Part 14

“There Thames runs by beneath the wall, Where pass the merchant vessels all, From every land, both high and low, Where Christian merchants come and go.” TRISTRAM, _c._ 1175.

Footnote 3:

The substance of this chapter was read at the Society of Antiquaries about 1917, but it has not been printed before.

_FIRST BRITISH CITIES._—Ancient cities were not planted down by an act of will, they sprang up on lines of communication as centres of control and commerce. On a geological map it appears that a chalk belt passes from Kent to Hampshire towards the south bank of the Thames. From the north bank another wide belt diverges to the north-east. The backbones of these chalk regions are the North Downs of Kent and the Chiltern Hills; they contain between them a long triangle of gravel drift and marsh flats through which the Thames flows to the sea. These downs, as we know to-day, when we find ourselves on them, are pre-eminently walking grounds, and they must have been the prehistoric ways of communication. “Primitive man traversed the ranges lengthways: in the valleys were forests almost impenetrable, whereas the backbone of each ridge would stand bald above the ocean of trees.” The oldest roads were “Ridge-ways.” On the high Wiltshire downs at, or near, a point where the southern system of downs converge, stands Stonehenge, and I cannot doubt that it was in some way conceived as being a centre and “capital” of the country. The Gauls recognised such a centre, or “omphalos,” near Chartres. Since writing this, I find that Sir C. Oman has said, “Britain must have had some focus corresponding to that for Gaul; possibly among the prehistoric monuments of Salisbury Plain.” Stonehenge, I may say in passing, is a monument of wrought stone set out with precision, and I cannot see how it can be earlier than about 500-700 B.C.

The ancient trackway along the Chilterns, known as the Ickneld Way, reached the Thames near Wallingford. Travellers going south and east from this point struck across the narrow space of low broken ground between the two chalk ranges by a short linking road. Silchester, the capital of an important Brito-Belgic tribe, lies on or near the course of such a road in a corn-bearing region. Silchester was the key of the old road system over the Thames fords. It is known to have been one of three most important pre-Roman British centres, and we may, I think, look on it as the first British city.

The British city of Verulamium lay to the south of the Ickneld Way, in the same great triangle between the two chalk regions which is here much wider. The rise of this centre suggests that a road linking the two chalk ranges had been found across the river valley much lower than Silchester. The later Roman Watling Street, directed straight on Verulam, formed such a link, and there are many reasons which suggests that some underlying British trackway must have been the cause why Verulam became important. Later, again, Colchester came to be the chief city. Possibly it was favoured as being more remote when the Romans should make an attack. It seems to have been named after the Celtic war-god, and this may be significant. (In Roman days as in mediæval, there was probably a ferry from Gravesend to Tilbury for direct access to Colchester from Kent. This seems to be suggested by the Peutinger roadmap.)

_Origin of London._—By origin I mean the beginning of a development which led to the establishment of a port and commercial town. Doubtless the site may have been occupied by some dwellers in the Stone Age. For many centuries before the Roman conquest Britain had been in commercial relations with the Continent. Just before the conquest Verulam was the capital of the leading Celtic kingdom. This Brito-Belgic kingdom had its southern boundary along the Thames and its eastern at the Lea, and these are still boundaries of Middlesex. If this kingdom, with its capital some twenty miles inland, had any sea-borne trade, its port must have been on or near the site of London. It is even probable that this port was the cause of the pre-eminence of the little kingdom to which it belonged. The port was to Verulam what the Piræus was to Athens, Ostia to Rome, Dover to Canterbury, and Southampton to Winchester. London was doubtless the source of the wealth of King Cymbeline, and we might very well look on him as the founder.

