Iron Age Coinage in Kent
IRON AGE COINAGE IN KENT: A REVIEW OF
CURRENT KNOWLEDGE
DA YID HOLMAN
Over the past few years, Iron Age Kent has seen increasing research
on settlements, cemeteries, ceramics and numismatics. As part of this
research, efforts to improve the general level of co-operation between
Kentish archaeologists and responsible metal-detector users has led
to a very significant increase in our knowledge of several classes of
metal artefact. Perhaps most important has been the dramatic rise in
the number of Iron Age coins recorded, particularly from the east of
the county. In 1976, Allen commented that Iron Age coinage in Kent
was as yet poorly understood and that there must be many more coins
to be found and recorded (Allen 1976, 100). In September 1991, an
independent (part-time, unfunded) research project - the Kent Iron
Age Coin Project - was established by the writer in an attempt to
make a record of all new finds. By the end of 1999 a total of 1,318
previously unreported Celtic coin finds from the county had been
recorded, tripling the number to 1,974. Recording is continuing as
new finds are made. Full details have been forwarded to the Celtic
Coin Index, held at the Institute of Archaeology, Oxford.
The work of the Project has not only greatly increased the number
of coins known, but has also significantly altered the distribution
pattern of certain types, whilst confirming the previously mapped
distribution of others; many gaps have now been filled. As noted by
Haselgrove (1987, 213), single coin finds are of particular value in
understanding coin circulation patterns and these patterns have
provided perhaps the most beneficial results of the current study,
allowing new ideas and interpretations to be put forward.
AN OVERVIEW OF IRON AGE COINAGE IN KENT
An attempt at reconstructing something of the political history of late
Iron Age Kent can be made by using the coin evidence (Mack 1975;
205
DA YID HOLMAN
Van Arsdell 1989 - also abbreviated to VA). The large number of
additional coins now recorded might be expected to clarify matters,
but if anything the situation appears to be far more complex than
previously thought. The lack of any other contemporary evidence for
most of those individual rulers named on the coins means that any
attempt to establish the political history of late Iron Age Kent from
the numismatic evidence will always be open to question. Even the
names themselves are not always certain.
The general sequence of Iron Age coinage in Kent was established
by Mack (1953) and has been further refined by Rodwell (1976),
Haselgrove (1987) and Van Arsdell (1989). The most recent review is
that by Hobbs (1996, 9). The following overview is based on all the
numismatic evidence now available. All dates given are subject to an
error range of plus or minus ten years. Kentish coin types are found
scattered across much of South-East England, and occasionally beyond,
but they concentrate in Kent and can be seen as originating in
the County.
The development of Iron Age coinage in Britain can be sub-divided
into phases. Haselgrove has studied the subject in some detail (1987,
75-101 ). The phasing used in this paper, which is in general usage for
Iron Age coin studies, is based on Haselgrove's work, but includes an
additional category for the Kentish Primary Series potins and some
minor amendments to the other phases to take account of recent research.
Although in part contemporary, non-gold Continental imports
and potin coinages are kept separate from phases 1-5 as they appear
to have had a different function. The same applies to the individual
potin coinages.
Potin coinage - a total of 789 coin finds from Kent (British types
only), comprising:
Kentish Primary Series ('Thurrock type')[446 coin finds]
The French term potin is used to describe high tin content, cast
bronze coinages in both Britain and Gaul. The earliest coinage to be
produced in Kent, and indeed in Britain, was the series of potin coins
now commonly referred to as the 'Thurrock' type follo wing the
discovery of a large hoard of these coins at Corringham , near
Thurrock, Essex in 1987, but which is described here as the 'Kentish
Primary Series' in order to more accurately reflect its apparent area of
origin. The only other hoards of this type are both from Kent :
Folkestone (Holman, forthcoming) and possibly Gravesend. Oce
very rare, these coins are now by far the commonest Iron Age cows
found in Kent. On archaeological evidence (e.g. Haselgrove J 988,
206
IRON AGE COINAGE IN KENT: A REVIEW OF CURRENT KNOWLEDGE
TABLE 1. COINAGE PHASES USED
Phase Notes Date (+/-10 yrs)
P:KP Earliest British potin coinage 2nd century BC
(Kentish Primary)
P:FLI First potin coinage of 'flat' Late 2nd - l st
module (Flat Linear I) century BC
P:FLII Latest British potin coinage Mid 1st century
(Flat Linear II) BC
C (Potin, Imported base metal and Mostly c. 60-30
AE, AR) silver coinage BC
1-5 (AV) All imported gold coinage Late 3rd century
and earliest British - c. 50 BC
types
6 Kentish Uninscribed Series, c. 50-25 BC
other uninscribed
British types
7 Dubnovellaunos; Sam--; c. 25 - 1 BC
Vosenos; Tasciovanus
8E (early) Eppi!lus, early Cunobelin C. AD 1- 25
types
8L (late) Later Cunobelin types; c. AD 25 - 40
Amminus
9 Latest British coinage C. AD 40 -
Conquest
I 03), they are likely to be of mid-later second century BC date and
copy Southern and Central Gaulish prototypes. The precise prototype
is uncertain (Haselgrove 1993, 37), but most likely is a medium size
bronze of Massalia (Marseilles) or a close copy.
