“Now let’s see what happens
…’
From Ronald Searle’s Merry England, etc.,
1956.
Not much space for swelling into hugeness; no great wastes overwhelming in their dreariness, no great solitudes of forests, no terrible untrodden mountain walls; all is measured, mingled, varied, gliding easily one thing into another, little rivers, little plains, little hills, little mountains … neither prison, nor palace, but a decent home.
William Morris on England, The Lesser Arts. Collected Works, vol. XXII
If I were deposited on the site of my flat in Kentish Town, north London, in prehistoric times, would I recognize it?
Would I find the top of the hill on which my late-Victorian villa sits, concealed by all that jungle and those woolly mammoths? Would I find my way down to the Thames, by tracking the route of its tributary, the Fleet? – it still runs down the bottom of my street, albeit cut and covered with roads and housing in the nineteenth century.
What about the hills at Woolwich, where the chalk downs approach the Thames? Could I identify the flat tributary valleys of the Thames, the Ravensbourne emerging at Deptford Creek, or the Wandle, running through Wandsworth?
The chances are, I’d be lost, even with the help of the most prominent features in the landscape. Still – whether I recognize them or not – the reason why my flat is where it is, is precisely because of those ancient landscape features, many of which survive today.
Most English cities are where they are for an ancient geological, or sociological, reason. You may be a Buddhist living in Canterbury, but it’s because St Augustine picked this spot for his church that the city – and you – are there.
Ipswich – one of the biggest cities in Suffolk – is significant today because it was significant yesterday. It was the only major urban settlement in East Anglia until the Viking invasions ended in the late ninth century.
In the sixteenth and seventeenth centuries, Norwich was the second-biggest and richest city in the country after London. In the fourteenth century, its walled centre was even bigger than London’s. Its modern prominence really depends on an even earlier incarnation as a Saxon town, around a spot called Westwic, near the west end of modern Westwick Street.
Those ancient reasons for setting up a town are so deep-rooted that they can survive long periods of obscurity. Colchester – originally Camulodunum, after Camulos, the war god of the Catuvellauni, a pre-Roman tribe – was the capital of Roman Britain, but disappeared from view under the Saxons; it isn’t mentioned in a Saxon document until the tenth century.
Access to water dictated the site of most of these early English settlements. Monasteries were built close to water supplies – for cooking, washing and drainage. The Norman cathedral of Sarum, Wiltshire, was deserted in the fourteenth century because it had no water supply; as a result, Salisbury Cathedral was built in the valley below, and the town grew around it. Neighbouring Wilton, once Wiltshire’s most important town, was eclipsed by Salisbury when the River Avon was forded by Harnham Bridge, diverting the main road away from Wilton.
If rivers were medieval A-roads, estuaries were Roman, Viking and Saxon motorways: the places where most riverborne traffic entered England; where the water was so deep that the biggest ships could dock easily. The ancient importance of river and sea traffic explains why so many English towns are still sprinkled around estuary mouths. Many of our eighty-one estuaries – a fifth of all the North Sea and Atlantic tidal inlets in Europe – are still given over to industry and ports.
Even if you’re a Londoner who never uses the Thames, it’s because the Romans did rely on it so heavily that you’re living there. And it is water that still dictates the boundaries of Londoners’ lives. Like rats, the water table is never far away from you: when a London basement is dug out, you almost always have to build in pumping systems and flooding alarms to keep the ground water at bay.
The plan of the City of London remains broadly Roman, nestling next to the Thames, and roughly skirted by its Roman wall. Made out of Kentish Ragstone, probably quarried in Maidstone, the wall was built in around AD 200; Kentish Rag was also used on St Paul’s Cathedral, Canterbury Cathedral and the Tower of London.
Londinium was settled at the most easterly fordable point of the river before it grew too wide; a point where the tide can bring ships forty miles inland. Anyone downriver of that lowest fordable point – that is, anyone without a boat – necessarily had to use the London crossing first. Upriver, as the water narrowed, you could build any number of bridges.
Improved technology over the centuries has allowed rivers across England to be forded further and further downstream: like the Queen Elizabeth II Bridge, the M25 crossing over the Thames between Thurrock and Dartford, opened in 1991. When the Severn Bridge was opened in 1966, it was the longest bridge in the world; now it’s flanked by the 1996 Severn Bridge, further downstream and more than three times the length.
