enclosure in a low-lying position just to the north (NT19NE 025). A large group of hut circles has been found traversing an east/west ridge, 200 m south of Clune Craig (NT19NE 021,022 & 024). m k 4 1. , site s tumulu e possibl A . find-spot e th f o h sout e th o t s site t significan o als e ar e Ther e Bronz A . 1927-8 n i ) 001 E (NT19S l Hil e Tolli n o d excavate s wa , Cartmore r Weste f o t south-wes Age cemetery was also identified there. A major cropmark site of a possible henge (NT19SE 009) has been recorded 1.5 km south of the find-spot and has recently been conserved. Other e th d an s working l coa d an e ironston f o s remain e th e b o t d prove e hav y vicinit s it n i s cropmark . landscape l ritua s thi f o r survivo y onl e th e b o t s appear e heng

ILLUS 2 Wester Cartmore cumulative n i e ar s age ; diagram n polle uncalibrated radiocarbon years Sandy peat Gyttja y gyttjaSand Silty gyttj) a conversions r fo 1 e Tabl e (se P B 8 199 , SCOTLAND F O S ANTIQUARIE F O Y SOCIET | 8 9

E SIT E AS TTH CONDITION L ENVIRONMENTA d an r diamete m 0 20 c f o , depression d deepene y glaciall a n i d locate d fiel a n i s lie t find-spo e Th often containing standing water, in an area of glacial till derived from Carboniferous deposits. To e th t eas d an h sout e th o t e whil , occur s degree 0 1 o t p u s slope n depressio e th f o t wes d an h nort e th e Th . substrate t consisten a d reveale d fiel e th f o a are e th r ove g Corin . gently y awa s fall d lan t lowes e th h whic f o , infill f o m 3 2. c s contain d an t outle r o t inle m strea o n s possesse n depressio 0.5 m is non-polleniferous (illus 2). The basal 0.5 m consists of sands, silts and clays and are succeeded by gyttja (an organic lake mud) which between 0.9 m and 0.7 m depth becomes much more mineral. A return to more organic gyttja then occurs with the organic content rising to its maximum (95% loss-on-ignition) at 0.3 m. From that point the organic content declines to 70% e th h wit t consisten s i e sequenc e whol s Thi . surface d lan t loss-on-ignitiopresen e th o t p nu accumulation of sediments which typically characterize the period since the removal of the last ice from Fife (Edwards & Whittington 1997). The steep rise in the organic content of the deposits at 0.7 m appears to mark the onset of the major temperature recovery at Wester Cartmore after . Scotland t lef Readvance) d Lomon h (Loc e ic t las e th n (polle s sediment r lowe e Th . views e thes o t t suppor d len n undertake s analyse n polle e Th e th h wit s change h whic a flor n polle a e hav , here d considere t no , 2) s illu n i 7 wc-s wc- o 1t zone e th m 3 0. t A . age e ic n Devensia e th f o e clos e th s toward d occurre t tha e climat n i s fluctuation apparent arrival of the thermophilous species of trees, such as Quercus (oak), Ulmus (elm) and Alnus glutinosa (alder) occurs and those taxa better adapted to cooler conditions, such as Juniperus m 3 0. From . decline d marke a o int o g r o r disappea r eithe ) Salixd an communis (willow ) (juniper to the present-day land surface the pollen spectra are dominated by Betula (birch), Alnus and d woul ofe Betula Corylus-avellana-typet Gale)dominanc e Swee Th . r Myrica— and/o gale l (haze be consistent with the name of the parish — Beath, derived from the Gaelic for birch.