Dr. Guest argued that London was founded as a Roman camp at the time of the Claudian conquest; but it is now agreed that the name is Celtic; and it must not be forgotten that London is and always was a port. When we first hear of London only seventeen years after the Claudian conquest, it was already, as Tacitus says, famous for the number of its merchants, and this must imply that it was a principal port. Dr. Haverfield, while admitting that the name is Celtic, went on to say: “The name Londinium, the place of Londinos, witnesses at most to nothing more than one wigwam or one barn.” This “at most” can only mean that every town presumably begins with one building; in London, however, the building is not likely to have been a barn amid the bare gravels, but rather a boatman’s house. Further evidence for the existence of a pre-Roman town is brought out by the large number of Celtic objects found on the site and in the neighbourhood, but they have never been properly catalogued as a group. Dr. Haverfield allowed that three pieces of imported Samian ware in the British Museum might belong to the period A.D. 10-40. “We might then conclude that through the influx of Roman traders London had been noted as a suitable trading centre a few years previous to the Roman conquest; but the minute dating of these potsherds is not easy, and we must leave the question of pre-Roman London unsettled. Either there was no pre-Roman London, or it was an undeveloped settlement, which may have been on the south bank of the Thames” (_Journ. Rom. Studies_, vol. i. p. 146). The evidence of such early imports is greatly strengthened by other discoveries at Silchester. Mr. May, speaking of the early “Samian” ware, says: “The Silchester examples are of much significance. Together with the contemporary Belgic imitations they prove that the inhabitants of the capital of the Atrebati were importing costly luxuries in considerable quantities from Italy and Northern Gaul at the beginning of the Christian era.” Early Belgic pottery has been found in London as well as “Samian,” and there is in the British Museum a wine jar of an early type found in Southwark. Some British pottery was doubtless made in Londinium itself before the Roman conquest. Mr. Lambert has described specimens of coarse wares in _Archæologia_. Of one of these he writes: “Bead-rimmed pot, coarse grey ware, irregularly burnt. A pre-Roman type, surviving into the Roman period.” He dates it A.D. 50-80, I suppose thinking that it cannot have really been pre-Roman.

London above bridge is an inland city, the English capital; below bridge it is a great seaport. In a description of England, published in 1753, I find this: “That part of the Thames, which is properly the harbour, is called the Pool, and begins at the turning out of Limehouse Reach and extends to the Custom House quays. In this compass I had the curiosity to count the ships, and I have found about 2000 sail of all sorts of vessels that really go to sea.” In a twelfth-century rhyme on English towns are the words, “London for ships most.” Bede describes London as a great ship port. The city is placed just where the Thames widened into an estuary. At Battersea the river was little wider two thousand years ago than at present; it overflowed wide spaces of marsh about Westminster and again contracted by London. Her high ground came close to the water on the north, and on the Southwark side there was only a narrow margin of low ground. Directly to the east of London was the low land called in the Middle Ages “Wapping Marsh” (_Middlesex Feet of Fines_). In the Pepys collection at Cambridge I have seen an engraved plan of “Lands by Wall or Wapping Marsh, 1683: seven acres of land in which the millponds and ditches did all over dispersedly lie.” Stow tells of Limehouse marshes being “drowned.” Before the lower Thames was embanked the river must have been two or three miles wide, at every tide, a little below London, where the considerable little river, the Lea, runs into it. The higher ground of the site of London is in the angle formed by the Thames and Lea, and is the extremity of the northern hills, Highgate and Islington. From the hills several streams flowed through deeply excavated beds into the Thames. The most considerable of these was the Fleet; the smaller Walbrook intersected the site of the city. Conyers in his MS. at the British Museum noted how the Fleet was embanked in 1675 with material taken from old St. Paul’s, “to narrow-in the spreading breadth of Fleet River.... The waters overflowed these parts in the old times.” The general topographical conditions were well observed by Drayton in _Polyolbion_—The city was built on a rising bank of gravel and sand, surrounded by lower ground: the tide flowing up the Lea and Fleet prevented the town from growing too long: to the north and south of the Thames were ranges of hills: “And such a road for ships scarce all the world commands.”

