by Marc Morris
Again, we know these kinds of details because the processes of construction are detailed clearly in surviving records. The same records also give us the precise number and kind of workmen on each site. Most of the labour force were of the unskilled variety, press-ganged in the counties. There were, however, scores of skilled artisans on every site, including carpenters and plumbers, and in the later stages, glaziers. When work at Harlech was at its height in the summer of 1286, in addition to the 546 general labourers, there were 115 quarriers, 30 blacksmiths, 22 carpenters and 227 stonemasons.
Stonemasons were the key to the whole operation. They were skilled workers, and paid at the accordingly high rate of three to four pence a day. They were sometimes known as freemasons, not because they dressed up in aprons and indulged in arcane ceremonies, but because they had the ability to carve the more expensive ‘freestone’, which could be chiselled in any direction without splitting. Working any kind of stone was a slow process, especially if the patron demanded that every inch of a building was finished to the same high standard. Typically working in teams, masons travelled the country from one project to the next, and worked mostly in the warmer months of the year.
The written records, for all their exhaustive detail, are rather inexpressive when it comes to the actual processes involved in the construction of a castle. To understand these, we have to turn to the illustrations in contemporary manuscripts. These provide a wealth of additional information: how scaffolding was erected, how stone was carved, and how blocks of stone were moved around the site. Heavy blocks were either dragged on sleds, or wheeled from place to place in carts. The method of lifting them off the ground and up to the level of the builders was particularly ingenious – at each site, carpenters constructed several windlass cranes, powered by men in a treadmill. One such crane survives at the top of the spire in Salisbury Cathedral.
The treadmill crane at Salisbury cathedral.
Because the castles had to fulfil a vital military role, Edward drove construction on at lightning speed. Astounding as it seems, the records show that Conway Castle was substantially completed in just four years. Harlech scarcely took longer, and was finished by 1289. Only at Caernarfon did work drag on into the 1290s and beyond. The speed and scale of construction meant that these castles were not cheap. Harlech, with a price tag of £10,000, was the least expensive of the three. Conway, being larger and more sophisticated, cost something in the order of £15,000. When the builders finally put down their tools at Caernarfon, the project had absorbed at least £27,000.
So, from the summer of 1283, we have to imagine thousands of men being conscripted from all over England, marched to north-west Wales, and beginning to build Conway, Harlech and Caernarfon. The big question that remains is, who was in charge? Edward was of course the prime mover behind the project, but who masterminded the layout and design of the new castles, supervised the workforce, and ordered the materials? Today we would expect these jobs to be shared by a number of individuals – an architect would come up with the initial design, a surveyor would inspect and assess the site, a foreman would take charge of the labour force and so on. In the Middle Ages, however, all these tasks fell to one man. He was the master mason, a uniquely skilled and talented individual.
All kinds of jobs might fall to the master mason. He was responsible for sourcing the stone and devising the machines for lifting the blocks. He carried with him the ‘moulds’ or templates that were used to mark the uncut pieces of stone. Unlike a modern architect, he had no formal professional training; his skill with stone and his knowledge of geometry and mechanics were all acquired on the job. Master masons worked their way up from the ranks of the ordinary masons, and started their careers cutting blocks just like the rest. Moreover, although his skills might elevate him above his peers, a master mason was never entirely removed from the workplace. Even though he might conceive of the design for a whole building, and have hundreds of men working under him, he was not confined to an office; he might still be found, chisel in hand, working alongside his fellows.
Who was Edward I’s master mason? Until recently, all that historians had to conjure with was a name: Master James of St Georges. He first occurs in the records in the spring of 1278, when he appears at Flint and Rhuddlan ‘to ordain the works of the castles there’. After the second Welsh war, he is styled ‘Master of the King’s Works in Wales’, and receives the very high salary of three shillings a day for his efforts. Clearly this is our master mason. But who was he? He appears in 1278 out of nowhere. Where had he come from? Nobody could say until, just sixty years ago, a historian called Arnold Taylor became Chief Inspector of Welsh Monuments (1946–55), and set out to solve the mystery.
In the course of his work, Taylor came to know Edward’s Welsh castles inside out. He soon became puzzled by some unusual and apparently unique features. In the first place, he noticed that there were small square holes in the sides of the towers. This in itself was perfectly normal; you find such holes in the sides of castles everywhere. They were for the wooden scaffolding supports or joists to fit into while the castle was being built, and are known as ‘putlog holes’ because (wait for it) they are the holes where they put the logs.
Something, however, was not quite right about the ones at Conway and Harlech. Normally they wrapped around the walls or towers of a castle at the same horizontal level. At Edward’s Welsh castles, however, the holes spiralled around the towers, suggesting that the original scaffolding had been sloped, rather like the slide on a helter-skelter.