DESCRIPTION OF THE AXEHEAD o t e curv e gentl a n i e diverg s side e th ; butt d arche d broa ; profile g lon r lenticula ; 3) s (illu e ax t Fla meet the strongly curved cutting edge which is backed by a bevel, 7-9 mm wide; generally a s area e som d an n rubbed/wor e wher s highlight d gold-coloure h wit , colour n brow t mat h smoot : Dimensions . texture n i r roughe d an n gree g bein e surfac e th n i d resulte s ha n corrosio r mino e wher . g 5 691. : Weight ; mm 2 1 h T ; mm 5 48. ) (butt W ; mm 1 8 ) edge g (cuttin W ; mm 3 15 L X-ray fluorescence analysis of the axehead by Katherine Eremin (1995) showed that it was . tin d an c arseni , antimony f o s trace h wit , copper f o d compose

DISCUSSION r coppe f o s version e wher , Ireland n i s i s metalworkinf o e island e evidenc thes t n gi earlies y ver e Th axes found in continental Europe suggest that native production may have been under way soon after 2500 cal BC. The typological features of this copper axehead, particularly the thin-butted s axe e Ballybeg/Roseisl s a d classifie p grou a o t s belong t i t tha e indicat , sides g curvin d an m for n I . iii) , al1985 et m (Needha s axe 2 s Clas c generi s Needham' r o ) 27-9 , 1981 s Burges & t (Schmid f o f hal d secon e th o t y confidentl d date e b n ca e Cartmor r Weste m fro d axehea e th , terms d broa y Earl e th o t c Neolithi e Lat e th m fro n transitio f o d perio t tha g durin , BC l ca m millenniu d thir e th Bronze Age when metal was in production and use among what were essentially Neolithic t fla e bronz d thin-butte t firs e th f o s precursor e ar m for s thi f o s axehead , Typologically . societies WHITTINGTON, COWIE & YEOMAN: COPPER FLAT AXEHEAD FROM WESTER CARTMORE 99

0 10

d axehea r coppe e ILLUTh S3

axes, which would have appeared following the adoption of full tin-bronze metallurgy during the last quarter of the millennium. n illue (se compariso s y n 1b ) Britai n norther m fro n know e ar s axe r coppe w fe y Relativel with Ireland, where very large numbers have been recovered (Harbison 1969). The Wester e typ e axeheadf o s s ofBallybeg/Roseisl example n doze a y onl t abou f o e on s i n specime e Cartmor & t (Schmid s Highland e th o t e Yorkshir m fro d distribute y widel d foun , Britain n norther m fro Burgess 1981,114; another recent discovery from Galston, Ayrshire, should also now be added to their corpus). It should be stressed, however, that the present general scarcity of copper axeheads in this zone may be misleading: while the numbers in circulation may have been less than in d an l surviva l differentia o t d le e hav y ma n depositio f o s , differencecircumstance e Ireland th n i s recovery. An unknown proportion of the original total in circulation may have ended up being recycled; this would leave only those which were taken out of circulation in other ways (eg deliberate deposits or casual losses) potentially recoverable as archaeological finds. r Weste e th s circumstance l environmenta t wha n i d an y wh , how o t s a s arise n questio e Th e wer s analyse n polle e close-sampl s Thu . found s wa t i e wher e b o t e cam d axehea e Cartmor e th h whic n i s sediment e th , latter e th From . obtained e wer s date n radiocarbo d an n undertake axehead was presumed to have lain are dated to 9310 + 60 BP (8450-8140 cal BC, Beta-104917). This date clearly suggests that the artefact is not in a correct stratigraphical context. Today the Wester Cartmore field has the appearance, especially in winter, of being an ill- drained area of rough pasture with an irregular surface, rather than a bog, as it supports very few n i s deposit e th f o e featur g strikin t mos e th s Perhap . feature a h suc f o c characteristi s specie e th f o this field, and their pollen, is that the lower 2 m represent some 4000 radiocarbon years while only o o

O •d > H Oc > 52

O

9080*80- z a 9310*60- 30 8590-60-

SOJ

y peat Sand d ban t Sil t pea d UmhumifieHumifie t pea d ILLUS 4 (above and facing) Wester Cartmore percentage pollen diagram for selected taxa for the upper 0.5 m of the sediments; ages are in uncalibrated radiocarbon ) conversions r fo 1 e Tabl e (se P B s year | 10 E 1 CARTMOR R WESTE M FRO D AXEHEA T FLA R COPPE : YEOMAN & E COWI , WHITTINGTON 102 | SOCIETY OF ANTIQUARIES OF SCOTLAND, 1998