Flat Linear I {231 coin finds]
The Kentish Primary Series was superseded by a new series of potin
coins referred to here as 'Flat Linear I'; these had probably appeared
by I 00 BC and may be connected with the imported gold coinages
known as Gallo-Belgic A and B (Haselgrove 1995, 6) (see below). A
rare transitional type (VA 102) exists with features from both the
Kentish Primary and Flat Linear I series. Both Flat Linear I and II
potins (see below) have been classified chronologically by Allen as
classes A to L (I) and M to P (II) (Allen 1971 ). The Flat Linear I coins
were cast with thinner flans and a linear design, the earliest of which
207
DAVID HOLMAN
show a stylistic affinity with the early potin coinage attributed to the
Sequani tribe of Eastern Gaul (Haselgrove 1993, 37), imported examples
of which are known from Kent. Of the earlier classes of Flat
Linear I, the commonest type in east Kent is Allen type D although
type B is fairly well represented. Allen type L, which it has been
suggested may be concurrent with Caesar's activities in Britain and
Gaul (Haselgrove 1995, 5) is still the commonest of the later types.
Previous tentative suggestions that these coins originated in the
Medway valley (Haselgrove 1987, 151-157), based on the proportion
of early Flat Linear I types there, no longer bear scrutiny and origins
further east must now be postulated. Flat Linear I potins were
frequently hoarded. In Kent, hoards are known from Birchington,
Broadstairs and Lenham Heath.
Flat Linear II [Jl2 coinfinds]
The design of the Flat Linear I po tins became progressively simplistic
until their final degradation around the middle of the first century BC
by which time they had become smaller with childlike designs. This
latest potin series is referred to as 'Flat Linear II' and is generally
scarce in east Kent, except at Canterbury and Folkes tone. Their overall
distribution seems to indicate an origin primarily in west Kent and
beyond, with an increasing number coming from the northern Home
Counties. Evidence from excavations in Canterbury seems to indicate
that the later Flat Linear I and Flat Linear II potins circulated for
many years, perhaps even until after the Roman Conquest (Blockley
et al. 1995). In contrast, there is some evidence that the Kentish
Primary Series and earlier Flat Linear I types had largely ceased to
circulate before the end of the first century BC.
Potin coins probably ceased to be produced in the third quarter of
the first century BC, at which time struck, rather than cast, bronze
coinage started to make an appearance in Kent. This period may be
illustrated by a hoard reportedly found near Deal containing late
Class I and early Class II potins, supposedly with clay moulds, and a
Caption for Plate I opposite
I) AV Stater, Gallo-Belgic C (VA 48) 2) Patin, Kentish Primary series (VA !406)
3) Potin, Flat Linear I series (VA 129) 4) Potin, Fial Linear IT series (VA 139)
5) AR Unit, Kentish Uninsc. series
(Uncatalogued type) 6) AR unit, Dubnovellaunos (VA I 78)
7) AE unit, Dubnovellaunos (VA 180) 8) AR unit, Tasciovanus/Sego (VA I 85 I)
9) AE 1/2 unit, Sam--- (Uncatalogued type) 10) AE unit, Eppillus (VA 451)
11) AE unit, Cunobelin (VA 1973-1) 12) AR unit, Amminus (VA 194)
208
IRON AGE COINAGE IN KENT: A REVIEW OF CURRENT KNOWLEDGE
PLATE I
•
2 3 4
•
• -
5 6 7 8
••
9 10 l l 12
Some examples of Kenrish Iron Age Coins
209
DA YID HOLMAN
Kentish Uninscribed struck bronze coin (Haselgrove 1995, 6);
however, this discovery has not been verified and is not included in
the statistics owing to its dubious authenticity.
