The Roman Station of VAGNIACIE at Springhead, Gravesend

( 177 ) THE ROMAN STATION OE VAGNIACJ3 AT SPRINGHEAD, NEAR GRAVESEND. BY GEORGE M. ARNOLD, ESQ. GIBBON,* in his able summary respecting the union and internal prosperity of the Eoman Empire in the age of the Antonines, writes thus :— " All these cities were connected with each other and with the capital by the puhlic highways, which, issuing from the Porum of Home, traversed Italy, pervaded the provinces, and were terminated only hy the frontiers of the empire. If we carefully trace the distance from the wall of Antoninus to Home, and from thence^ to Jerusalem, it will he found that the great chain of communication from the north-west to the south-east point of the empire was drawn out to the length of 4080 Eoman miles. The public roads were accurately divided hy milestones, and ran in a direct line from one city to another with very little respect for the obstacles either of nature or private property." I t is the design of the present paper to shew that a link in this great chain of communication was the route from Eochester through Southfleet to London, and that the Eoman station of Vagniacce was situate upon it, where the three parishes of Northfleet, Swanscomb, and Southfleet meet, at a place which for many years has been known as Springhead. We find the word Vagniacis once only in the Itinerary of Antoninus, where it occurs in the second Iter. The somewhat similar term Vagnaca occurs in the fifteenth Iter of the supposititious Richard of Cirencester, in a circular route from London to Southampton and back by the eastern coast of Eent. The only derivation, of which I had heard, for this name, is the adjective " Vagus," indicative of wandering ; possibly in relation to a stream "meandering." Nothing * Decline and Fall of the Soman Empire, vol. i., c. ii. VOL XVIII. ' -N1 7 8 THE ROMAN STATION OP VAGNIACJ3 specially favourable or adverse to its application to this locality has, I am bound to state, come to my knowledge; but recently Sir Edwin Arnold has suggested " stagni aquse " or the " marsh waters;" the quantity of marshland would not, however, have been particularly noteworthy or distinctive at Springhead. At the inception of such a station there may have been no houses nor any settlement then existing, for we find, at twenty miles from Maridurnum (Carmarthen), a station denoted simply as ad vigesimum, equivalent to " at the twentieth milestone." To the north of the road, however, there was a British Oppidum lying in the forest of Swanseomb, through a part of which forest this Eoman way is believed to have held its course. The initial V is not very unusual in Latin proper names of places ; of the 160 names which occur in the Itinerary I should suppose some twenty commence with that letter. Taking Vagniacce as a nominative plural, it does not appear, that many stations possessed a similar noun of multitude as cognomen. The next station, however, was Duroorivce* (Rochester), the name of which may denote camps by a running water, and be well applied to the Roman stations (Strood and Rochester) on opposite banks of the River Medway, and it is probable that the station Vagniacce may have derived its plural name from being divided more or less into two by the fleet, or stream, which, taking its rise at Southfleet, flowed to Northfleet (hence the two names), and there discharged itself into the River Thames. * By some -writers JDurolriv® is rather regarded as equivalent to fluminis trajecttis, a ferry, which presents the same idea of duality in the two banks. The Roman remains in and about Roohester have never been adequately treated of, hut we cannot hastily dismiss it from our attention, because, as will be seen in the sequel, it will be very material to start from the proposition that it occupied the station mentioned in the second, third, and fourth Itinera of Antoninus as DurobrkAs, the dative of Dwrdbrvom. This claim can scarcely be disputed. In Gibson's comment on the fifth Iter he writes: " Though the name of Durobrwcs ocours before both in the second, third, and fourth Iter of Antoninus, it may not be unworthy of notice that all antiquaries have agreed in those three Iters, fixing that station at Roohester in Kent, though it is not quite answerable in its distance from London to the miUiary numbers of the Itinerary, whioh places it twenty-seven miles from the last-mentioned city, whereas it is about thirty, which in the main answers to twenty-seven of the Itinerary according to Horsley." Durolriva} in the fifth Iter was fixed hy Camden at Dornford, by Gale at Brig Casterton, and hy Gibson at Caistor in Northamptonshire, AT'SPRINGHEAD, NEAR GRAVESEND. 