0 10,00 e som r fo e existenc n i n bee w no s ha h whic , period e Holocen e th o t g m7 belon 0. r uppe e th s period o tw e th o t g belongin s sediment f o s depth e relativ n i y discrepanc s Thi . years n radiocarbo could be explained by an extremely slowed deposition rate in the Holocene section of the core. d woul s activitie l agricultura e wher , situation d lowlan a n i l unusua t mos e b s thi d woul y onl t No be expected to have enhanced erosion rates on the steep slopes surrounding the depression, but the pollen diagram does not show a complete Holocene vegetation history (Whittington et al e th m fro l materia f o l remova e massiv a n bee s ha e ther t tha d suggeste s i t I . 1991b) ; 1991a depression which could once have held a raised bog. e wher , lochs d an s river , bogs s a h suc , 'wetm contexts fro ' k metalwor h muc f o y recover e Th h suc y man f o n interpretatio e th o t d le s ha , impossible r o t difficul n bee e hav d woul y recover deposits as votive offerings. It could be speculated that if the axehead is evidence of a votive or d explaine e b d coul e dat e Holocen y earl y ver a s ha h whic m stratu a n i n locatio s it , offering l ritua d consolidate y poorl f o e surfac y contemporar e th w belo l wel t i g carryin t objec e th f o t weigh e th y b sediments of the bog. For this suggestion to be supported, it would have been necessary, at the o t r wate f o s pool t intermitten g containin g bo a r fo , circulation n i s wa d axehea e th n whe e tim e Th . suggestion s thi o t s objection l palynologica , however , are e Ther . existence n i n bee e hav e existenc n i g bein s plant d lan t we r o c aquati f o n indicatio o n y virtuall s (illum ha ) s4 diagra n polle . level t tha t a y dramaticall f of s fall ) (sedges e Cyperacea f o e presenc e th n eve ; m 3 0. e abov Illustration 4 is a selected taxa pollen diagram from the find site. The presumed level of the axehead correlates with the top of pollen assemblage zone wc-9 and it is noticeable that below e som y b d reveale , history n vegetatio d recorde e th n i e chang r majo a n bee d ha e ther l leve t tha e th t a d an b wc-8 e zon f o p to e th s toward n representatio e spor d an n polle n i s change g strikin r Wate d alterniflorum(Alternate-flowere Myriophyllum f o a tax c aquati e Th . wc-9 e zon f o e bas Milfoil), Myriophyllum spicatum (Spiked Water Milfoil), and Potamogetonaceae (Pondweeds) collapse, and the presence of Equisetum (Horsetails) increases nearly eightfold and then collapses Sphagnum n i e increas d thousandfol d an d shortlive a y b d replace e b o t , risen d ha t i s a y swiftl s a (MeadowsweetFilipendula / Equisetum, s a h suc e thos o t t firs y wa e gav s plant r wate e Th . spores . land t we y ver r o r wate y mudd n i e thriv h whic ) Menyanthesd (Bogbean an ) trifoliata Dropwort They in turn were replaced by mosses, over which the layer of silt occurs. Thus by 9310 BP the s plant d wet-lan o n y virtuall d an c aquati o n e ar e ther s a t exis o t d cease d ha g bo e Cartmor r Weste represented above this level. As a result, the possibility of a deliberate Bronze Age deposition of e th s i e sit e th f o e featur t importan r furthe A . abandoned e b o t s ha a are y water a o int d axehea e th existence of 10 mm of coarse silt just above the level at which the axehead was recovered and which appears to represent an inwash of sediment. An explanation of the way in which the axehead achieved its resting place is difficult. It had been hoped that close radiocarbon dating of the sediments above and below the presumed level of the axehead would be illuminating. As can be seen from illus 4 the dates are reversed; the youngest date, 8590 + 60 BP (7695-7505 cal BC, Beta-104918), is lowest in the stratigraphic column and is succeeded by the oldest date, 9310 BP, above which occurs a date of 9080 + 50 BP (8130-8020 cal BC, Beta-104916). The pollen record suggests that the youngest date is too young. t Quercustha d w an sho ) 1991b ; 1991a al et n (Whittingto e Fif n i s record n polle r othe , BP 0 859 y B Ulmus were well established; that is seemingly not the case at Wester Cartmore. While, as will be shown below, the central date does appear to be correct, the upper date also seems to be anomalous. g durin d an e befor t jus s record e spor d an n polle y man f o e natur e th n i s change d marke e Th g followin n basi e th o int n depositio t sedimen n i n dow g slowin a o t e du y probabl e ar 9 wc- e zon e th d ha e hav o als d woul t Tha . slopes g surroundin e th n o d woodlan f o t developmen e th n upo WHITTINGTON, COWIE & YEOMAN: COPPER FLAT AXEHEAD FROM WESTER CARTMORE 103

