THE CANTERBURY-RICHBOROUGH ROMAN ROAD:
A REVIEW
P.H. PANTON
INTRODUCTION
Margary's account in 19671 of the Roman Road to Richborough from
Canterbury begins 'This road leads east from Canterbury
(Durovernum) and must have been one of the most important in
Britain, for the port of Rutupiae was one of the principal points of entry
from Gaul, and perhaps the most used of all at that time'. Nevertheless,
despite its paramount importance, the exact line of the road has yet to
be established beyond reasonable doubt. This present paper reviews
efforts made over the years to point to a solution to the problem, and
attempts to summarise conclusions which may be drawn from the
present state of knowledge.
A basic problem is that the geography of the Wantsum area has
changed radically since Roman times; the Wantsum has disappeared
and Rutupiae is no longer an island. Dowker in 18722 published a
Map of the Coast round Richborough in Roman times (see Fig. 1)
and there have been a number of variations of such a map published
subsequently. Perhaps the most authoritative is that of J.D. Ogilvie3
in 1977 (see Fig. 2). From these conjectural maps it may be seen
that, however excellent an harbour and anchorage Rutupiae may
have provided, the establishment of a direct road to Canterbury from
the island through tidal marsh land with deep and changing inlets,
must have presented severe problems to Roman engineers.
1 Ivan D. Margary, Roman Roads in Britain, Rev. Ed. London, 1967, 36.
2 G. Dowker, 'Account of the Society's Researches on the Roman Castrum at
Richborough', Arch. Cant., viii (1872), 1-17.
3 J.D. Ogilvie, 'The Stourmouth - Adisham Water Main Trench', Arch. Cant., xciii
(1977), 91-124.
1
P.H. PANTON
...
,. <
,: ...
2
l0
.i,!,':4,
j-·\'!j ,f, :-' t'• "J'•f :ij:j I
f... d , "14
l}, :h A -0::1.,. ' '+ .
i ROMAN ROAD
0
DUBRIS
' ·,r';:i j ,. :, z
. · ·,:cl•. · , i ,,' ::' . ·. •i/h'•&· 'ls ffi ' •'t.J ...
r:·, t·.
,,! "' . is·'. ::· 0:,. ,, ,·•
··,·
o . 'l,"i ••• :c
0 ; : :t ' CJ g
t- ·'; .,, - ,' ' g §
Q
<
0
0:
z
<
:& 0
0:
·<11·1 ·' 0:
.J,J'
-' ,j;,• - "
' : J
z "5 :,
0 '\j
0:
.."'... i 0 u
1
t
R
......
..:
0 Cl
.; !::
::!
"'
s ·.o
§
s
0
.s .r:I b.()
0::I
0
'O
0::I ..."'.... Cd
CJ0
'v"-")' -p il-l' i.s h.. .P, a u.n.c!.a. l'"l e. & -·············
£ S Cd.e.
§. ¼. -rmz.les.
)
Fig. 6. Roman road from Richborough to Canterbury (Source: Knox, I 942).
:-
(")
I
c::::
i
I
;o
0
►
;
P.H. PANTON
the consequent production of the long straight mound, which cannot
therefore be of Roman construction.
Regarding the 'Margary line' to Ash, Ogilvie noted that support
would be added by finding a causeway at Cooper Street, where the
route crosses a marsh similar to but narrower than the Fleet Channel.
Although no structure was found, the deep dyke crossing the neck of
the marsh yielded several pieces of Roman tile and flint similar to those
from the Fleet Crossing.
STOURMOUTH - A DISHAM WATER MAIN
More recent work and discoveries in the Thanet area seem rather to
have complicated than eased the problem. Ogilvie in his 1977 paper3
reported on the watching brief on the excavation of a Stourmouth -
Adisham Water Main Trench. (A map of the line of the Trench is given
at Fig. 7). While the work yielded evidence of three new sites -
Neolithic, Bronze Age and Medieval - the original objective was to
look for traces of the Roman road, and although the trench was
examined for all but about 200 ft. of its length, no trace of the road was
found. Ogilvie noted that although the O.S. maps mark the A275 as a
Roman road, the absence of deep road metalling, and the presence of
medieval pottery below the road make this unlikely. Indeed, it must be
that since along the length of the water main no trace of a Roman road
was found, the validity of both the Margary route via W ingham,
Shatterling and Ash and the 'Andrews Line' must be in some doubt.
Ogilvie also noted that the area round Preston Village must have
been an important centre, continuously inhabited for a considerable
period, and he concluded that 'this populous centre must have
significance when considering the layout of Rutupiae and the Roman
routes from Richborough to Canterbury'. In this connection, it should
be noted that the water main trench stops about a mile north-east of
Preston, most probably just short of the proposed 'Knox line'. The
trenching, therefore, and lack of evidence from it of a road, does not
affect the 'Knox line'.