That lowest fordable spot on the Thames was where the Romans built the first London Bridge out of wood, 169 feet downstream from today’s London Bridge. It was rebuilt by the Normans in stone in 1176. That bridge – the only fixed crossing east of Kingston upon Thames until 1730, when Putney Bridge was built – was replaced in 1831 by the bridge that ended up in Arizona in 1971.
The southern end of London Bridge became a natural settling point, too. It’s where Southwark Cathedral, the oldest cathedral in London, was begun in AD 606. Roman London’s road network revolved around the bridge, and the river – Roman settlements at Fulham and Putney were also river crossing points.
When the Romans invaded in AD 43, the London area contained a few ancient British settlements. Otherwise, it was little more than a sprawling series of marshes flanking the banks of the Thames; marshes formed by tributaries leading down to the flat river basin.
Above these marshes were two gravel-capped hills, now in the City of London – survivors of a 50ft-high gravel terrace that once ran parallel to the river.
These hills produced reliably dry positions from which to launch a boat. The need to cross from south to north – connecting Kent and Sussex, the counties where the Romans landed, with the rest of the country – was of paramount importance.
It’s telling that the Romans chose the further bank – i.e. the north – as the basis for Londinium, embattled with walls and a fortress; the area that still roughly marks the square mile of the City of London. Both those crucial gravel-capped hills were on the less marshy north side of the river. The marshiness of the river’s southern side also meant fewer train and Tube lines were dug through the subsoil – thought to be quicksand – as late as the nineteenth century. As a result, many more trains are run on viaducts through south London than they are north of the river.
The marshiness of the south bank explains why the north was settled first; and why the major administrative and commercial areas of the city were founded there. It also explains why south London has been unfashionable for centuries. This polarized approach just doesn’t happen in other cities like, say, Paris, where the Rives Gauche and Droite share the commercial, administrative and architectural honours.
That derogatory attitude to south London has been around for a while. Even in 1901, the novelist Walter Besant called south London ‘a city without a municipality, without a centre, without civic history … with no intellectual artistic, scientific, musical, literary centre, with no local patriotism or enthusiasm. One cannot imagine a man proud of New Cross.’
The north–south divide also explains why, Lambeth Palace apart, the majority of the city’s grandest buildings with the oldest origins – Westminster Abbey, the Houses of Parliament, St Paul’s Cathedral among them – are on the north bank of the Thames.
Those two gravel hills north of the river were the hub of what became Londinium, founded in around AD 50, seven years after Claudius’s armies invaded England. On one hill was the Temple of Venus – still London’s principal religious hill, it now supports the dome of St Paul’s, the fifth cathedral to stand on the site since AD 604.
The other hill was occupied by the heart of Roman government – the Forum. The spot continued to be strategically significant: the old Royal Exchange, now a series of upmarket shops, is on the site today, but, from 1565 to 1939, it was the City’s commercial centre, right next door to the Bank of England.
Between the two hills ran a stream called the Wall Brook; the street running along its course is still called Walbrook. Other small streams, like the Fleet at the bottom of my street, were buried later – as was the Holbourne river, whose deep valley now passes beneath Holborn Viaduct. The Westbourne was dammed to form the Serpentine; the Tyburn stream produced the lake in St James’s Park.
Beyond the confines of the City, other islands, poking above the marshes, were the first villages to be settled. Medieval London parish churches were scattered along the Thames. Their names – Battersea, Chelsea and Putney – reflect the Saxon for island, ‘ea’ or ‘ey’. Westminster Abbey, the site of a Saxon church, was also built on one of these ‘ey’ islands – Thorney Island. Eyots, small, mid-river islands, have the same etymological root.
Risk from flooding obviously increased the closer you got to the river. In the suburbs, villages like Wimbledon and Clapham Common were settled on higher ground, often ranged around flat grazing commons.
Ancient villages downstream from London – Erith, Purfleet, Gravesend, Woolwich, Greenwich – were built on chalk hills. In between these high points, the soft alluvial land was easily dug out to produce the London Docks. The Thames was the site of the first great enclosed docks in the country: the West India (1799–1806), the London (1800–1805) and the East India (1803–6).
Right up until the creation of the canals in the eighteenth century, roads were so unreliable that the Thames was one of the principal arteries of England, carrying stone, grain and timber in vast quantities. The Thames pumped life into towns all the way along the riverside – Oxford, Reading and Henley among them.
London’s significant political, religious and commercial buildings are still on the river bank or near it, including Parliament, Westminster Abbey and St Paul’s. And most royal palaces are on or close to the river: Hampton Court, Windsor, Whitehall, Buckingham Palace, Greenwich and Richmond.
