Newsletter 23 - January 2002 - Roman Finds Group

36 Pages·2011·9.16 MB·English

Iucerna
Roman nds Group Newsletter 23
Fi
Contents Editorial
Romandecoratedironstyli ...... 2 Tttis issue is very full, and I am actremely grateful
Wax spatula handle from Yorkshire . ! . . 5 to all the contributors fur their hard work.
, . . . . . ! . . . ,.8
NextRFGmeeting
Writing eguipment has been the subject of much
.
Gold amulet-pendant from Shropshire . 9 study on the continent in recent years, and now a
Romanwatgrworks . . . , r . r . , . 10 rema*able group of s$li ftom Elms Farm,
Bronzevessel fromMiddlewich . .. . 11 Heybridge (a site which has produced many
Ivory knife-handle from Silchester,, . . 12 exceptional small frnds), has provided the
Nomoremystery?.. . r r . r . . . stimulus for Hilary Major's typology of deconted
13
Boxappeal . . . . . r r . . . ! . r L4 styli in Britain,
Bookreviewr . ! . . . r . . . . . . . 14 There is a good crop of small articles, and a long
FRGTOO-1700meeting. . r. . . . . 16 contribution from Bavaria. Though this is a study
Spindlerwhorlrpot. , . , , . . ,., L7 of some unusual deposiB in two late ?ffi-century
The autumn meeting at Exeter . . . . 28 grav6, the piece has major implications reaching
Crosswordrby'Diggler'..,..... back into the Roman period. The paper is an
32 object /esson in avoiding preconceptiofrs, and
Confgrences.,rr!rt...r,33
examining pot contents arefully.
Booksetc. . r r ! . . . . . . . . t .34
For your light relief and enteftainment during a
peaceful tea-break, this issue includes the first of
what I hope will be a senes of ctosswords.
'Digger's' archaeologial-historical crosswords
have been a feature of 'Essex Past and Prsent'
for some time, so apologi* to Essex r*idents
who may have long memori* for clues and their
answers!
Our new Meetings Co-ordinator, Ellen Swifr, has
been working overtime, ananging venu6 for this
autumn and spring 2003, as well as completing
arrangemenB for Segedunum this spring. If you
would like to host a meeting, or have ideas for a
topic, let her know. A booking form for
Segedunum is included in this issue, and if you
want to know the full programme, do contact
Ellen by email, snail mail, or phone. Contad
details on p 34.
Infofmation for contributors Ellen would be grateful if you could find the time
to frll in her questionnaire also enclosed. She
Post your contributions to Nina Cruffiffry, 2 aims to present the findings at a conference at
Hall Road, Copford, Colchester CO6 1BN, or Durfiam, Promoting Roman finds in context and
email them to [email protected], theory, ifi the autumn.
Deadlines May 30th for inclusion in the luly
issue, and November 30th for the January Richard Hobbs, our Secretdry, will shortly be
moving departments within the British Museum.
issue.
He is leaving Portable Antiquities, where he made
Emailed text should be either a .txt, .rtf, or a tremendous cantribution to the prajed, to go to
.doc file. Illustrations should preferably be Prehistoric and Early Europe, replacing C.atherine
simple line drawings or uncluttered photos. Johns. She will be sadly missed, but may then
Emailed pictures should be .tif or .jpg files. have more time to write articles for Lucerna.
Remernber not to breach copyright law when Nina Crummy
sending illustrations.
lu*rna !#$f
-Iariuary
Roman decorated iron styli
Manning (1985, 85) produced a baslc from X-rays. None of the decorated styli
typology of Roman iron styli which has been from Elms Farrn were recognised as such
widely used, and is based princlpally on the prior to X-ray. Few reports specifically
degree of separation of the point and stem. mention X-rays, particularly older finds
All decorated styli were classed as type 4, a publicatioils, and in some reports the objects
very broad and heterogeneous group. It is have obviously been drawn without the use
clear that the classification of decorated styli of X-rays, implying that decorated styli may
wlll bear refinemert, and the following is a have been rnissed during cataloguing.
preliminary note on an attempt to group a Published drawings of styli are often
number of published examples. Styli are the unsatisfactory, since they are usually at 1:Z
only class of Roman iron object which are (the normal reduction for ironwork), making
regularly embellished with moulded bands it difficult to see the details of the
and inlay in other metals, and consequently decoration. The text doesnt always clarify
are potentially more closely datable than the drawing a lt ffi;ty, for exampl€, mention
rnost other ironwork, particularly if different inlald bands, but not how many. Without
groups can be identified and dated. It is also seeing the original objects, some styli can
difficult to imagine a village blacksmith only be provisionally grouped. Howev€r,
making the more elaborately decorated styli, given the tendency of ironwork to
and they surely must have been a disintegrate during storage, I suspect that
specialised product, produced at a limited not all decorated styli survive, particularly
number of workshops, and possibly an those from older excavations.
import from the continent.