FURTHER EVIDENCE F OR THE 'ANDREWS LINE'
In 1965, officers of the Ordnance Survey, working from Maidstone,
reported on a previously recognised stretch of agger of the road in Pine
Wood. 12 This seemed to be a continuation of the straight line of the
12 Archaeological Notes from Maidstone Museum, Arch. Cant., lxxx (1965), 279-80.
10
THE CANTERBURY-RICHBOROUGH ROMAN ROAD: A REVIEW
/ LINE OF TRENCH
ONE MILE
Milne
/ ELMSTONE
Longmete Road
HOADEN
SANDWICH
->
OURLOCK STREAM
Fig. 7. Stourmouth-Adisham water main trench (Source: Ogilvie, 1977).
11
P.H. PANTON
road from Canterbury, where it reaches the outskirts of Littlebourne. At
this point the modern road takes a wide swerve to avoid Pine Wood
(N.G.R. TR 196578). The O.S. workers found that the stretch of agger
'incorporated a change of alignment which suggested that the road ran
along the ridge towards Wickhambreaux, to cross the Little Stour in the
vicinity of Deerson Farm, and thence direct to the Wantsum crossing
immediately west to Richborough. A good route, but no confirmatory
evidence was found.'
Work (so far unpublished, but in the possession of the Canterbury
Archaeological Trust) in the 1960s and 1970s, centred on lckham
Gravel Quarries (N.G.R. TR 233590) has yielded evidence of RomanoBritish
industrial sites, including water mills, and timber lined wells.
Also found in the area was evidence of road alignments, including
metalling and side ditches built on pegged brushwood. It would seem,
however, that the road structures found may not have been substantial
enough to indicate a major route. Evidence was also found of flints,
perhaps of a ford or bridge, across the W ingham River, period
A.D. 100-400, and of a zig-zag road up the incline on the Great
Wenderton side of the Wingham River. The location of these features
are more or less in alignment with the direction change of the agger
reported in Pine Wood, and towards the causeway from Richborough
Island; the 'Andrews line'.
C.A.T. WORK 1992
In 1992, work by the Canterbury Archaeological Trust at Each End
near Ash revealed a stretch of Roman road with associated settlement
of quite early date (N.G.R. TR 585303). The alignment was roughly
NE, heading towards Richborough. The location of the site is, however,
too far east along the Ash-Sandwich road for the line of the excavated
road to be fitted into Margary's or Winbolt's postulated route from Ash
to Richborough. The metalling may have been too light for a major
road, and in any event, the line of the road would most probably have
met tidal marsh land before reaching Richborough. The discovery,
therefore, may not be relevant to the central question of the main
Roman route out of Richborough to Canterbury.
FIELD WALKING 1994
On 5 January, and on 30 April, 1994, attempts were made to verify on
the ground evidence of the road as noted by previous authors. The field
walking party consisted of Mr J. Bradshaw (who had been largely
12
THE CANTERBURY-RICHBOROUGH ROMAN ROAD: A REVIEW
responsible for the Roman sites at Ickham Gravel Quarries); Mr and
Mrs. A. Mauduit, local residents at Ickham; and the author of this
present note.
On the 5 January reported sites at Littlebourne and Ickham were
walked over, and on the 30 April, the reported routes from Ash to
Richborough Fort were traversed.
Attempts to find the agger in the Pine Wood, Littlebourne (N.G.R.
TR 196578), as reported, 13 failed. It would seem that parts of the wood
have been grubbed out and put to the plough since the original
observations were made in 1965, and this may account for the present
lack of evidence.
On the other hand, a stretch of track was clearly to be seen in a field
north of Seaton Mill, between the Wickhambreaux road and the Little
Stour (N.G.R. TR 225588-227589). Further east, (N.G.R. TR 235593)
evidence for a crossing of the Wingham River was still visible
including some old timbers driven in the banks on both sides, which
may provide evidence for a bridge. Across the Wingham River, on the
Great Wenderton side, evidence of the track zig-zagging up the incline
could still be seen. That there was Roman activity in this area, roughly
on the 'Andrews line' is clear, but there may have been a local road
only associated with industrial use in the area.
On 30 April, 1994, an attempt to trace Margary's suggested route
from the Ash Road to Richborough via Cooper Street and Fleet Farm
was made. No definite identification was made of features mentioned
by Margary, though it has to be noted that parts of the area have been
much changed by the new Sandwich by-pass road. It was, however,
possible to identify the location of the line of the Fleet Causeway as
given by Ogilvie.