Early Gold (206 coin finds: all British and Continental, phases 1-5)
Circulating at the same time as the potin coins were a large number of
imported gold coins and their insular derivatives (Haselgrove I 987,
78-92). The very earliest imports are gold staters dating from the
third century BC which imitate coins of Philip II of Macedon (359-336
BC); three such imports have been recorded from east Kent, at Ringwould
(two) and Alkham. There are also a few half staters of a type
known as Gallo-Belgic XA, of late third or early second century BC
date. None of these presumably valuable coins are likely to have been
intended for circulation use.
It is only from the mid-late second century BC that gold coinage,
principally consisting of several series of coins classified as GalloBelgic
A to F (Allen I 960, 99- I 3 I), was imported in any quantity.
This has previously been considered to be the result of immigration
(Allen 1960, 98) although this idea has met with less favour in recent
years, with trade, mercenary activities and political alliances being
other possibilities (Fitzpatrick I 992, 16; Hobbs 1996, 9). According
to Caesar, there was immigration from Gaul, and this was probably
the period when the high king Di viciacus ruled Britain and Gaul
(DBG II, 4). Gallo-Belgic A and B types are scarce finds in east Kent
and are more commonly found along the north coast and in the west
of the County. Imports into east Kent only begin to increase with
Gallo-Belgic C in the early first century BC. Some of the Gallo-Belgic
C staters may be British in origin (Fitzpatrick 1992, 8, note 27) and
forgeries are also known. The following stater series, Gallo-Belgic E,
is very common and coins of this type were imported into Britain in
substantial numbers, with many examples known from Kent. Again,
some may be British copies (Haselgrove 1993, 39). Gallo-Belgic F is
unknown from Kent, although a single example of the late GalloBelgic
XF series inscribed 'CRICIRU' is recorded from Tonbridge.
There is also a large series of quarter staters known as Gallo-Belgic
DB and DC; dating from the early-mid first century BC, these are also
common finds in Kent and were subsequently copied. Caesar (DBG
V, 12) mentions the use of gold and bronze coins in Britain and these
are presumed to have been the Gallo-Belgic gold imports and the
potin coinages.
Following the Roman conquest of Gaul, native gold coinage was
largely suppressed and gold imports into Britain consequently ceased.
210
IRON AGE COINAGE IN KENT: A REVIEW OF CURRENT KNOWLEDGE
Interestingly, the distribution of imported gold coinages differs
greatly from the (mostly later) imported base metal coinage, suggesting
that they served different purposes, the gold perhaps primarily
used as bullion and wealth storage and the bronze as small change for
daily trade.
Kentish Uninscribed Series (I 81 coin finds)
From early in the second half of the first century BC, coinage struck
from dies rather than cast from moulds began to be produced in Kent,
the inspiration for this change coming from Gaul. The earliest of
these coins, lacking any attempt at inscriptions, are referred to
collectively as the Kentish Uninscribed Series, and they are known in
gold, silver and bronze. The large number of imports in circulation
had obviated the need for much native Kentish coinage prior to this,
apart from the potins. Early Kentish gold is generally very rare with
the exception of a series of quarter-starers with several varieties
based on Gallo-Belgic D (VA 147) and a slightly later, more debased,
type (VA 151).
Contemporary with these gold coins are the first native struck
bronzes. Probably starting around 40 BC, this is a large and varied
group with close affinities to the coinages of Northern Gaul commonly
found in east Kent, particularly those coins from the Ambiani and
Morini tribal areas which were closest to Kent and which provide
many of the prototypes for the Kentish bronzes (Haselgrove 1993,
43). The Kentish coins are considered to have been issued over a
period of perhaps 15 to 20 years, ending c. 25 BC, but their relative
chronology is difficult to determine and remains a matter of contention.
Some types may yet prove to be Gaulish, the lack of a detailed
study (and the non-recording of detector finds) in the relevant areas
of France (the Pas-de-Calais/Nord region) leaving many unanswered
questions. As an example of the difficulties of this period, one type
(VA 154-11) attributed to the Kentish Uninscribed Series on the basis
of a solitary specimen with an illegible reverse can now be shown to
be almost certainly later. Other examples are now known which show
that the reverse design is a capricorn, probably copied from a coin of
the Roman emperor Augustus and thereby post-dating the Kentish
Uninscribed Series. A clear specimen is still awaited, but this type is
likely to turn out to belong to the later inscribed series, perhaps of
Eppillus.
Silver coinage of the Kentish Uninscribed Series is very rare. Although
three or four different types have now been recognised, they
are all known from only a few examples. One uncatalogued type in
211
DAVID HOLMAN
particular shows close links to the bronze coinage, both in style and
in the use of common symbols in the field. The range of types, both in
silver and bronze, is now rather greater than those listed by Mack
(I 975) and Van Arsdell (1989).