The impetus for this study was the finding of THE RANGE OF DECORATIVE
a relatively large number of styli at Elms TECHNTQUES EI|PLOYED
Farm, Heybrldge, Essex. Thirfy sryli were
The styli from Elms Farm employed a variety
fwionuectlrruedd , deae lolc omnreaadt eeod fo ifn tih rovena ,r rioonfuo sws htw icahy tsh.i rTteheeny omfo dste coof rtahteiv ere tpeecrhtoniriqeu feosu nwdh iocnh sintcyollr.p oTrhaetey
elaborately
decorated examples from this country (fig l, fall lnto three main groups: moulded bands,
9i)5. T, haen ds itfeu,n wdehidc hb wy aEsn egxlcisahv aHteedr itiang 1eg agn3d- minceistaedls .d eTchoerraet ioisn ,a lasnod r aarpep tuiesde noof ns-tfaemrropuesd
Bovis Hornes, was a Late lron Age and decoration.
Roman rural settlement, with a central The moulded bands (the term 'moulded'
temple (see Atkinson and Preston lgg8 for does not here imply casting) can take
an interim report). The Late Iron Age site various forms, from simple beading to
seerns to have been of some importance, but polygonal sectioned bands (stylus handles
a slow decline appears to have set in by the are normally circular in section). Three of
second century. This is not the sort of site the Elms Farm styli had bands with polygonal
where one would expect to find large sectiofis, some with non-ferrous applied
numbers of styli, especially the rnore bands, a feature which has been rarely
elaborately decorated ones, which must have recognised elsewhere. It is not something
been valued (and valuable?) possessions. whlch can be easily seen on X-rays, and can
The reason for the presence of so many styli in most cases only be identified during
at a relatively low status slte is not, cleaning. It was therefore probably more
however, considered here. I d0, however, widely used than is evident from the
think it is unlikely that they were being made publisfied examples.
on the site.
Incised decoration most commonly takes the
form of one or more circumferential grooves.
COI{STRAITTTS Oil THE STUDY other incised decoration includes spiral
groov€s, fine rilling, cross-hatchlfig, and
Iron styli of any kind may be difficult to obligue lines. Grooves may originally have
recognise, as they tend to resemble nails contalned non-ferrous inlay, so lt is, perhaps,
wish euns cuoavlelyre hd idind econn bcryet icoonr.r oAsniyo nd,e caonradti oitns wrniitshle aindlianigd tgor odoiffveerse natinadte t hboestwe ewenit he xgarmoopvleess
is
recognition dependent on identification but no inlay.
-r
-a
\!
ll
g h Flr
E
\J
a,
k
hiI.,
aA
b c e s
r)
SOmm
trr
H
Fig. 1 Group 1 a) Gestingthorpe b) Caister-on-Sea c) Elms Farm; Group 2 d) Harlow; Group 3 e) London; Group 4 Q Verulamium;
Group 5 g) Gestlngthorpe; Group 6 h) Elms Farm; Group 7 i) Elms Farm il Gorhambury Group I k) London, St. Magnus House.
i#rry##' )dnuarTo !##l
Applied non-ferrous metals were most Roman (Elms Farm). The bulbous point is
commonly ln the form of wire, or narrow also present on a stylus from London wlth a
strlps, set into a groove. The applied metals copper alloy handle and iron polnt (Manning
from the Elms Farm styli were analysed, and 1985, N25). The handle is decorated wlth
proved to be exclusively brass. Virtually all incised cross-hatching and polygonal
known examples are described as 'copper sectioned panels. It ls prcbably from the
alloy', the exception being one from Walbrook, and thereforc likely to be lst to
Brancaster which has both brass strips and 2nd-century in date, suppofting an early
sllver lattice (Group 5; Sparey Green and Roman date for the type.