There were other oddities. Conway and Harlech had several archways that were perfectly semicircular. This might not sound especially strange, but Taylor knew from his experience of other castles in England and Wales that such archways were not to be found elsewhere in the UK on buildings of this date. Similarly, Taylor realized that the windows in the great gatehouse at Harlech, as well as being exceedingly large and handsome, were apparently unique. Despite his extensive knowledge of castles, he had never seen anything like them. And then there were the toilets. Nothing unusual there, you might think. Taylor, however, was intrigued by the peculiar projecting funnel-shaped design he saw at Harlech. Again, he’d never seen anything else quite like it.
Could these clues – putlog holes, archways, windows and toilets – help reveal the identity of Master James of St Georges? Taylor decided to find out. The evidence in the architecture seemed to suggest that, whatever else he might be, Master James was certainly not British. And so, in the autumn of 1950, Taylor left Britain on a quest to find him. Following what was no more than a strong hunch, he headed for Switzerland, and the tiny alpine province of Savoy.
Today, Savoy is a region shared between Switzerland, Italy and France. In the thirteenth century, however, it was an independent state, ruled by a dynasty of counts. Although small in size, its position made it powerful. The counts of Savoy controlled the Alpine passes between France and Italy, and therefore controlled traffic and communication between the kings of England and France to the west, and the Emperor and the Pope to the east.
Even so, it might seem an odd place to go looking for clues about Welsh architecture. Taylor, however, had done his homework. He knew, for example, that since the middle of the thirteenth century, there had been strong ties between the counts of Savoy and the English royal family. Precisely because of their powerful role as brokers of international relations, Henry III had married into the Savoyard family and, for a time, his court was dominated by his relatives from Savoy. Edward grew up in the company of Savoyard uncles and cousins and, when he was king, continued to cultivate the connection between the two dynasties. Many of his best friends were from Savoy, including the great Otto de Grandson, whom Edward put in charge of north Wales after the conquest.
Having reached Savoy, Taylor headed for the eastern edge of Lake Geneva, and the castle of Chillon. As Taylor says in his writings, ‘It is easier to remember than to communicate one’s impressions on visiting this marvellous building for t
he first time.’ The castle stands on a small island right by the edge of the lake, and seems to rise directly out of the water, like a ship in harbour. Unlike Edward’s castles, it has been much restored and rebuilt since the thirteenth century, but what it occasionally lacks in medieval authenticity it more than compensates for in overall ambience. Seen from the west, framed by the snowy peaks of the Swiss Alps, the castle and its setting are nothing less than breathtaking.
Windows at Harlech
Chillon.
The castle at Chillon was originally established in the first half of the twelfth century, but most of the existing walls and towers were built in the middle of the thirteenth century, at the command of the then count, Peter of Savoy. It was here that Arnold Taylor made his first major discovery. Several of the windows, although much restored in the late nineteenth century, were identical in design to those in the gatehouse at Harlech. Not only were they a similar shape; they were exactly the same size. When Taylor measured them from top to bottom, he found that they differed in height from their Welsh equivalents by just a quarter of an inch. Two sets of windows, a thousand miles apart, but with less than half an inch between them – a big clue, surely, that Taylor was on to something.
Other discoveries soon followed. Heading further into the Alps, Taylor came to two smaller castles at La Batiaz and Saillon. Both were built in the thirteenth century by Peter of Savoy as part of his struggle with the bishops of Sion to control the Rhône valley. As castles, neither is as striking as Chillon, but their locations are similarly stunning. La Batiaz is perched dangerously on top of a spur of rock, many hundreds of feet above the small town of Martigny. It was here that Taylor found his second major clue. On the side of the castle walls is a pair of projecting garderobes, built to exactly the same funnel-shaped design as the solitary example at Harlech. Subsequent investigations have uncovered no similar examples. In other words, Taylor had stumbled across another rock-solid, utterly distinctive architectural parallel between the castles of Wales and Savoy.
Garderobes at Harlech
La Batiaz
Archways at Saillon
Harlech.
The final pieces of architectural evidence were located a few miles up the valley in Saillon. A marvellous little place, the town still has an authentic medieval feel, and is built at a ridiculously steep angle on a small foothill of the Alps. Little remains of Peter of Savoy’s castle today – just a single round tower. The town walls, however, were also built at that time, and contained all the clues Taylor needed. The towers along the walls had the same spiralling putlog holes as those at Conway and Harlech, and the arches of the town gates were of the same semicircular design as the Welsh ones. Furthermore, beyond the clinching evidence of these identical features, the appearance of the walls as a whole gave Taylor an unmistakable feeling of déjà vu: seen from a distance, they are entirely reminiscent of the town walls at Conway.
All the unusual features of Edward I’s Welsh castles, therefore, had identical counterparts in medieval Savoy. What was the connection? The counts of Savoy, like the kings of England, kept detailed financial records of their building projects, and luckily these records still survive in the archives at Turin. It was there that Taylor found the answer. When he unrolled the fragile accounts for the 1260s, he found that Peter of Savoy’s castles had been built by two men, a father and son team, called Master John and Master James. The later accounts showed Master James working alone and, in particular, working on a brand new castle site, owned by the counts of Savoy but located over a hundred miles away near the French city of Lyon. The name of this castle was St Georges-d’Esperanche. This, Taylor recalled with delight, was the very place that, in 1273, Edward I had stopped on his way back from crusade to visit his Savoyard cousin. It seems very likely, since work on the castle was still underway at the time, that the king may have been introduced to a skilled master mason and castle-builder par excellence, who went by the name of Master James of St Georges.