Dubnovellaunos ( 127 coin finds : Kentish issues only)
From around 25 BC, inscriptions started to appear on the Kentish
coinage, suggesting that there were at least some people capable of
reading the names of those rulers who issued coins and wanted to
emphasise their rule. Even with names, the absolute and relative
dating of these coins is difficult and the problems and ramifications
are discussed by Fitzpatrick ( 1992, 25-26). The first chieftain in Kent
to issue coinage in his name was Dubnovellaunos (c. 25 - c. 5 BC) wh o
probably also held power in Essex for a time (see below). Some of the
later Kentish Uninscribed issues were probably also produced in the
early years of his reign and there is clear evidence of the same
die-cutters producing both uninscribed and inscribed issues.
Dubnovellaunos produced gold, silver and bronze coins, until a few
years ago all extremely rare but which are now far more widely known.
Opinions on whether the Dubnovellaunos who issued coins in Kent
was the same Dubnovellaunos who issued coins in Essex have differed
over the years. Mack ( 1975) regarded them as the same person,
but Van Arsdell (1989) disagreed. De Jersey (I 996) also considered
them to be different while Hobbs ( 1996) indicated a preference for
them being the same. This discussion shows no sign of coming to an
agreed conclusion, but a pointer in favour of them having been on e
person is provided by an uncatalogued bronze type which, although
apparently uninscribed, is stylistically attributable to Dubnovellaunos;
this type has been found in similar numbers in both Kent and
Essex and includes features of both series. A case for there being only
one Dubnovellaunos has also been put forward in a study of the gold
coinage (Kretz 1998).
Dubnovellaunos has often been considered to be the refugee
Dumnobellaunos named in the Res Gestae as a suppliant to Augustus
(Brunt & Moore 1967, 32), which would give a terminus ante quem of
AD 14. However, this is conjectural and the person named may
equally as well be the Dumnovellaunus named on certain coins of th e
Corieltauvi (cf. Hobbs 1996, 28) or someone not connected with an y
coinage series. Nor is there any way to date this event within Augustus'
long reign (27 BC - AD 14). The tenuous link with the Res
Gestae can be put in further perspective with the recent discovery of
two hoards found at Alton in Hampshire, which showed that the
212
IRON AGE COINAGE IN KENT: A REVIEW OF CURRENT KNOWLEDGE
Atrebatic (West Sussex, Hampshire area) chieftain known as Tincommius,
usually assumed to be the Tim--- mentioned therein, was in
fact named Tincomarus. This clearly illustrates the difficulty in
extrapolating the full form of personal names from shortened versions
(Cheesman 1998). Apart from the tentative Augustus connection,
Dubnovellaunos is unknown to history with the exception of his
coinage, although the quantity and range of this suggests that he was
a ruler of some importance. He may have been one of the 'friendly
British kings' recognised by Augustus before 7 BC.
SAM--- and Vosenos (48 coin finds)
The next coinage to appear in Kent was a series of silver and bronze
coins inscribed SA. The silver coins have so far been found in west
Kent only but the bronzes have a county-wide distribution. An
uncatalogued bronze half unit extends the legend to Sam---, still only
an abbreviation of a longer, probably personal, name. Contemporary
with these are the extremely rare coins of Vosenos (or Vodenos),
known from only a handful of examples in gold and silver, mostly
from east Kent. Both Sam--- and Vosenos can be dated to the very end
of the first century BC, perhaps overlapping with the latter stages of
Dubnovellaunos' reign. However, apart from their coinage, these
individuals are again unknown.
Sego (Tasciovanus) (22 coin finds)
Another inscribed series to appear in east Kent at the very end of the
first century BC or start of the first century AD was that bearing the
name Sego, known in gold, silver and bronze. Clearly associated stylistically
with the coinage of the north Thames (i.e. the area including
Hertfordshire and Essex) chieftain Tasciovanus and the Verulamium
(St Albans) mint, the appearance of these coins in Kent is difficult to
explain. The gold stater names both Tasciovanus and Sego. Perhaps
the most likely explanation is that Tasciovanus attempted to gain a
foothold in east Kent with its attendant advantages of closer Continental
trading links and these coins were minted for use here, either at
Verulamium or in Kent by an itinerant moneyer. Finds of gold coins
of Tasciovanus in west Kent have previously been considered a sign
of political dependence (Haselgrove 1988, 159). Sego may be an
abbreviated personal name, but an alternative interpretation is that it
was a title meaning 'the powerful one' (Celtic sego = powerful) as a
direct reference to Tasciovanus himself. This view takes into account
comparable inscriptions found on certain coins of Cunobelin and
Amminus (see below) which probably have the same meaning (Ho!-
213
DAVID HOLMAN
man 1999). There can be no question of any connection with the
Segovax mentioned by Caesar, who ruled much earlier.