Hinchliffe 1985, 53, no77).
The decoratlve motifs present include rilllng,
spiral and circumferential gnooves,
T}IE GROUPS sometimes inlaid with wire, moulded beads,
The grioups are based on a sample of polygonal-sectioned bands and moulded
decoration.
seventy-six styli, from thirty-four sites. The
sample includes four styli which I consider to
be dubious, and sixteen styll which were not GROUP 2 (Fig 1, d)
gtlaerocnuktap oteivdf e,s leiyml tihidlaeerr n sdtiutfiyeel dit. o ( aEinnigcdho mtm gpraleoytue pnysee swst e broeer Sshtyalpi eo fo gf rtohuep e2r aasreer ,c haanrdac ttehreis esdt ybley athnde
positioning of the decoration. The eraser is
subject to change!), based on the form of
waisted, and the decoration consists almost
the stylus, and the style of decoration. The exclusively of low beaded elements with
criteria used for the form were the method of flanking grooves or slight ridges. Six
separation of the point and handle (as with
examples were identlfied, from Bancroft vllla,
Manning's typology), the general shape of
Brough-on-Humber, Harlow, Lullingstone and
the handle (i.e. straight or bulbous), and, to
some extent, the shape of the eraser. Richborough. Where datable, they are frcm
the are Late Roman contexts. Waisted erasers are
Unfortunately, erasers often
damaged or missing, so although I consider probably a late Roman form.
that the eraser shape may be relevant (and
is one of the particular features of Group 2), GROUP 3 (FIg 1, e)
it has generally been afforded less
importance in this scheme than other Tohfe t hthirse eg rdoeuflpn itaer ea ntdy polnfeie pdr obbya btlhe emire mlonbge,r
features,
slightly splayed erasers, and single reels at
It should be emphasised that, on the whole, the base of the handle. The three definlte
the groupings are rather loose, and do not examples are all in the British Museum, and
necessarlly imply that the styli aFe were seen by Manning as a dlstlnct group,
contemporaty, ot were made in the same his type 4a (Manning 1985, N28-N30). The
place. In a few cases, however, styli from probable example is from Baldock, and is
dlfferent sites are similar enough to suggest slenderer, with a shorter splayed eraser.
that they did come from the same workshop. The reel is present, but not flanked by a
groove. One of the lncomplete Elms Farm
GROUP I (Fig 1, a-c) sWli probably had a splayed eraser, and
could belong to this group. None of the
This is the most coherent gtuup, linked by examples is dated.
the unusual form of the point, which is
strongly separated from the handle, and
bulbous. The erasens are semi-clrcular or GROUP 4 (FIg 1, f)
straight-sided where present. Five examples This group features seml-circular eraserc,
were identified, including hvo from Elms with a handle that tends to swell sllghtly
Farm; the others are from Caister-on-Sea towards the point, but not invariably. There
(Darling 1993, 101, no 422), Gestingthorpe ls sllght separation of the point and handle.
(Manning 1985a, 51, no 200)
and The decoratlon consists principally of grooves
Magiovinium (Neal 1987, 52, no 106). Three at the Junction of the point and handle,
of the examples are similar enough to sometimes with similar decoration below the
strongly suggest that they are the products eraser. The grooves are sometimes inlaid
of the same workshop (Ftg 1a-c; from Elms with non-ferrous bands. Eight examples
Farm, Caister-on-Sea, and Gestingthorpe). werc identlfled, five of which came from
The only securely dated example ls early dated conte)cs (although only one was
4
luceme Janua*,2#$J
closely dated), and a 3rd-century date seems with an undifferentiated point. It lncludes the
likely for most of the group. Although they most elaborately decorated stylus flom Elms
are not simllar enough to suggest a common Farm, which has the handle entirely covered
source for all of them, most are from East in decoratlve elements. Most, however, have
Anglia/Herts, including three from various simple decoration consisting of
sites in Verulamium, which suggests that circumfercntial grooves, often rcund the
they are an eastem type, even lf not all from thickest part of the handle (le at the Junction
the same workshop. of the handle and point). Five are from
dated contexts, from the second to 4th
century, and there are possibly at least two
GROUP 5 (Fig 1,9)
sub-groups nepresented, mid-Roman and
Thls is a somewhat disparate group in terms late Roman.