_Way to the Port._—The men who first came to the site of London must have come from the higher ground of Islington and Highgate; they did not cross the Lea or the Fleet. Before some engineering was done the natural way was from the direction of Verulam. Now, an ancient road lies along this course from St. Albans to Aldersgate. As it approaches London it passes between the Walbrook and the Fleet, pointing towards what the old tablet near St. Paul’s says is the highest land in the city. The Walbrook where it fell into the Thames must have had steep clean gravel banks containing a tidal inlet—a perfect landing-place where small ancient ships could be brought alongside. This creek, afterwards known as Dowgate, must have been the original port of London. Along the old road wine, pottery and bronzes were carried into the interior, and corn was brought for export. Dowgate is known as a port for foreign ships from Saxon days (Round’s _Commune of London_). It is especially interesting to find from Stow that in the fifteenth century the Abbot of St. Albans had a quay by Dowgate. (Old writers supposed that “Dow” represented the British word for water; recent scholars equate it with _Dove_; but even so there is the curious analogy with Dover and such like place-names.) The Roman gates of London, of course, opened on important routes, and “the street from Aldersgate to Islington” is mentioned in the twelfth century (_Middlesex Feet of Fines_). Stow says: “From the further end of Aldersgate Street straight north to the Bar is called Goswell Street. Beyond leaving the Charterhouse on the left hand the way stretcheth up towards Iseldon.” Again on the old woodcut, usually called Aggas’s map, the street out of Aldersgate is inscribed “the way to St. Albans.” That excellent old book, John Nelson’s _History of Islington_, carries the account of this road forward, and he thought that it was Roman. He quotes a passage from Norden, to the effect that it passed east of Highgate through Tollington Lane to Crouch End, Hornsey Park, Muswell Hill, etc. “Tolentone,” he points out, is mentioned in _Domesday_. This road is laid down on old maps. Recent modifications at Islington may be made out by comparing maps given by Nelson and by Lewis in 1842.

I now quote the passage relating to this old ridgeway road to Verulam from Norden’s MS. (British Museum, 570). He begins at Clerkenwell instead of from the City: “It is not to be omitted to declare the old and ancient highways heretofore used by our fathers though the new be of greater regard and account for that they yield more ease unto the travellers. There was an old way that passed from Clerkenwell as also from Portpoole [Gray’s Inn] towards Barnet and so to St. Albans. From Clerkenwell it extended as the way now is unto a bridge or brooke between Gray’s Inn Lane and Pancras Church, near which brooke it entered into an old lane leaving Pancras Church on the west. It is called Longwich Lane, through which lane it passed along leaving also Highgate on the west and passed through Tollington Lane, whence it extended to Crouch End and thence through the Park to Muswell Hill near by Colney Hatch and so to Friern Barnet, from thence to Whetstone and there meeteth the new way. The cause why travellers left this old and ancient way was the deep and dirty passage in the winter.”

The road is well described in Pennant’s Tour (1782): “On quitting St. Albans I passed the wall of Sopwell Nunnery mixed with quantities of Roman tiles. After London Colney on the Colne I reached Ridgehill (!), a most extensive view. At South Mimms enter Middlesex and about a mile farther made Barnet; in Saxon times a vast wood filled this tract. From this town is a quick descent. Just beyond Whetstone the road passes over Finchley Common, infamous for robberies, and often planted with gibbets. About a mile beyond stands Highgate, a large village seated on a lofty eminence overlooking the smoky extent beneath. Here, in my memory, stood a gateway at which in old time a toll was paid to the Bishop of London for liberty, granted between four and five hundred years ago, for passing from Whetstone along the present road instead of the old miry way by Friern Barnet, Colnie Hatch, Muswell Hill, Crouch End, and leaving Highgate to the west by the Church of St. Pancras. After resting for a small space over the busy prospect, I descended into the plain, reached the metropolis, and disappeared in the crowd.”

The old miry way by Crouch End is, I cannot doubt, the original British road from Verulam to Londinium. (St. Pancras, it may be mentioned here, must be a very old settlement; near by was a bridge over the Fleet River, at a later time called “Battle Bridge,” on which name theories have been founded, but I think the bridge may have taken the place of “Bradford in the Parish of St. Pancras,” mentioned in the _Feet of Fines_, 23 H. viii.)