Eppillus (l 0 1 coin finds : Kentish issues only)
The next ruler to issue coins in Kent was Eppillus (c. AD 1-15). The
prototypes for his coins appear to be mostly near-contemporary
Roman issues of the emperor Augustus. Associated with the northern
Atrebatic centre of Calleva (Silchester) by a separate coin series
produced there, on which he styled himself 'rex Calleva', Eppillus
may have held both seats of power simultaneously for a time before
losing Calleva (Bean 1991 ). Eppillus described himself as a son of
Commius, a (presumably) Atrebatic chieftain possibly descended
from the Gaulish Commius, 'the Atrebatian • mentioned by Caesar. At
one point during his reign, Eppillus appears to have formed an alliance
with Tincomarus and Verica, who also describe themselves as
sons of Commius, to judge from silver coins (VA 442, 443) apparently
naming them in addition to Eppillus (Bean 1 991 ). Eppillus is
again unknown to history except from his coinage. The comments of
Caesar (DBG V, 14) on polyandry may be of some relevance in any
discussion of dynastic issues.
The Kentish series of Eppillus is varied, with gold, silver and
bronze issues known, but it is distinct from his Calleva series which
did not circulate in Kent. The gold and silver coins are still very rare,
but the bronzes are now known in some quantity. There are four catalogued
types of bronze for Eppillus with three or four other extremely
rare uncatalogued types including half-units. A number of types we re
produced showing a Victory figure which has been interpreted by
Nash (1987, 137) as an indication of a political success. The same
would perhaps appear appropriate for Cunobelin (see below).
Cunobelin (201 coin finds : all types)
Sometime around AD 15, Eppillus appears to have been replaced by
Cunobelin (c. AD 10-40), a leader of the north Thames tribes based at
Camulodunum (Colchester) and probably the most powerful man in
Britain at the time. He is attested by Roman sources as 'Britannorum
Rex' (Suetonius, Gaius, xliv). Cunobelin, who described himself as a
son ofTasciovanus, was apparently successful in his attempt to annex
the whole of Kent to his kingdom and many interesting features are
found in his Kentish coinage.
The coinage of Cunobelin has a wider and more even distribution
across Kent than any other of the inscribed dynastic issues. For the
first time, inscribed gold coins appeared in Kent in some quantity,
214
IRON AGE COINAGE IN KENT: A REVIEW OF CURRENT KNOWLEDGE
most bearing the well-known 'corn ear' obverse design (e.g. VA
2010, 2025). However, it is the silver coinage of Cunobelin which
provides particular interest for Kentish coin studies. Rodwell (I 976,
274) noted that no silver coins of Cunobelin were known from Kent
at that time; this situation has now changed to such an extent that a
probable east Kent mint can be deduced as producing silver (and
bronze) coinage. One type (VA 2067) has a clear Kentish distribution
and is far removed from the well-executed issues of the Camulodunum
mint with its crude engraving, degenerating almost to 'matchstick'
style. The same die-cutter was also responsible for an uncatalogued
silver issue of similarly crude style, again with a clear Kentish
distribution. Another silver issue of Cunobelin which may possibly,
although less certainly, be associated with Kent is the late issue
bearing the enigmatic legend 'SOLIDV'; this is found in east Kent
but is extremely rare. 'SOLIDV', the meaning of which is unclear,
could refer to a mint site (Rodwell 1976, 276), a local official, or even
the name of the moneyer.
Fitzpatrick ( 1992, 26) has suggested, based on distributions, that
certain coins of Cunobelin were struck for use in particular regions,
such as those inscribed CVNO/T ASCI F found in Hertfordshire. It
now appears that certain bronze issues of Cunobelin were probably
struck in Kent and were primarily used there. As with the Kentish
silver coins, these bear the legend CVN or CVNO on both sides and
make no reference to the mint at Camulodunum (CAM). The bronze
type VA 1981 is a strong candidate as a Kentish mint product. Another
type (VA 1989), bearing on its obverse a stylised ship design,
may also be a Kentish issue with the reverse legend SE meaning
'powerful' (Holman 1999). One obverse die of this type shows that
the legend was originally intended to read 'CAMV', but this was
changed to 'CVNO' (Muckelroy et al. 1978, 439-444), suggesting
that although this die was cut by someone familiar with the Camulodunum
mint, that was not the intended mint for this type. Some other
coin types of Cunobelin were minted at Camulodunum primarily for
use in Kent to judge from their distribution, and these probably date
from soon after his acquisition of Kent; the most common of all
struck bronze Iron Age coins found in Kent, VA 1973-1, one specimen
of which appears to bear a CAM mintmark, is a clear example.