The town walls at Conway
Saillon.
Arnold Taylor’s hunch had paid off – he had discovered the true identity and origins of Master James, and proved beyond question that he was one of the world’s greatest architects. Not only was he responsible for the great royal castles in Wales; he was also the designer of a string of castles in Savoy. When you look at his early handiwork, you can see how Edward I, travelling through the Alps in 1273, must have been impressed, not to say envious. It is always slightly galling when you drop in on relatives and discover that their new house is quite a lot nicer than yours. Sitting in the magnificent chambers of Chillon Castle, gazing at Lake Geneva through the ‘Harlech’ windows, one is tempted to think the unthinkable. Was Edward really driven by necessity and outraged honour, as he would always claim? Or did he just want a lovely new set of fabulous castles in the mountains, like his cousins?
We will never know. What we do know is that when Edward formulated his castle-building plan in 1278, there was only one man for the job. The king summoned Master James and put him in charge of the entire operation. Since the results were Conway, Harlech and Caernarfon, we can say with some certainty that his faith was not misplaced. Edward was clearly delighted with the castles because, as well as paying his architect a handsome salary, he also gave him a wonderful gift. In 1290, he appointed Master James as constable of Harlech castle. What better reward for the architect than to live in the magnificent gatehouse that he himself had created?
Harlech is in some respects Master James’s most striking castle, owing largely to its situation on a great outcrop of rock, with views out to sea to the west and vistas of Mount Snowdon to the east. Impressive as it is, however, the castle was intended to house only a garrison and a resident constable – it would have been a little pressed to accommodate the king and his court. At Conway and Caernarfon, though, the story was different. These were castles built as palaces, where Master James designed and built spectacular suites of rooms for the king, his family and the royal household.
The best place to appreciate the luxury of Master James’s designs is at Conway, since its royal apartments still survive. Domestic comfort has taken several steps forward since the twelfth century and its great towers. Not only do we find plenty of grand fireplaces, but also many more toilets – even the tiny watching-chamber for the king’s chapel is thoughtfully provided with en suite facilities. The biggest advance, however, is in terms of windows. Although the windows on the outside of the castle are few and tiny, the windows on the inner courtyards are large and stately, and would have been fitted throughout with stained glass of many colours.
Soldiers using the crossbow-loops at Caernarfon.
At Caernarfon, meanwhile, you can appreciate the military sophistication of Master James of St Georges’ design. On the exposed southern side of the castle, two passageways run through the curtain wall for its entire length, one above the other, punctuated at regular intervals with crossbow-loops. The garrison, therefore, with archers and crossbowmen shooting from within the walls as well as behind the battlements on top, had treble the defensive force. On the north side of the castle, the arrangement of the defences is even more ingenious. What from the outside seems to be a bank of five conventional crossbow-loops is revealed on the inside to be a cunning arrangement of interlocking slots, with each loop capable of being shared by up to three archers. The result, combined again with the potential number of men shooting from the battlements and the towers, has been justly called ‘one of the most formidable concentrations of “fire-power” to be found in the Middle Ages’.
The best place, however, to stand and admire the military might of Caernarfon Castle is at the main entrance, the so-called King’s Gate. Any would-be attacker trying to get in this way was faced with a daunting set of obstacles. In the first place, there was the great ditch in front of the gate, which could only be crossed by means of the drawbridge. If, by some miracle, the attacker managed to get past that, he was confronted by five sets of sturdy o
ak doors, alternated with no less than six portcullises. Above his head were holes in the ceiling, called meutrières or murder-holes, conventionally believed to have been for boiling oil or rocks, but now generally (and rather boringly) considered to have been water chutes, for use if a fire was started in front of the gates. Even if we grudgingly accept this new hypothesis, we can still draw comfort from the fact that the gatehouse is liberally provided with crossbow loops and flanked by the other mural towers and wall-walks on this side of the castle. Boiling oil or no boiling oil, our would-be assailant was going to have a hard time getting in this way.
The King’s Gate is unquestionably a mighty piece of military thinking, but its functions are not purely practical. Above the doors is a carved statue of a king, which makes Caernarfon the earliest example of a castle with an ornamented gatehouse. As with the forebuildings on our twelfth-century castles, this is an entrance designed to impress the friendly visitor as well as deter the attacker. Some of the military paraphernalia can also be read in this way. One portcullis, for example, is a jolly good idea, and two portcullises might likewise be interpreted as very prudent. But does anyone really need six portcullises or, for that matter, five sets of oak doors? Another way of looking at these barriers is to see them as part of a ceremonial procession route. We have to imagine a distinguished visitor entering the castle, and being treated to the spectacle of doors swinging open and portcullises being raised in an intentionally elaborate ballet. It seems clear that a deliberate degree of theatricality has been incorporated into the architect’s design.