odifs dtienccotlrya tdioifnfe arenndt ieartaesde rp sohinatpse. , Tbhuet jaulnl chtaiovne The Elms Farm stylus (Fig L, i) ls an
exceptional piece of craftsmanship, though
between the handle and point is sharply
now in poor condition, and cracked, wlth the
defined, with the point forming a splke set in
the mlddle of the end of the rylindrical eraser missing. The decoration consists of
bands of applied circumferential brass strips;
handle, which is sometimes slightly flared at
the point end. The decorative techniques three moulded polygonal-sectioned segments
used are variable, including polygonal- wlth triangular facets pointing alternately up
sectioned bands on one example from and down, wlth an applied strip of brass
and round each segment; and three panels of
London, circumferential grooves,
sometimes inlaid. The group includes the applied lattice made from very thin brass
strips. It ls unclear whether the lattice was
above mentioned stylus from Brancaster,
made up before application to the stylus, or
with a silver lattice inlay. It is unlikely that it
the group ls the product of a single whether was laid strip-by-strlp dircctly
onto the iron. It came from a pit flll dated to
workshop. Dating ranges from early 3rd to
the mid 2d to mid 3d century.
4th century, wlth one example from Maiden
Castle (Wheeler L943,286, no 2) exhibiting
I'
the waisted eraser typical of group 2, and of Group 8 (Fig k)
possible 4th century date.
The group has straight slded erasens, slender
points.
handles and undifferentiated The
GROUP 6 (Fig 1, h) decoration ls slmple in all cases, consisting of
one or more grooves and/or inlaid bands.
The members of thls group, while somewhat
slmilar to Group 4, have parallel-sided Nine examples were identified, plus two
erasers, and a copper alloy bead (or a epxoasmsibpllee se, xfoaumr palrees .f rOomf t2hned -fcievnetu drya,t eodr
moulded lron bead intaid with copper alloy)
between the handle and point. Copper alloy posslbly 2d-century, contexts.
bands are present on the handles. Only four To conclude, this preliminary look at the
examples were identlfied, two from grouping of decorated styli has .already
Aldborough (Bishop 1996, 34, nos 195 and produced some evidence for the o<istence of
L97) and two from Elms Farm, one of which workshops producing a partlcular style of
is very similar to the Aldborough styli. The stylus (eg Group 1), and for a restricted date
second is rather more elaborate, range for some types (eg Group 2). It would
incorporatlng a polygonal-sectioned panel be interesting to compare the range of
and crimping on the bead, and is possibly decorative technigue and forms wlth those
not a member of this group. Both the Elms found on the continent. Are the very
Farm styli are from dated contexts: the one elaborately decorated styli such as Fig 1, i
(fig l, h) more closely resembling the continental imports? If anyone knows of any
Aldborough styli is late 4th century, whereas parallels in Britain or abroad for this
the more elaborate ocample is from an early exceptional stylus (or any of the other styli) I
Roman context. would be very grateful if you could contact
me.
GROUP 7 (Fig 1, I-r) Hilary Major
This group, of which there were nine Essor County Council Fielcl Archa@logy Group
Faifielcl C.ourt,
examples, has a swelling at the junction Fairffeld Road
between the handle and point, more Braintr@ CM7 sYQ
pronounced on some examples than otherc, hila ry. maior@esso<cc.gov. uk
rucerna Januery J##;'
References Manning, W H, 1985 Abbgue of the Romano-Eritish
Iron Tools, Fittings and Weapans in tfie British l{useum
Atkinson, M, & Preston, S, 1998 'The Late Iron Age
and Roman settlement at Elms Farm, Heybridge, Essex, Manning, W H, 1985a 'Ironwork' 46-58 in Draper, J.
excavations 1993-5: an interim report', Bribnnia 29, 1985 Fscavations by-Mr. H.k C-oopu on tfie Roman site
85-110 at Hitl Farm, Cratingtlwrpe, Essex, E Anglian Arciaeol
25
Wheeler, R E M, L943 Maiden C-astle, Dorset Rep Res
Comm Soc Antiq London 12 Neal, D S, 1987 'Excavations at Magiovinium,
Buckinghamshire, 1978-8A', Recards of Buclcs 29, I'I24
Bishop, M C, 1996 Finds from Roman Aldborough
Oxbow Monograph 55 Sparey Green, C & Hincfiliffie, J, 1985 'The small finds
from the L974 and L977 $(ctvations' ln Hinchlift, J.