17 9 Another explanation of this plural form may be found in the circumstance that the Springhead Valley bifurcates (so to speak) toward the south; the one hollow running in the direction of Southfleet Church, and the other to the south-east toward Hazells and Northfleet Green Parm; and these twin-terminals of the Pleet might well occasion and account for the plurality of the nominative Vagniacce. At the period of Caesar's* invasion it may well have been that this station Vagniacce was non-existent as a populated place; there is no need to suppose that it, any more than Londinium itself, was of any considerable importance. Neither Vagniacce nor Londinium finds any mention in Caisar's Commentaries. The probability is that we owe it, together with the main Eoman military roads of Britain, to the efforts of Julius Agricola, Propraetor in A..D. 78 under the Emperor Vespasian, a consideration which raises the question of the probable date of the Itinerary of Antoninus. Horsley thinks it should be (as to the British portion) ascribed to Caracalla, since he spent some years in Britain, and was present at York at the death of his father Severus, Pebruary 4 A.D. 211. If it were ascribed to Antoninus Pius other questions would arise. This Roman military way naturally leads our thoughts to London, and there to the relic of the " London Stone " in Cannon Street. It is clear that the island of Britain became a province of the Roman Empire under Claudius Drusus, A.D. 45, and it is I believe unquestioned that London became a * The object of this paper being limited to a claim " that the road by Springhead was an integral part of the great line of communication between the Kentish coast and London, and that Springhead itself was the station upon such line called Vagniacce" it would answer no useful purpose to mix the inquiry with any question as to the routes taken by Julius Cassar upon his two successive incursions into the county of Kent, for there is no necessary connection between them. The British roads or trackways had at the period of such invasion neither the directness of line nor the solidity of construction of those which were afterwards formed hy the conquerors. "Whether Csesar's passage of the river was over the Thames at " Cowey Stakes/5 or really over the Medway, at Aylesford, or elsewhere, is a question which is still left to the research of the antiquary. Let me say however, in memory of one who lived near to Springhead (Mr. A. J Dunkm), who dearly oherished the renown of Cantium and its four kingdoms and who contended that Julius Csesar's most extended progress from the seashore was still confined within the confines of Kent, that his contention carries With it many arguments that are worthy of notice, * 3 180 THE ROMAN STATION OP VAGNIACiE colony about the time that Julius Agrieola was in Britain, under the Emperor Domitian A.D. 85, when it soon began to lay the foundations of that pre-eminence which, once acquired, it has ever since maintained. A strong reason for the compilation being ascribed to a later date is that it comprises Constantinople, and some other places which were not founded till long after the last of the Antonines. Mr. Wright gives the date as A.D. 320. Taking our stand at the London Porum in the middle principal street, or Prsetorian way, afterwards known as Watling Street, we should observe the four military ways which anciently led out from their common centre in the City. I. The first of these, with which we are most concerned, being the Trajectus or ferry over the River Thames, from Stoney Lane in Southwark, entered London at Dour (water) Gate (alias Dowgate) and directed its course to London Stone (in Cannon Street). II. The second Watling Street, or the Prsetorian way, directed its course from south-east to north-west, through London City to Newgate. III. The third, Ermyn Street, directed its course from south-west, northward to Cripplegate, taking its way by Highbury Barn to Stroud Green. IV. The fourth, the vicinal way, led through Oldgate (later Aldgate) by Bethnal Green to the Trajectus or ferry at Old Pord. Thus these four military ways passed through the four original gates of the City, Newgate, Cripplegate, Aldgate, and Dourgate. During the government of Julius Agrieola (under the Emperors Vespasian, Titus, and Domitian) the civilization of the Britons, after the Roman manner, made great advances both in apparel and in the erection of houses and of temples, with all other symptoms of the possession of material resources and of advancing knowledge, accompanied by the cultivation of arts and sciences. The Roman road to Dour Gate, from the Kentish coast, proceeded with the directness of an arrow from the sea coast to Rochester, whence, if my readers will consult the AT SPRINGHEAD, NEAR GRAVESEND. 