There is some evidence that this type was mass-produced within a
short time span with one die showing signs of having been crudely
repaired and the striking frequently being off-centre. All of the
bronze coins mentioned here feature a Victory figure in the design
and it is possible that this is a reference to Cunobelin gaining power
in Kent. Rodwell (1976, 274-276) has previously suggested 'wreath'
215
DA YID HOLMAN
and 'Victory' types were produced following the annexation of the
Verulamium region and there is no reason why this should not apply
to Kent also. A possible parallel for this under Eppillus has been
mentioned above.
It has been noted previously (Fitzpatrick 1992, 27) that late issues
of Cunobelin heavily outnumber his earlier issues, a situation found
on the majority of sites in the north Thames region and clearly
illustrated for Northamptonshire (Curteis 1996, 27). It appears
however that the situation in Kent is very different, the early issues
forming the bulk of Cunobelin' s coinage in the region. Possible
explanations for this will be discussed elsewhere including the
possibility that some of the Kentish mint products are later than
generally supposed (Holman & Parfitt, forthcoming).
Amminus (25 coin finds)
The final Iron Age coins to be produced in Kent were those bearing
the name Amminus, probably Adminius, a son of Cunobelin who fled
Britain to seek refuge under the Roman emperor Gaius about AD 39
(Suetonius, Gaius, xliv). These are known only in silver and bronze
and are still very rare. They bear some comparison with the later
issues of Cunobelin, in particular the silver unit VA 192, the obverse
of which copies a silver issue of Cunobelin. One silver type of
Amminus bears the legend SEC, again probably meaning 'powerful'
(Holman 1999).
However, it is the appearance of an apparent mint signature,
DVNO, which provides the main interest of Amminus' coinage. This
has been attributed to Canterbury (Haselgrove 1987, I 43 ), but there
must be strong objections to this since Canterbury was known as Durovernum
(fort by the alder-swamp) and is unlikely to have ever incorporated
a DVNO- element. DVNO would seem to refer to a site on a
hill (Celtic dun; Rivet & Smith 1979, 344) and the principal distribution
of these coins lies to the east of Canterbury. The site of DVNO
must therefore remain a matter of conjecture (Holman 1999); Bigbury
is a possibility, but there is no evidence of late occupation here from
the limited excavations undertaken to date.
Post-Conquest
Following the Roman Conquest, native Iron Age coinage in SouthEast
England ceased to be produced, but it continued in circulation
for perhaps another generation (Haselgrove 1993, 62) before
probably being totally replaced by the new Roman coinage when te
coin supply increased after AD 64 although, as noted by Curteis
216
IRON AGE COINAGE IN KENT: A REVIEW OF CURRENT KNOWLEDGE
(1996, 33), conclusive evidence is elusive. An analogous situation
existed in Gaul for many years after its conquest (Nash 1978, 13-14,
27), as reflected at Chilly and Digeon although these sites were
probably sanctuaries (Allen 1995, 33). The displacement of Iron Age
coin types, particularly later issues, is primarily attributed to the
Roman army (Haselgrove 1993, 50, n.66). One likely reason for the
continued use of native coinage for several years after the conquest is
that the supply of Roman coinage was initially very restricted owing
to the failure of the Roman authorities to supply the quantity required
in either Britain or Gaul. Copies of Claudian aes were produced in
some quantity on an unofficial basis, but it is unlikely that they alone
would have filled the void (Boon I 988, 118-124 ). The native coins
would have been expected to continue to circulate in this situation,
perhaps as an emergency coinage until the newly conquered province
could be properly organised and supplied.
Claudian coinage has been found at a number of sites across Kent.
Many of these are contemporary copies, as is usual. Although of
larger size than Iron Age coins, they appear to be scarcer as metaldetector
finds. Pre-Claudian Roman coins are also present but none
can be shown to have derived from a pre-Conquest context and their
frequently extremely worn condition suggests a long life for many of
them. Apart from the Richborough conquest period losses, the only
certain Claudian coin deposit from Kent is the Bredgar hoard containing
a number of aurei and dated to AD 43.