Darling, M J, 1993 Caister-on-Sea: exavations by Exavations at Brancaster 1974 and 1977 E Anglian
Charles Green 7957-7955, E Anglian Archaeol 6O Archaeol 25, 4I-62
WAX SPATULA HANDLE FROM YORKSHIRE
On the Portable Antiquities Database there is The late George Boon suggested that these
an entry for a metal-detector find from Minerva bust handles were from razors
Beadlam, near Helmsley, ifi North Yorkshire. (1991, 30-2), but recent work has shown
Though described as a knife-handle in the that they, and other related handles
form of a helmeted bust of Mars, the object described by Boon, are from spatulae used
has been so clearly described that it can to smooth over sectiotrs, or whole pages, of
readily be identified as the ninth Minerva writing on wax tablets. Wax spatulae have
bust wax spatula handle from Britain:'In been the subject of considerable study on
form of a helmeted bust with rounded the continent over the past few years, with
shoulders. The head is rounded, while the three main forms, A-C, identified by Michel
body and slot are flat, apart from a heart- FeugEre (1995). Type B splits into two
shaped decoration on the breast. The face is subtypeS, and type A into five, of which
worn, the nose completely eroded, with one those with Minerva bust handles form
eye visible. At the bottom, there is a slot to subtype A5 (Fig 3).
take a blade. Length 45 mm.'
Unfortunately, no image was made of the
Helmsley handle when it was report€d,
though its description matches the handles
illustrated here perfectly and it is also the
right size. The heart-shaped decoration is
the aegis worn by Minerva and instantly
identifies this image as that of the goddess
rather than Mars, as originally supposed on
the database.
0 1cm
,,t
FDigra 1w: iAn5g hbanyd leA fnrodmre twhe H Vaiclalr.s CFoarpmy rsigiteh,t CCaammbbrrididggee. WFighi t2in.g C 1o9m2p3l.e Cteo pAy5ri gwhat xM sFpeautgudlare .f uSocma leO sinp rcimngse., after
University Archaeological Unit.
lucema January 200!
FeugEre's initial research produced a list of 5. Museum of London 12845, accessioned 1932,
eight examples of variant A5 from Britain acquired earlier, Prcvenance unknown. Undgren
(1995 , 326). Unfortunately, one was a 1978, 96, pl 68; PitB L979,95, no 215; FeugEre
duplicate. (I am grateful to Jenny Hall for 1995, 326, no 6 (and duplicated as no 5, fig 4, d).
clearing this up.) To the remaining seven can 6. Ospringe, Kent. Whiting L923, 66, pl oppmite
be added a recent find from Cambridge (Fig 65; Whiting et al 1931, pl 55, httom; Dunning
1), and now the Helmsley example. (A 1958, 16, pl 3; Boon 1991, 30, fig 4, gi Feuglre
Minerva bust from Cheshire, cited in lackson 1995, 326, na 7, fig 2, a. Here Fig 2.
& Potter 1996 as a parallel to the handle 7. Stonea Grange, C.ambridgshire. futter &
from Stonea, is of unusually thin curved Jackson 1982, fig 3, 4; Jackson & Potter 1996, fig
section and more likely to be an appliqu6.) LLz,95; FeugEre 1995, 326, oo 8, fig 4, e.
The revised British list is as follows: 8. Cambridge, Cambridgeshire. Crummy 2001.
Here Fig 1.
1. Kenchester, Herefordshire. Boon 1991, 3t, fig
4, h; FeugEre 1995, 326, no 1. 9. Helmsley, Beadlam, North Yorkshire. Portable
AnUquiUes Scherne: Finds ID yorymm4O4.