181 annexed plan, (reduced from the Ordnance Survey,) they will see that it continued its course up Strood Hill, and running onward parallel with Crutches Lane to the north of the village of Cobham, reached Springhead at a distance somewhat under, and not over, ten miles, and thence towards London, its ultimate terminus. I do not mean to contend that the Continental route was then, as now, wholly from Dover to London; because it is clear, from the incidental statements in such authorities as Lucan, Tacitus, Ammianus Marcellinus, Ausonius, and others (whose testimonies have been gathered by Mr. C. Roach Smith in his interesting work on the great Kentish Castra), that Riehborough was the chief marine portal of Britain, so much so that Butupice gave its name to the whole British coast opposite Gaul, aud even to the neighbouring sea itself. I t is not, however, a matter which affects our inquiry; since, whether we take the Continental trajectus to have terminated at that port, or at Dover (Duoris), the Roman way in both cases is short and direct to Durovernum, (Canterbury), and thence it continues its westward course to the capital. Before proceeding further let us set down those parts of the second, third,-and fourth Itinera which immediately concern us, using the English names of the probable sites for the purpose of better elucidating the argument. ITER II. From the Great North "Wall to Riehborough 481 miles. To London . . . 407 miles To Noviomagus - - 10 To Springhead - - - 18 To Rochester - - - 9— 37 „ To Ospringe - - - 18 To Canterbury - - - 12 To Riehborough - - 12— 37 ITEB III. From London to Dover 66 miles. To Eochester - -. - 27 To Canterbury - - - 25 To Dover - - - -14 ITER IV. Prom London to Lymne 68 miles. To Rochester - - - 27 To Canterbury - - - 25 To Lymne - - - 16 66 68 182 THE ROMAN STATION OE VAGNIAOE Westward of Springhead, the direction of the road to Londinium under the second Iter has been a matter exercising the arguments of the learned with very conflicting results. Some are for continuing its course with the same directness through Dartford. Others contend for a south-western deflection to Holwood Hill at Keston, or to Woodcote near Croydon (alleged sites of the intervening station of Noviomagus) turning thence to the north to reach London. Touching this digression to Noviomagus, so palpably out of the direct route to London, it has been alleged that the swamps between Dartford and Crayford, arising from the converging of two rivers, the Darenth and the Cray (not then embanked), created so boggy a tract that the Roman way was necessarily diverted by it towards the south-west, and further that the magnitude (thirty acres) and importance of the enclosure at Keston, assuming it to be Noviomagus, justified such a diversion. Whatever may be ultimately decided as to the true site of 'Noviomagus, there is nothing which need be held fatal to the claim of Springhead to be the site of the antecedent station, on the way from Dover to London. Hasted, writing of the camp at Keston, says : " Others have supposed this to have heen the remains of Noviomagus, the first Roman station from London towards Dover; in support of which they urge that the ancient Roman road, instead of going through Deptford as the present one does, directed its course much more southward, making a circle to avoid the marshes between Lambeth and Deptford, which were then passable with much difficulty from their lying so very low; and as the river was not then embanked were overflowed upon all spring-tides, as they are now upon all extraordinary ones." (Vol. ii., p. 39, octavo.) Here we may turn aside to see what other places have claims to be the station of Vagniacce. They are chiefly Northfleet, Crayford, and Maidstone, all in Kent. Por the latter there is the high authority of Camden, who connects the word vagus with the River Medway, upon which Maidstone stands, and because he thought it complied with the true distances in the Roman Itinerary. He was supported in this by Bishop Gibson, his translator, also by Mr. William Burton and Dean Gale; aud in Ainsworth's Dictionary I observe it is stated as a matter of fact. AT SPRINGHEAD, NEAR GRAVESEND. We have no instances, however, whatever o£ Maidstone being so designated in the archives of antiquity. On the other hand, Mr. Somner, Bishop Stillingfleet, and Dr. Harris