In summary, evidence for Roman activity on the Andrews line in the
lckham area was confirmed, but roads may have been local only,
associated with industrial activity in the area. The forays threw no new
light on the Margary line, apart from locating the line of the Fleet
Causeway as evidenced by Ogilvie.
DISCUSSION
During its four centuries of Roman occupation the Rutupiae site was
put to various uses, and it cannot be excluded that at different times in
its history different routes may have been found suitable and
appropriate.
13 cf. n.8.
13
EH.PANTON
In the first century or so of Roman occupation, the Wantsum
channel, guarded at the ends by Reculver and Richborough forts, was
the route into Britain from the Continent, and Richborough itself was
the chief port of entry. Among the remains of the Roman encampment
of first and second centuries A.D., is a massive concrete platform, some
100 x 80 ft. in dimension, on which is based a large cruciform
foundation.14 Various theories have been advanced to account for these
massive foundations; a Pharos or watch-tower of unusual height, or a
triumphal arch of some description. An imaginative suggestion by
Knox15 fits the plans for a praetorium at Lambese (Algeria) onto the
platform at Richborough, with the cruciform structure, raised a few feet
above the platform, fitting neatly inside the postulated praetorium. The
whole would have provided a triumphal ceremonial entry into the
province of Britain, with the cruciform space providing a platform on
which dignitories could stand to receive visitors in a ceremonial
fashion.
Whatever were the structures on the concrete and cruciform bases, it
can hardly be doubted that the existence of these massive bases implies
buildings of importance matching the role of Rutupiae as chief port of
entry in the Province. We must query whether, given such an
impressive entrance, it is likely that the road from the island would take
such a devious and anticlimatic line as to Each End and Ash as
proposed by Winbolt, Margary and others. Is it not likely that a straight,
impressive route would be preferred, even though that would present
problems (though not perhaps insuperable ones) to Roman engineers in
negotiating the marshy areas through which it would have to pass?
In the late third century, after a period of decay and obvious decline
in importance, Richborough entered a period of use as a fort of the
Saxon Shore, part of the defence of the civil zone of the province
against raiders from the sea. In this period, the emphasis would be
more on enabling the Saxon Shore fort to draw on the resources of and
provide protection for the local populace rather than acting as a
triumphal port of entry. The need for a correspondingly impressive
route out of the island may therefore have diminished, and the original
straight road (if it had existed) may indeed have fallen into disrepair or
have been overwhelmed in part by tidal waters. An easier but less
direct route might then be preferred. In the period of decay and
abandonment in the late fourth and early fifth centuries, upkeep of
difficult routes could scarcely be expected.
14 See Dowker, op. cit., and G. Dowker, 'On the Cross and Platform at Richborough',
Arch. Cant., xxiv (1900), 201.
is C. Knox, 'Richborough - Lambese', Arch. Cant., xliv (1932), 165.
14
THE CANTERBURY-RICHBOROUGH ROMAN ROAD: A REVIEW
What all this may imply is that different routes at different times may
have been constructed, according to the use and importance of the site,
and that, at times, more than one route may have existed. On these
grounds, none of the three main routes proposed - the Andrews line,
the Knox line and the Margary line - can be ruled out as improbable.
However, evidence for Knox's route to Upstreet is slight and
disputed, and would have presented difficult engineering problems. For
Andrews' direct route equally, engineering problems would have been
difficult. This route, skirting Littlebourne and going through lckham
and Wickhambreaux, has some evidence of aggers and Roman sites
along the line, but lacks support of other features, such as parish
boundaries, footpaths, etc. However, it may be that this route, if it
existed, had effectively disappeared before parish boundaries began to
be defined in Saxon times. Margary's route, following in the main
existing routes to Ash and Each End, and thereafter relying on evidence
of aggers and embankments is perhaps better attested. Note must be
taken, however, of the monitoring of the water main trench from
Adisham to Preston, which found no evidence of a Roman road
crossing the line of the trench, and therefore casts doubt on both the
Margary and the Andrews line, but no evidence against the Knox line.
SUMMARY AND CONCLUSION
In sum, while Margary's proposed route remains the most plausible, it
does not seem possible to dismiss the Andrews direct line entirely. This
may have been the first route used when Rutupiae was the first port of
triumphal entry to the province, subsequently falling into disuse and
disrepair when the importance of Rutupiae declined before rising again
as a Saxon Shore fort. Support for either of these routes is diminished
by the negative findings from the Adisham - Preston water main. The
Knox line is not affected by the water main evidence, but it remains
comparatively theoretical and speculative.
The problem is unsolved and is likely to remain so, in default of hard
and significant evidence.
15