A summer’s day journey to London, such as Matthew Paris would have known it, must have been of beauty unimaginable when the miry lane was not too wet. Mention is made in the time of Henry VIII. of “a capital messuage called Muswell Farm in the parish of Clerkenwell and Hornsey, and the site of a certain chapel in the said parish, now dissolved, lately called Muswell Chapel” (_Middlesex Feet of Fines_, 35 H. viii.). A memory of the view of St. Paul’s rising from the midst of the walled city is given in a little sketch by Matthew Paris himself. I find this of Highgate in 1753: “On the summit of the hill a view over the whole vale to the city, and that so eminently that they see the ships passing up and down the river for twelve or fifteen miles below London.” Of Hampstead: “The Heath affords a most beautiful prospect, for we see within eight miles of Northampton, and the prospect to London and beyond it to Banstead Downs, Shooter’s Hill, Red Hill, and Windsor Castle is uninterrupted.”

A note of Camden speaks of another old road striking across to Edgware. “Hampstead Heath, from which you have a most pleasant prospect of the most beautiful city of London and the lovely country about it, over which the ancient Roman military way led to Verulam by Edgworth and not by Highgate as now, which new way was opened by the Bishop of London about 300 years since.”

Drayton showed remarkable perception when, describing the hills about London, he wrote of Highgate:

“Appointed for a gate of London to have been When first the mighty Brute that city did begin; Its holts to the east stand to look Upon the winding course of Lea’s delightful Brook.”

When Walbrook Creek was a landing-place having a road connecting it with the interior, we may be sure that boating passages across the Thames would be common, and very soon a link with the road to Dover would be formed on this line. Thus, the road through Southwark must have followed the foundation of London immediately. As is well known, Ptolemy put Londinium in Kent, but he—as Dr. Bradley pointed out—was frequently very wrong in regard to inland places.

An ancient bronze mace-head was discovered in the gravel taken from under old London Bridge, which, I believe, has never been illustrated (Fig. 171). It was one of the mace-heads which are classed in the British Museum as of the Bronze Age, but they are, I think, early British. I have found a drawing of the mace-head in question in some interesting volumes of sketches by Fairholt at the South Kensington Museum. Fairholt’s note reads: “Bronze mace found at Barnes, November 10, 1841, amongst the gravel taken from old London Bridge.” Fig. 172 is an early bronze mace-head from Italy, in the British Museum, given for comparison.

The conditions were favourable for establishing a way in the line of London Bridge, for hard ground here approaches near to the south bank of the river. That the Roman city spread from Walbrook Creek as a centre is now generally agreed. Mr. Lambert’s plan of the finding-places of Claudian and pre-Claudian coins shows them distributed near the primitive port. Again, the city Watling Street is probably the beginning of the old road from the port. Wren found traces of an old street running aslant under the end of old St. Paul’s, and this probably formed part of the way towards Aldersgate. The acceptance of such a route as the main street of the oldest London would solve the difficulty of the “fault” in the lines of Newgate Street and Cheapside. I suppose that the Roman street through Newgate (which all would agree was formed at a late time when the walls and gates were built) branched westward from the old Verulam road I have been describing. In a similar way, the Roman road on the course of Old Street probably branched to the east out of the same ancient Verulam road. Mr. Codrington and others have supposed that the road to the east was continued also westward, but no evidence of this has been found. Stow, in his account of Aldersgate Street, says: “On the east side at a Red Cross turneth the Ealde Street, so called for that it was the old highway from Aldersgate Street for the north-east parts of England before Bishopsgate was builded.”