fix Noviomagus at Crayford and Vagniacce at Northfleet. In favour of Springhead we possess the later, and therefore probably the riper, researches of Mr. John Thorpe and Mr. Roach Smith. In Thorpe's Custumale Boff., p. 249, we read: "But what renders Southfleet most remarkable is that here was the Vagniacce of the Romans, their second station from London, according to the Itinerary of Antoninus." His authority is confirmed by Hasted, who says : " Somner and some others have placed the station of the Romans called Vagniacce at Northfleet, not far distant, hut the objection to this is that the valley between Northfleet Hill, leading to the bridge, and the opposite hill westward from it, was at that time a broad sheet of water . . . . therefore it is reasonable to suppose . . . . the Eomans shaped their course more to the southward towards Southfleet, where it was more narrow, and where they had the benefit of a fine spring which rises there, still known by the name of Springhead." (Hist, of Kent, ii., 423.) Mr. Roach Smith, to my mind, has solved the whole dispute by a judicious and natural solution, which, accepting its statements, involves no contradiction of the great Antonine record. It is as follows. In Archceologia Cantiana, Vol. X., p. 171, he says: "Noviomagus therefore it may be inferred was not situated upon the nearer and direct road to Durobrivis. It must then be placed at some spot southwards, where a road going to it and then proceeding to Vagniacce would give an additional ten miles. If we try to adopt the opinion of Gale and others we have a difliculty in the distance, which would much exceed the required eighteen miles, and then the retrogression to Rochester to regain the direct route. If we place Vagniacce at Springhead we are supported by all the requirements for such a station, in extensive foundations of buildings and in antiquities of a general character; and also the distance on the direct military road to Eochester. As regards Noviomagus being located, as proposed hy Mr. Kempe, at Keston, there is the same objection as to distance. It is probable the place has yet to he discovered so as to answer this requirement. But although it appears to me that we are more justified in placing Vagniacce near Springhead, yet I have no doubt a Eoman road ran near or through the site of Maidstone, probably direct from London to the Partus Lemams. 184i THE ROMAN STATION OE VAGNIAC-E The painstaking and courteous Hon. Secretary of the Kent Archseological Society, the Rev. Canon Scott Robertson, in the same work (Vol. XV., p. 68), writes : "The learned Camden iu his Britannia sought to identify Maidstone with Vagniacce, a Eoman station mentioned by Antoninus as forming one stage on the second Iter or route given in his Itinerary. He says it stood twenty-eight miles distant from London, between Noviomagus and Durobrivce (Eochester), on the road to Durovernum (Canterbury). He founded its identification mainly upon a fancied resemblance between the Saxon name of Maidstone and the word Vagniacce. "We cannot endorse Camden's conjecture, although it was adopted by many learned writers after him.".... The Rev. Beale Poste, in Archceologia Cantiana, Vol. I., p. 154, says: " Many who read these pages will be doubtless aware that in fixing a determinate site for the Eoman military station, Vagniacce, I am assuming to do that which has never been attempted to he done by the old antiquaries, as Camden, Gale, Burton, Leland, or Lambarde, or by the moderns, as Gibson, Gough, Reynolds, Hasted, Thorpe, or Hatcher, or even been supposed by Newton himself, the learned historian of the place. This undoubtedly is so ; for though a fair proportion of those eminent persons in the antiquarian department of literature may have thought generally that the Eoman station was at Maidstone, yet none of them have advanced so far as to point out in what quarter of the present town the pre-, cise spot was situated," But to maintain his thesis the reverend author has to contend that the Roman way, from the Rutupine coast to London, traversed the route to Rochester and then started off to the south to Maidstone, and thence to Noviomagus at Keston, by Wrotham Heath and Oldbury Camp, by the mere statement of which detour the essential weakness of the theory is manifested. Again, speaking of "the Roman way," he says, "When the road has crossed (the Medway) at Radford (quarter of a mile below Maidstone) it ascends the gradually rising ground for 300 or 400 yards, and having reached a species of plateau, or level, it joins nearly at right aggies the ancient Roman road, proceeding in one direction to "Rochester, and in the contrary direction communicating with the Weald. . . . . I venture to place Vagniacce at the junction of the two roads of which I have just spoken." {Ibid., p. 156.) He AT SPRINGHEAD, NEAR GRAVESEND. 