Other pre-Conquest coins (217 coin finds)
Apart from the many Gaulish imports and coins minted by other
British tribes, e.g. Iceni, Corieltauvi, etc., a small number of early
coins manufactured and used in the Mediterranean region are known
from Kent. Although some may be connected with the Roman Conquest,
brought in by the army, the numbers are such, particularly for
certain types, that the possibility of a much earlier importation date
cannot be discounted. In particular, a bronze issue of Carthaginian
Sicily and a bronze issue of Ebusus (Ibiza), both of fourth to second
century BC date, occur fairly regularly (ten and three respectively).
The reason for such coins appearing in Kent is uncertain, as is the
date of their importation, but mercenary activity and trade are both
possibilities. The idea that they may have been lost in the eighteenth
century (Laing 1968, 15) can safely be discounted in the case of the
Kentish finds.
217
DA YID HOLMAN
THE DISTRIBUTION OF IRON AGE COINAGE IN KENT
Kent is a well defined geographical area, being a large county with a
long peninsular coastline. Its closeness to mainland Europe makes it
a likely place to find evidence of Continental influence. The name of
Kent derives from the Celtic Cantion, later Romanised to Cantium,
which translates as rim or border, i.e. the rim of the land (Rivet &
Smith 1979, 300). Topographically, the elevation of Kent ranges
from marshland around the coast to chalk downland reaching in
excess of 250m above OD in the extreme west. The spine of the North
Downs cuts across the central part of the county, separating the north
from the south. The Downs are cut by the three major rivers, the
Darent, Medway and Stour, and numerous smaller watercourses give
rise to downland valleys. There is a broad range of soil types and the
countryside is productive, both for woodland and farming. Everitt
( 1986) has classified Kent into a number of distinct landscapes. It has
been found that the vast majority of Iron Age coins come from his
'Foothills' and 'Downland' areas although an attempt to isolate
separate regions of coin circulation by this method failed to produce
any convincing results. Nevertheless, the geography of Kent provides
a useful basis for a division based on the rivers.
A general description of Kent is given by Caesar who wrote that it
was by far the most civilised part of Britain with a lifestyle little
different from that in Gaul. In the archaeological record, many details
of the late Iron Age in Kent are still unclear. There were apparently
few hillforts in the county, the most well known being Oldbury and
Bigbury, and valley oppida are presumed, on archaeological and etymological
evidence, to have existed at Canterbury, Rochester and
Loose/Boughton Monchelsea. Much of Kent was densely populated
and covered with homesteads according to Caesar (DBG V, 12). It is
becoming increasingly clear, however, that there were a number of
other major settlements whose names are now lost, some of which
have only come to light since the advent of metal detecting.
Archaeologically, Kent falls within the region covered by the core
area of the late Iron Age Aylesford-Swarling cu Iture (Cunliffe 1991,
133). It was inhabited by the tribes known collectively as the Cantiaci
(Rivet & Smith 1979, 299), although this name is not attested
by Caesar. Caesar (DBG V, 22) mentions four kings in Kent in 54
BC, namely Carvilius, Cingetorix, Segovax and Taximagulus, but
does not link them to any specific tribe; from this, it has long been
postulated that the county was in some way sub-divided into four
separate kingdoms (Cunliffe 1991, 146). The possible boundaries of
these kingdoms have seldom been considered in detail. Cunliffe
218
I. Alkham
2. Bigbury
3. Birchington
4. Boughton
Monchelsea
5. Boxley
6. Bredgar
7. Broadstairs
8. Canterbury
9. Chilham
10. Deal
11. Elham
12. Faversham
13. Folkestone
14. Frinsted
15. Gravesend
16. Higham
Land above 100 metres O.D.
I J
0 km 10
10
22
English Channel
Fig. I Regions and Zones of Kent (showing places mentioned in the text)
tl'l
(")
0
z
►
C)
z
z :-:l
►
17. Keston I
18. Lenham Heath ...::;
19. Oldbury
20. Radfield
21. Richborough
22. Ringwould
23. Rochester
24. Ryarsh
25. Springhead
26. Stoke
27. Tonbridge
28. Tunstall
29. W esterham
DAVID HOLMAN
( 1982, 48) has suggested that the principal rivers of Kent, the Stour,
Medway, Darent and Thames served as central routeways with kingdoms
extending on either side of them. An obvious alternative is
that rivers and other topographical features served as the boundaries
(Rodwell 1976, 279; Haselgrove 1987, 137). Canterbury and Rochester,
both major late Iron Age settlements, are situated on the principal
rivers. For the purposes of this study, Kent has been divided
into four topographical regions using the sea and principal rivers as
the boundaries. These provide convenient units within which the
observed differences in coin distribution can be studied. These
regions are as follows:
Region A : east of the River Stour, including the Isle of Thanet;
Region B : between the Rivers Stour and Medway, with a possible
sub-division along the line of the North Downs
escarpment (north and south zones);
Region C : between the Rivers Medway and Darent, subdivided
as Region B
Region D : west of the Darent.