_The Westminster Crossing._—It was remarked above that the emergence of Verulam into importance probably followed on the use of a river crossing at Westminster and a trackway in the course of Edgware Road, which is known to have been part of a later Roman highway—the Great Watling Street. It is generally allowed, as by Dr. Rice Holmes, that a British trackway underlies the general course of this great Roman highway from Dover to St. Albans and beyond. The monk Higden, writing about 1360, said that Watling Street passed to the west of Westminster, but it has been objected that what the monk thought was not evidence. However, in his time and until about two centuries ago, an important river crossing was maintained at the Horse-Ferry. The Horse-Ferry Road appears to have been made to divert a direct passage at Westminster when the great hall of the new Palace was built, about 1100. Sighting the line of Tothill Street, we see that it would have passed by the old Palace, but that the Hall blocks the way. The Abbey lies at the side of this line, which seems to mark the boundary of St. Margaret’s churchyard. Here, too, were found the Roman tomb and what appears to be a terminus mark (T II). The Horse-ferry is mentioned in an order of 1246: “The Bailiff of Kennington is to cause a barge to be made to carry people and horses over the Thames” (Hudson Turner’s _Domestic Architecture_, vol. i.). Canterbury documents show that the ferry was later in the charge of the Archbishop of Canterbury, and doubtless Lambeth Palace exists here as being on the great road. In the more direct line there still exists a short street called Stangate, which is an old name for a paved way. When Elizabeth, daughter of Henry VII., died at Eltham, “her body was conveyed to Stangate over against Westminster” (Sandford). A way to the river also was long maintained through New Palace Yard to a landing-place (see Fig. 173), from Norden’s map, _c._ 1600). Matthew Paris, in his route map of the way to Jerusalem, shows London Bridge and also the Westminster and Lambeth route, because these were alternative crossings. Tothill Street is mentioned in mediæval documents. It is possible that in early days there was a ford at Westminster, for Mr. Lambert has given reasons for thinking that formerly the tide did not rise so high as at present by 6 ft. Testimony on the course of the way from Westminster to Edgware Road is given in Ogilvie’s _Road Book_, 1675: “Piccadilly ... on the left falls in the way from Westminster by Tuttle Street; four poles from this corner, you have a way on the right by the side of Hyde Park into the other road at Tyburn.” The ancient road from Westminster would have crossed what is now Green Park in the direction of Tyburn Lane, now Park Lane. Here, in Tyburn Lane, was “Osulstone,” which gave its name to the Hundred in which London city is situated (see map recently reproduced by London Topographical Society). Tyburn, close by, was the place of execution, and doubtless the place of meeting of the old folk-mote of the Hundred, because it was at the cross roads. I have more detail establishing the continuity of this route (on which Dr. Haverfield expressed doubt), but I will pass to a few general final considerations.

The primitive road in Kent as far as Greenwich was on high ground, but beyond was the wide river valley. By bending to the left on the edge of higher ground, through Camberwell where Roman objects have been found, the river might be more nearly approached opposite Westminster, and there was solid land on the opposite bank also. Beyond, at Park Lane, the higher firm ground pushed down towards Westminster, between two little streams—the road here, indeed, was a low ridgeway. All evidence suggests that a British road to Verulam passed the Thames at Westminster. In Allen’s _Lambeth_, it is said that three “Celts” were found in digging the foundations of Westminster Bridge. Now, in Fairholt’s _Albums of Sketches_, at South Kensington, are drawings of three bronze weapons thus described: “Swords and spear found August 1847, under Westminster Bridge by a ballast heaver.” The swords (Fig. 174) were 28½ in. and 23½ in. long, the spear-head or dagger was 16½ in. long. Other pieces of British bronze work have been found in the river in the neighbourhood of the Westminster crossing. Westminster Bridge itself still carries on the tradition by crossing the river at this point, and it is interesting to find recorded that the building of the bridge in the line of the Horse-ferry was the first intention. The importance of the Horse-ferry about 1700 is shown by the list of charges given in Hatton’s _New View_ (1708).

My general results in regard to the British and Roman road systems may be summarised thus:

1. A primitive trackway along the North Downs near the south bank of the river.

2. An ancient river-crossing by a ford at Westminster and thence north-west through Britain.

3. The growth of Verulam on this road, and the rise of London as a port in connection with it.

4. A direct London-Verulam road made over Islington—a ridgeway.