Today, though, the Thames has never been so empty since before the Romans arrived. London may be built around the Thames, but it turns its back on the water. Londoners can go years without crossing the river or walking along it. Once a week, I bicycle across the river to teach in south London. As I thread my way through the traffic jams on Waterloo Bridge, I’m struck by how little traffic there is down below me, on what was once the busiest river in the world.
Still, for all its modern emptiness, the river remains at the heart of London’s foundation, as it does with other English cities. Manchester began life as the Roman fort Mancunium – thus Mancunians – in the first century AD, settled on a spur of land by the River Medlock. The medieval city was built on a sandstone hill where the River Irwell met the River Irk.
Leicester – the best spot to cross the River Soar – was a town for half a century before the Romans arrived in AD 43.
Cambridge, the highest navigable point of the Cam, began life as a river-crossing town, called Durolipons by the Romans, on a gravel ridge north of the River Cam. Four Roman roads crossed the river here: Akeman Street, running north-east to south-west; a connecting road to Ermine Street; the road to Braughing and Great Chesterford; and the road from Colchester to Leicester. Any traffic from further north was squeezed down towards Cambridge by the near-impenetrable Fens.
Southampton is where it is because the Channel narrows between Cap de la Hague, near Cherbourg, and Portland Bill. The city’s unique position – affected by tidal oscillations along the Channel, as well as across it – gives Southampton a double high tide. With three hours of high water, that still gives enough time to dock and turn round the biggest of ships.
There are usually further geological reasons why a particular riverside spot is chosen for an ancient settlement. Liverpool was settled on a peninsula between the Pool of Liverpool, a tidal creek, and the River Mersey; it was also built around a hill, a low one made of the same sandstone used to build much of the city.
Winchester was sited on a gravel spur projecting into the Itchen Valley; the cathedral crypt is still often flooded because it was built on the valley floor. Worcester was first settled in the fourth or fifth century BC, on a ridge of gravel and sand next to an ancient ford across the Severn.
York is on a terminal moraine – a ridge of accumulated glacial debris across the marshy plain – producing a dry east–west walkway above the ancient swamps, across what is now the Vale of York, itself formed by a lake dammed with ice. Where the Ouse crossed these heights, the Romans, in AD 71, built Eboracum, which became York.
Access to the River Avon, and from there into the Bristol Channel, is the main reason for Bristol’s site. The city nestles in the Avon valley, between, to the north, the Cotswolds, and the Mendip Hills to the south. Bristol was England’s main port for trade with Wales, the West Country, Ireland, Western Europe and America. The port specialized in West Country goods – leather, wheat, food and wool. The geology surrounding the port was helpful, too – with good building stone, alluvial clays, gravels, sands and coalfields nearby.
Bristol and Bath were the only Roman towns of any size in Somerset; in large part because of their site on the River Avon.
Bath had another reason for its prominence – it had the only naturally occurring hot springs in England. That’s why the Romans built a spa there in AD 43; why it became England’s first big modern spa town in the late seventeenth century, and why it later became the pre-eminent Georgian city in the country.
It’s also why Bath got both its modern name and its ancient one, Aquae Sulis – the Waters of Sulis, the Celtic goddess worshipped at the spring before the Romans came.
Earlier this year, I caught the last sunshine of a spring Bank Holiday weekend, and took a thirty-mile bike ride between Shotover Hill, Oxfordshire, and Ivinghoe, Buckinghamshire.
After climbing Shotover Hill, the road flattened out along a ridge, on the old route from Oxford to London. Ivinghoe Beacon was quite a climb, too, but the profile of the hill was slightly different: with a slightly concave slope leading up to the peak, dipping down to a flat ridge beyond.
I only know these things because I bicycled past them. If I had been Margaret Gelling, the leading historian of British place names and president of the English Place-Name Society for twelve years, who died in 2009, aged eighty-four, I wouldn’t have had to look at a map, or read a book, to work out what these places looked like.
Just from their names, Dr Gelling would have known not only what these places looked like, but also a pretty good chunk of their history, too. She would have known that the ‘over’ in Shotover derives from the Anglo-Saxon term ofer, meaning a flat-topped ridge with a convex shoulder. The ‘shot’ comes from sceot, meaning a steep place.
The ‘hoe’ in Ivinghoe is also a precise term. Literally meaning a heel, a hoe looks like an upside-down foot, with the heel sticking up in the air, and the concave indentation of the instep sloping down to ground level.