185 places another Vagniacce at Aylesford, and thus justifying a plural nomenclature adds, " that iu some states of the tides the traffic would cross the river at Radford (Maidstone) and at other times by the ford at Aylesford." (Arch. Cant., I., 174.) I t has often been wished that the numerals of the Itinerary could be verified with the original MS,, but it appears there were no less than nine excellent editions in existence prior to 1711, and Dr. Gale, in the Preface to his Commentary, says that in these the variation in distances is very inconsiderable, that in Iter II. the only variance consisted in the fact that at Luguvallwm the Oxford copy for twelve reads fifteen; and at Deva Bentley's edition for twenty reads ten. That the embarrassment arising from the difficulty of tracing the route through Noviomagus has been real and has provoked many attempts to solve it, is evident from the circumstance that the Rev. John Landon, Curate of Southfleet, in the last century proposed that, after exhausting the ten miles from Durobrivce to Vagniacce, the march should still be continued for another eight miles by overlapping and reincluding the prior ten miles traversed, thus making up the eighteen; and that this enumeration would then yield a situation on the River Cray, which he would style Noviomagus. In this view he was partly followed by Mr. A. J. Dunkin, but the following extract from Mr. Landon's interesting letter to Mr. Hearne, the eminent antiquary, will throw additional light upon our inquiry: " There were lately found in Mr. Pedder's Barkfields (or perhaps more properly Bagfields, from ' Vag,' the first syllable in Vagniacce) in Southfleet two small silver pieces There is upon one of them a very fine and bold profile of a woman's head with ' PlautiUa' round it very, plainly to be read; the reverse we have not made out. The other seems to be something emblematical of Eome, but the hammer having in some places struck too nigh the edge, the letters are not very legible. I do not remember whether I ever told you that near the place where these pieces were found there was some few years ago a very fair milestone discovered. It stood upright in the ground with its crown about four or five inches below the surface. I measured it soon after it was dug up. It was two feet and a half long, two .of its sides were sixteen inches each, the other two fourteen, its corners were chiselled, but its faces were very 186 THE ROMAN STATION OP VAGNIACJE rustic. However, upon one of the sides was a very fair X cut, which was undoubtedly to shew that it stood ten miles from some particular place. Now if we only suppose the Eomans in placing their milestones reckoned upwards, that is to say, from Durobrivis to Vagniacce, ten miles, and place the stone there, then this would be a convincing proof that Vagniacce must have stood near where this stone once did, because the distance between Vagniacce and Durobrivis is always called ten miles, and this I beheve is very nearly the distance between Durobrivis and the place where the stone was once found." We donotfind any elucidation of our subject in the Notitia, written about A.D. 450, nor in the celebrated Ravenna Geographer's Treatise, about the middle of the seventh century, and the same negative remark applies to the other authority upon such subjects as this enquiry, viz., "Peutinger's Table," written probably at a period between Constantine and Theodosius the Great; but let us quietly reexamine the case in favour of Springhead (confining our attention to Itinera IL, III., and IV.). Here we find the Iter from Londinium to Durobrivce stated thrice in succession to be only twenty-seven miles, which by this shortest length of measurement must be drawn through Springhead, the most direct and shortest route, for in no other direction could the distance be comprised within such limits, thus from London to Vagniacce \7\, from Vagniacce to Rochester 9^—27 miles. In reference to Roman as well as other milestones, only complete miles, and no fraction of a mile, are inscribed upon them, nor are fractions given in any part of the Itinerary, and hence when we come within half a mile of a new distance the stone would equally accord with the next following number. And surely these three Itinera, when placed side by side, remove all the difficulty which arose from the interposition, in the second Iter, of the station called Noviomagus. The compiler well knew his subject, and has said in effect, as explained by Mr. C. Roach Smith, if you proceed and pass through each station under Iter No. 2 it will involve the detour through Noviomagus, and this will make the distance between London and Rochester ten miles more than the thrice-affirmed twenty-seven miles—thus London to AT SPRINGHEAD, NEAR GRAVESEND. 