2. Kingscote, Gloucestershire. Henig & Paddock
1993, 89, fig 3, 9; Feugbre 1995, 326, no 2, ftg 4, In Britain the best-preserved A5 spatula, its
d. iron blade intact apart from some damage
3. Woodeaton, Oxfordshire. Kirk L949, 40, no 2, along the edges, came from a grave at
gl 4, F; Pitts L979, 96-7, pl 2L3,218; Boon 1991, Ospritrg€, Kent (Fig 2). Others from graves
fig 4, j; FeugEre 1995, 326, no 3, fig 4, b. (There on the continent have been found with
may be a second from Woodeaton: Kirk L949, pl inkwells and other writing eguipment, such
4, E; Pitts L979, pl 23, 2L7; Boon 1991, 32, as styli, and it is probable that at least some
footnote 76.) of the graves also originally contained writing
4. Silchester, Hampshire. Toynbee 1954, 81; Pitts tablets, long since decayed. The dating
L979, 97, no ZLg, pl 28; Feug6re 1995, 326, no evidence suggests that A5 handles belong to
4, fig 4, c.
,'A5
.
r.:, . r,
d,, ,,,,'H
ffi-
jll
i
, il f ,
irt
1,
ii
't ji i
ti'
ll
ii
li I :' '
li Ir:
.. ' _:, . . . i1ir!,lltll.,!.i, i'.!.:l' lil',ili'l'iri i ii.'.
Fig 3, Classification of wax spatulae.
Copyright M Feugere.
iutema -]dnuary .?##l
the 2nd century, perhaps running into the
early 3d century.
ReferenGes
The Yorkshire A5 handle is the most G Boon 1991 'Tonsar hunlglnas:. mzor and toilet knifr
northerly known in Britain, and it may be in antiquity', Britannia ?JL,2I'32
associated with the occupation of the villa at N Crumrny 2001 'A new Minerva bust wax spatula
Beadlam. The other examples are scattered handle from Britain', Instntmentum Bulletin l?,22
across the south, from London (probably), G C Dunning 1958 Guide tu Maison Dieu Museum
Cambridge and Stonea Grange in the east to
Kingscote and Kenchester in the west. The M FeugEre 1995 'Les spahrles i cire i manche figur€',
in W Czysz et at (eds), ftoviruialrfimisctrc Forsdutngen.
number from Britain compares to one each Fesfsdrrift trir Giinter Ulbert zum 65 GeburB,f;ig,32l-38
from Holland, Austria, Hungary, Italy, (Munich)
Portugal, Romania and Morocco, four from M Henig & I M Paddock 1993 'Metal figurines in the
Belgium, eight from France, and ten from Corinium Museurn, Cirencester', Trans Bristol
Germany (FeugEre 1995, 326-331). Gloucste$hire N&ao,l Soc 111, 85-93
Complete spatulas such as that from R Pl Jackson &TW Fotter 1996 Exavatiorlsat
Ospringe are evidence for literary, and the Stonea, C-am bridgeshire, 1980-8 5
use of the image of Minerva on a writing I R Klrk 1949 'Br?nzes from Woodeaton, Oxon',
utensil points to her function as goddess of Oxoniensia 1t[, 1-45
learning. There is, however, a sub-text to C Undgren 1978 Classial art frrns and Celtic
these handles. Detached from its iron blade, mutations
the handle may take on an independent life L Pitts L979 Raman bronze figurines of tfie
as an image of the goddess in any of her Catuvellauni and Trinovanfes, BAR Brit Ser 6O
aspects, of which perhaps the most T W Potter & R P J lackson 1982 'The Roman site of
important may be as goddess of the healing Stonea, Cambridgeshir,B', Antiquity 56, 111-20
arts, the Roman Minerva Medica, or in Britain I M C Toynbee 1964 Art in Britain under the Romarls
perhaps conflated with a local goddess, eg
W Whiting 1923 'A Roman cemetery discovercd at
Sulis Minerva at Bath. The two Minerva bust
Ospringe in 1920', Archaeol Qntiana ,61 65-80
handles (probably only one from a wax
spatula) found at the temple at Woodeaton W Whiting, W Hawley, & T May 1931 Report on tfie
ex@tration of tfie Roman cemetery at Ospringe, Kent
were probably votive offerings, as may have
been the Cambridge find, which came from a
small settlement with monolithic post
settings dating to the 3d and 4h centuries.
I would be grateful if anyone recognising
further e:<amples of wax spatulae with a
handle of any form, or spatula blades
without a handle, would get in touch with me
at the address below.
A bibliography of the continental material as
well as the British can be found in FeugEre
1995, while post-1994 references can also
be found at on www.instrumentum.net.