For thousands of years, this island has been intensively settled and cultivated by man – from Bronze Age man through Iron Age, Roman, Anglo-Saxon, Viking and Norman man. They left behind them a dense web of place names, describing the physical and sociological features of the place where they lived.
Even as much of England gets covered in concrete, tarmac and out-of-town shopping centres, those ancient place names survive beneath it all, giving clues as to how those places originally looked – and often still look.
The vast majority of English place names are derived from the Old English spoken by Germanic immigrants who came here in the fifth and sixth centuries, after the Romans left in AD 410.
There aren’t many Anglo-Saxon buildings remaining – there are thought to be only around 250 surviving churches with some Anglo-Saxon element in them. But the Anglo-Saxon names live on.
Our language was only given the Anglo-Saxon name ‘English’ – or englisc, in fact – in the ninth century. And it was under the Anglo-Saxons, too, that England started to develop a distinctive English look.
Not only did the Anglo-Saxons name most of modern England, but they also decided where most modern settlements are. Practically every modern village and town appears in the Domesday Book, a survey that largely consists of settlements founded before the Norman invasion.
Some Celtic terms for ancient landscape features, such as penn (head) and cruc (hill, mound or tumulus), were borrowed by the Anglo-Saxons from the Britons who were there before the Romans. The Thames comes from the Celtic Tamesa; the Avon from afon, meaning water, as does isca, the root of the Rivers Usk and Esk.
Celtic names survive in greater numbers in Wales and Cornwall, where the British fled the Anglo-Saxon advance. The popular Cornish prefix ‘Tre-’, as in Tremaine and Tregarn, comes from the Celtic trev or tre, meaning a village or homestead.1
But, otherwise, today’s place names were mostly first used 1,500 or so years ago by those Germanic immigrants, to describe landscape contours. The names practically always survive; as do the contours, unless they’ve been completely concealed by new settlements or motorways.
That connection between geography and name doesn’t exist in more modern countries that use borrowed names; like in, say, America, where Exeter, New Hampshire – unlike Exeter, Devon – doesn’t mean the castra, or Roman camp, on the River Exe.
In the mid-nineteenth century, the American writer Ralph Waldo Emerson said his home country was ‘whitewashed all over by unmeaning names, the cast-off clothes of the country from which its emigrants came; or named at a pinch from a psalm-tune’.2
To be fair, the American settlers weren’t always quite as random as that. Many of those American place names, particularly in New England, are directly borrowed from the East Anglian towns the settlers came from – such as Yarmouth, Ipswich, Haverhill, Cambridge and Boston. An estimated 71 per cent of the first Puritan pioneers to land in America in the early seventeenth century came from East Anglia.3
The scientific study of place names has been going on for about a century – and Dr Gelling was the most prolific and original of all place name historians. They are a meticulous group, keen on examining a small part of the country in detail, and looking into the origins of every village, farm and field. Most of them content themselves with studying one county over a lifetime. Dr Gelling mastered three.
For the English Place-Name Society, she wrote the official surveys for Oxfordshire in 1954 and Berkshire in 1976. At her death, she had done practically all the work for Shropshire.
Dr Gelling knew all the clues inside out, and the tiny differences between them. She knew that dun in a place name means not just a hill, but also a hill that was the site of a large village; that a settlement name incorporating versions of the word beorg tended to have a single farm or contain a church.
Leah – meaning clearing, wood, glade or forest – is by far the most common topographical term. Bexley, Bromley, Radley, Chorley, Headingley, Madingley, Washingley … all of them incorporate the term. Water also defines English place names to a great extent; another very common suffix is ‘ford’, as in Oxford and Hereford; ‘bridge’, as in Tunbridge Wells, Stockbridge and Trowbridge, isn’t far behind.
Once you know a few verbal tricks like this, place names begin to split naturally into their constituent parts. The Midlands are indeed Mid-Land. At Meriden, Warwickshire, there is a taper-ing cross which is supposed to mark the centre of England. Lambourne, Berkshire, splits easily into ‘lambs’ stream’. Burna is Old English for stream or brook (itself derived from the Old English broc). Burna is often used – in chalk country, in particular – to describe rivers that dry up intermittently.
Names can be extremely precise in their topographical descriptions: like the Gordano Valley, near Bristol, visible from the M5, meaning ‘an open valley with a triangular shape’; gelad, as in Crecca gelad, which morphed into Cricklade, Wiltshire, means a ‘difficult river crossing’. Sometimes, admittedly, place names can be a little more ambiguous. Endings derived from -ieg can refer to a place with drainage canals (like Wantage), or land liable to flooding.