The coin distributions in each of these regions show some reasonably
well defined, although not numismatically distinct, differences.
Since the number of coin finds from east Kent (Region A) is greater
than the combined total of the other three regions, this area has been
examined in greater depth and a detailed study is currently under
preparation. A summary of the numbers of coin finds from the four
regions of Kent for each phase is shown in Table 2 and i n histogram
form below (Figs 2 & 3). The numbers of coins of each metal are also
shown, as proposed by Rodwell (l 976, 314 ); plated coins, i.e. contemporary
forgeries, have been treated as being of the metal they purport
to be. The phases have been explained in Table 1 above.
It is instructive to compare Regions A, B and C in their entirety, for
a number of differences are apparent. Firstly, the ratio of Kentish Primary
potins falls sharply west of the Stour Valley until they account
for only 7 per cent of the recorded Iron Age coinage west of the
Medway in Region C. Recent detector finds confirm this pattern. In
Region A, Kentish Primary potins are still the most frequent finds; in
Region B, they are also evident, principally immediately to the south
of the North Downs escarpment, but less frequent; in Region C, they
continue to be markedly scarce compared with further east. This strongly
suggests that east Kent was the area of origin of these coins.
In contrast, Flat Linear I potins are more frequent in Region B than
in Region A, but they reduce to the Region A level west of the River
220
IRON AGE COINAGE IN KENT: A REVIEW OF CURRENT KNOWLEDGE
TABLE 2. COIN FINDS FROM KENT BY REGION, ZONE,
PHASE AND MET A L
Phase A B BN BS C CN CS D Total
P:KP 406 25 2 23 13 9 4 2 446
P:FLI 157 45 16 29 20 17 3 9 231
P:FLll 82 14 12 2 4 2 2 1 2 112
C 116 8 2 6 2 2 0 1 127
1-5 86 63 31 32 46 21 25 11 206
6 132 30 11 1 9 28 22 6 6 196
7 155 37 21 1 6 41 31 10 10 243
8E 199 36 15 21 28 19 9 5 268
8L 5 4 4 1 3 8 8 0 2 68
9 1 0 0 0 2 2 0 0 3
Un-ph 57 9 3 6 7 4 3 I 74
Totals 1445 271 114 157 199 137 62 59 1974
Potin 685 87 31 56 38 29 9 23 833
AE 524 60 18 42 76 59 17 7 667
AR 101 17 5 12 19 16 3 7 144
AV 135 107 60 47 66 33 33 22 330
Notes. AE=struck bronze: AR=Silver: A V=Gold. The Mediterranean coins
discussed on page 217 are not counted above, not being Celtic. lndividual
coins are each counted as a single coin find. Hoards have generally been
counted as one coin find; hoards including several coins from more than one
phase have been counted as having one coin find for each of those phases, on
the basis that such a hoard may be a deposit over a long period of time, e.g.
Stoke. The figures relate to Kent finds only and do not include Kentish coins
found elsewhere.
Medway in Region C. West of the Stour Valley, they exceed by some
margin the number of Kentish Primary potins, a fact not immediately
evident on the ground because of the lack of recorded finds in the
large B lean Forest area west of Canterbury, but increasingly clear
around the Medway Valley.
Flat Linear II potins show little variation between regions east of
the River Darent, always remaining at a low level. The Region A
figure is exaggerated by the large number of finds at Canterbury and
Folkestone which account for the majority of Flat Linear 11 potins in
that region. This may be seen to lend weight to the possibility that
Flat Linear II potins were, at least in part, a north Thames coinage
221
REGION A REGIONS
50.00 40.00
45.00
35.00
40.00
35.00
30.00
30.00 25.00
";/125.00
"/II 20.00
20.00 15.00
15.00
10.00
10.00
5.00
5.00
D.. 5 0 It) CD ... 11.1 15 .1 co C) z !2 :::; 5 0 '? CD .... !M .J 0, z 11. 11.
co a: 11. a: a: a: D.. 0 tl a: PHASE/METAL PHASE/METAL D.. >
<
Iv
s
tv ::i::
Iv ZONEBN ZONEBS
0
60.00 40.00 "".,,·.,.-.,u,,....· .,• .,, ,.,..,... ......- .·--· ....... ,..-......... _,,.., ........i i 35.00
!
50.00
30.00
40.00 25.00
;#!30.00
'#-20.00 ·
20.00
15.00
10.00
10.00
5.00
D.. :::; :::; 0 "?