Nina Crummy,
2 Hall Road,
Copfonl,
Colchster CO6 lBN
Essex
n i n a. cru m mv@ n tlworld.com
Acknowledgments
Richarcl Hobbs & Simon Holms of the Portable
Antiquitiw Scheme, and Oaig Barclay of the
Yarkhire Museum, helped track down cletails of
the Helmsley handle. Michel Feugire genetously
allowd Figur* 2 & 3 to be reprinted here, and
the Cambridge Univercity Archaalqial Unit
kinrlly gave permission far Fig 7 to be usecl.
j' tt cg;ll E Jgn i,,ari. -ffrE
A gold amulet-pendant from Eaton
Constanti ne, Shropshire
In July 2000, a small gold object was shown something of power or of special significance
to Mike Stokes, of Rowley's House Museutn, to the wearer is still current in the wearing of
Shrewsbury, by its finder, Mr Niall Mennice. lockets enclosing a picture or other memento
Mr Stokes immediately recognised it as a (eg a lock of hair). The Graeco-Roman
tubular Roman amulet-case pendant. The amulet-case combines that concept with the
object was duly reported to the local Coroner age-old protective power of a religious
as reguired by the 1995 Treasure Ad, an symbol, the modern equivalents of which
inquest was held in December 2000, and the include crucifixes, Stars of David, and the
pendant was found to be treasure. It has like. A Romano-Egyptian portrait of the late
since been acquired by Rowley's House second century which shows a young boy
Museum and the finder has received the wearing a gold amulet-case on a plaited
appropriate reward. leather necklace is illustrated in the
catalogue Ancient Faces (Walker and
The object is a simple cylindrical tube of gold
Bierbrier L997, no 94)
designed to hang horizontally from its three
suspension loops on a chain or ribbon. The There is a late-Roman gold amulet-case in
circular plates closing the ends of the tube the Thetford treasurc, deposited at the end
are very neatly soldered, BS is the lengthwise of the fourth century AD or the beginning of
join of the tube. There is no decoration the fifth (lohns & Potter, 1983, no 30), and
other than the lightly reeded surfaces of the the catalogue discussion includes a full
loops. The object has been damaged by survey of the type, including a wide selection
crushing. Overall it is 2.5 cm in length, has of parallels. An earlier, but still valuable,
a diameter of 6 ffiffi, and weighs 4.4 discussion is to be found in the catalogue of
grammes. The metal is of extremely high the T6nEs, Algeria, treasure (early sth
purity; semi-guantitative X-ray fluorescence century AD) (Heurgon 1958, 57-9). As far
analysis carried out in the Department of as I am aware, until the discovery of the
Scientific Research, British Museum, reveals Eaton Constantine pendatrt, the Thetford
a gold content of 99 per cent. example and one from York (MacGregor
L976, fig 8, 72) were the only specimens
recorded from Britain.
In the absence of any association or
contextual information, close dating of the
Eaton Constantine amulet is difficult. There
is no decoration to provide stylistic evidence
pointing to earlier or later Roman taste, but
the relatively small size of the object may in
itself hint at an early Roman or Middle
Empire date rather than one in the fourth
century. Unfortunately, none of the small,
decorated examples appears to have any
Gold amulet-case pendant from Eaton Constantine, archaeological context, but several of the
Shropshire. Length 25 mm. Photo: British Museum
larger specimens do have associations dating
them to the third century and later, possibly
Tubular amulet-cases, often containing a suggesting some rudimentary typological
magical inscription or a substance credited progression based on size. Barbara Deppert-
with supernatural power, have a very long Lippitz assigns a late Hellenistic date to a
history. In the Classical world, they occurred very small (13 mm) amulet-pendant with
from Hellenistic to late-Antique times, but granulated ornament (Deppert-Lippitz 1985,
similar objects were also used in Middle no 35), though on stylistic considerations
Kingdom Egypt, in India, medieval Persia, alone, The Thetford amulet is nearly 40 mm
and various Islamic and Jewish contexts up long, and some other examples from hoards
to post-medieval times and, fio doubt, of irie 3rcr-5th centuries AD elsewhere in the
many other times and places. The idea of Roman Empire are considerably larger. One
wearing an item of jewellery that contains of the two in the T6n&s treasure (Heurgon