You also get a feel for who founded a particular town or village. Norse words were incorporated into place names in the late ninth and tenth centuries, after the arrival of the Vikings. The Vikings landed on the east coast between the Thames and the Tyne and filled their place names with words like ‘beck’ (river), ‘fell’ (hill) and ‘thwaite’ (forest clearing). You’ll find them particularly in the Danelaw – the Viking-dominated land north-east of the line that runs roughly from Liverpool to the Thames Estuary.
Because of the powerful topographical effect on place names, villages with the same landscape end up with the same name, but with different pronunciations – they were named long before a standard pronunciation was settled on.
Gillingham – ‘a homestead of Gyllas family’, from Old English ham (village, homestead) and ingas (family, followers) – is pronounced with a soft ‘g’ in Kent, and a hard ‘g’ in Dorset. Mildenhall (‘a nook of land of a woman called Milde or a man called Milda’) is pronounced as it’s spelt in Suffolk. In Wiltshire, it’s pronounced ‘Minal’.
If only Dr Gelling had lived to see the row in the Kent village of Kenardington in July 2009. Victoria Cocking, a parish councillor in Church Road, Kenardington, was receiving misdirected post, meant for Church Road in neighbouring Kennington. And so she began a campaign to change the name of her street to Church Lane, although opponents of the move thought there were ulterior purposes.
‘It has always been Church Road – not a posh-sounding lane,’ said Geoff Cornes, another parish councillor, ‘The persons behind this change want to increase their social standing and up the value of their homes. Evidently Church Lane must have a nicer ring to it and be softer on delicate middle-class ears.’
Geoff Cornes was certainly right that there is a distinction between the terms, if not a class one. ‘Lane’ comes from the Old English, meaning narrow or hedged in. ‘Road’ is also derived from the Old English rad, meaning ‘a journey made by a horse’, i.e. a thoroughfare strong enough to withstand pounding hooves. ‘Road’ only became established in its current sense in the sixteenth century, used by, among others, Shakespeare (who spelt it ‘rode’).
Incidentally, Church Road, Kenardington, remained Church Road.
Dr Gelling wasn’t content, though, with merely compiling a list of suffixes and prefixes used in place names. She used these clues to compose an original history of how Britain developed over the last 1,500 years.
Before Dr Gelling came along, place name historians tended to think that place names were imposed by elites in a top-down way: thus names like Kington, Herefordshire – a king’s tun, or manor – or Knighton, Powys, meaning a knight’s tun.
Dr Gelling showed how many Anglo-Saxon place names must have been coined by ordinary locals, who named their villages after landscape features.
Where other historians concentrated on places that got their names from people’s houses and farms, she directed her attention to how the natural world inspired those names. Our Anglo-Saxon ancestors were skilled observers, adept at dreaming up accurate place names to fit the exact shape of the surrounding countryside.
Just as the Eskimos reputedly have dozens of different words for snow (in fact the Sami civilization in Arctic Scandinavia and Russia have many more), the Anglo-Saxons had forty different words for hill. And Dr Gelling worked out that they all meant distinct varieties of hill.
You can see why the Anglo-Saxons had to be so precise. In the days before maps or widespread literacy, the best way to direct someone to an unknown destination was to describe its exact physical features – ‘the tall, hump-backed hill by the stream’ or whatever it might be.
So, a place ending in -dun (the North and South Downs were originally Duns) meant a flat-topped hill that was easy to build on. Those ending in -hoh, like Ivinghoe, the destination of my bicycling trip, were sharply projecting, heel-shaped hills. Endings in -hop meant a valley or, more precisely, a remote place enclosed by hills.
Dr Gelling had one further triumph. She had always been convinced that the ancient Celtic Britons weren’t exiled to the fringes of the country by later invasions. Too many Celtic place names survive, predating the arrival of the Anglo-Saxons in the fifth century AD. Place names incorporating words like avon (river) or hamps (‘a dry stream in summer’) pointed to a continued Celtic presence.
Can you imagine, she said, Anglo-Saxons going up to the ancient Britons, and saying, ‘Look here, before I cut off your head, just tell me the name of this place.’4
It was more likely that the ancient Britons were allowed to stay on in their old haunts and stick to their old names. In 2001, Dr Gelling was proved right. DNA tests showed that most of us living in south England share DNA with pure-blooded Celts.
What’s in a name, Dr Gelling asked. And she got her answer – a staggering amount.