TABLE 1 Radiocarbon dates Lab code Material dated Years BP Calibrated dates (two sigma) Beta-104916 Humified peat 9080 + 50 8130-8020 calflc Beta-104917 Sandy peat 9310 + 60 8450-8140 calac Beta-104918 Sandy peat 8590 + 60 7695-7505 cal BC Calibrations are based on Stuiver & Reimer (1993)

effect of decreasing run-off into the basin, leading to its drying out and the demise of wetland taxa which could have been replaced by colonizing Betula. This may also have been a response to a marked rise in temperature which occurred at the close of the Devensian ice age and which allowed the colonization of thermophilous arboreal species, as represented in zone we-10 at Wester Cartmore. e th o t y wa s give t pea e wher y stratigraph e th n i t poin a w belo d axeheae situate Th s dwa band of inwashed silt. It is this which suggests not only an explanation for the position of the axehead in non-contemporaneous deposits but also why it occurs in this location. Clearance of trees would have led to erosion of the soils on the steep slopes north and west of the basin, d axehea e th t tha e episod s thi g durin s i t i d an d axehea e th e abov s lie h whic d ban t sil e th g providin would most probably have reached its final resting place in the ground, whatever the precise circumstances. Subsequent to this, peat would have grown again due to an increase in run-off from the slopes allowing the growth of Betula and perhaps of Myrica gale, if it is that taxon which is t apparen e th , occur d di s event f o n chai s avellana-typethi f I . Corylus e th pollen y b d represente n see s i P B 0 931 f o e dat e Th . rationalized e b o als n ca s date n radiocarbo e th f o e natur s anomalou s representative s a ofSalix,e , presenc Juniperus e th Cyperaceae h d wit an s accord t i s a e dat l rea a s a of the previous colder conditions, but also with the time when arboreal taxa were beginning to colonize. Some of this growth would have been on the former bog surface and subsequent root f o e dat e th g providin s thu , sediments e th o int n carbo r younge d introduce e hav y ma n penetratio s a y onl d regarde e b n ca d ban t sil e th e abov s deposit e th m fro P B 0 908 f o e dat e Th . BP 0 859 anomalous. It has been obtained by the AMS method from a very small sample of sediment and , occurred e hav d coul s thi e Whil . place n take s ha n contaminatio t tha y possibilit e th s i e ther e th n i d involve r erro f o e magnitud e th , depression e th o int g drainin r wate d har o t e du s perhap (Kerroc walrealisti et t 1984)no s i , . sediments e Ag e Bronz n i e li o t d axehea e th w allo o t , date The possibility that the axehead was deliberately buried might also be raised because this would have led to a mixing of sediments of differing ages. While the possibility of deliberate burial must remain, that cannot be used to explain the anomalous dates as the monolith used for the . site d fin e immediat e th m fro n take t no s wa s analyse n radiocarbo e th r fo d an s analysi n polle