187 Noviomagus, 10 ; Noviomagus to Springhead, 17£ ; Springhead to Rochester, 9£; total 37, which is not an incorrect computation, if we assign Keston or Bromley, or some proper place in their neighbourhood, as the equivalent of Noviomagus. As to the continuation of the Roman way to Londinium, Hearne writes : "On the Dartford Brent it is very visible, keeping its course south-east, and entering the enclosures appears again where the plough has not levelled it at different places " (Onstumale Boff., p. 150). So, again, Hasted (vol. ii., p. 292): " The Roman road shews itself very conspicuously on the south side of the high road between Dartford and the Brent" ; and I have myself understood, from Mr. Roach Smith, that he had a reliable account of the opening in the High Street of Dartford of an indubitable Roman military way constructed with paviors or squared stones; a fact to which I assign much importance, and I am anxious therefore to put it upon record. If any one will stand at Bexley Heath or Shooter's Hill, and look along the London and Dover road, climbing and descending hill after hill from Dartford through Welling to Blackheath, he will exclaim, " If those were not the lines of a true Roman way all ordinary experience is at fault." A straighter road, and one more regardless of the undulations of the country (and it embraces a section of some eight miles), could not be traced. A Roman way from Riehborough (the early port), and from Dover (the later port), to London, if once shewn to extend from Canterbury direct to the summit of Shooter's Hill, needs no further safe-conduct from me to conduct it surely to the Dour Gate of the Metropolis. The claims of Springhead would not, however, be complete, as being the site of Vagniacce, in the absence of Roman remains discovered upou the spot; and these have been considerable. I t appears that the Rev. John Landon took away with him many Roman coins found at Springhead, when he left the neighbourhood and accepted preferment in Lincolnshire; and we know, both from Hasted and from Hearne, that large 188 THE ROMAN STATION OP VAGNIAC2E. and continued discoveries were made hj Mr. Pedder, a former lessee of Springhead, comprising bronze articles and parched corn; and that Roman coins found at Springhead were in the possession of one Lane a beerhouse-keeper at Betsham (a hamlet o£ Southfleet); there are also interesting Roman sepulchral remains from Springhead, now at the British Museum, where they were deposited by a former rector of Southfleet, the Rev. Peter Rashleigh; and in my own collection, (wholly made during the present century by Mr. Silvester and his son, the late and present occupiers of Springhead,) I enumerate some 1400 Roman coins, ranging from Augustus to the departure of the Romans, besides bronze fibulae, stirrups, bells, rings, a fish hook, and pins; while in iron there are horse-shoes (part of a mass of some 500 which were found in a heap), a horse's bit, spur, spears, a billhook, and other implements, and since this paper has been in hand I have met with the foundation of a Roman building yet to be explored. Part of Mr. Rashleigh's discoveries consisted of a Roman family cemetery, enclosed with walls of masonry, some fifty feet square. It is in respect of these that in 1801 he wrote to the Society of Antiquaries as follows : "The field in which these remains were found adjoins the "Watling Street Road, at the corner of which Mr. Hasted, in his History of Kent, YOL i., p. 271, supposes a Roman milliare to bare been found, and which Dr. Thorpe conjectured to have been the ancient station of the Eomans called Vagniacce; this is likewise near the Springhead, which flows down into the Pleet at Northfleet Bridge, and. which was supposed to have been formerly navigable for vessels." . To conclude, let me add that the topographical situation was admirably calculated for a station in all that concerned openness of country, pure water supply, river communication open to the sea, besides its incidence upon the Via Militaris, which in its passage through Swanscomb Forest skirted the old British Oppidum on the right, as indeed it had similarly skirted on the left the Kentish settlement in Cobham Park.