CONCLUSIONS The most probable explanation for the location of the axehead seems to be that woodland clearance around the Wester Cartmore basin, an unsurprising activity given the wealth of t Subsequen . axes h suc f o y currenc f o d perio e th g durin d occurre , area e th n i y activit c prehistori to that date woodland once again became established in the area. The basin then accumulated , date e undeterminabl d an r late h muc a t A . axehead e th d burie r furthe unhumifieh whic t dpea e th o t d le l fue r fo d nee e th , area e entir e th m fro d remove n bee d ha d woodlan e onc y probabl cutting of peat from this site. This has left only remnant deposits in the basin, as can be seen from 104 SOCIETY OF ANTIQUARIES OF SCOTLAND, 1998

the truncated vegetation history revealed by the pollen analysis. Although the fine condition of l casua n tha r rathe explanatioe deliberatn f o s invit o t term es depositio n i stild tend l axehea e th loss or discard, the pollen analysis from Wester Cartmore shows the very real need for caution when interpreting the possible significance of archaeological finds from peatland, especially inorganic artefacts, and particular circumspection with regard to antiquarian accounts of finds o t t adjunc l vita a s become y palynolog t tha s circumstance h suc n i s i t I . mosses d an s bog m fro excavation.

ACKNOWLEDGEMENTS We are grateful to Derek Adie, Lochgelly, for reporting his discovery, and for his friendly assistance with the subsequent investigation of the find-spot. We are indebted to Sarah Govan of Fife Council for her help with the initial fieldwork. The location map (illus 1) was prepared by Alan Braby while the axehead (illus 3) was drawn by Marion O'Neil. The preparations for the pollen analysis were carried out by Lois Wood and the coring was undertaken by C Cameron and f o t cos e th r cove o t d grant-ai r fo t Trus l Russel e th o t l gratefu y especiall e ar e w , Finally . Jarvis J . dates n radiocarbo e thre

REFERENCES Edwards, K J & Whittington, G 1997 'A 12,000 year record of environmental change in the Lomond Hills, Fife, Scotland: vegetational and climatic variability', Veg Hist Archaeobot, 6(1997), 133-52. d Scotlan f o s Museum l 8980).Nationa (C&AR Fife Lochgelly, from Axe Copper of Analysis 5 199 K , Eremin Analytical Research Section report no 95/35. Harbison, P 1969 The Axes of the Early Bronze Age in Ireland. Munich (=Prahistorische Bronzefunde, IX/ 1). g datin n radiocarbo e th f o y stud e cas a : Pennsylvania , Bog y 198P , 4'Corr Fritz & G B , Warner , F P , Kerrow of marl', Quat Res Ass Techn Guides, 21 (1984), 326-36. Metalwork,Age Bronze British Hoards, Age Bronze Early 5 198 S H n , LawsonGree P S & , J A , Needham . AI-6.London Series Finds Associated Schmidt, P K & Burgess, C B 1981 The Axes of Scotland and Northern England. Munich (=Prahistorische IX/7), . Bronzefunde n 'Extende3 CALId 199 Calibratio e J 14 0 BCP 3. Ag dRevise Bas , a 14d CeDat an Reimer & M , Stuiver Program', Radiocarbon, 35 (1) (1993), 215-30. d an l pollen-analytica l J 199l C , a post-glacia 'Late d an - Caseldine & J K , Edwards , G , Whittington , (1991) 8 6 Palynol, Palaeobot t Fife' Rev , north-eas n i e sit l coasta r nea a m fro a dat l environmenta 65-85. k Blac t a e chang l vegetationa l 'Lateb post-glacia d 1991 an - R P , Cundill & J K , Edwards , G , Whittington Loch, Fife, eastern Scotland — a multiple core approach', New Phytol, 118 (1991), 147-66.