Archaeological Excavation of Eilean Dearg Castle
During the years 1964-1965-1967-1967 Hugo B. Millar lead an archaeological dig at Eilean Dearg. The archeological headquarters were at Port-an-Eilean and the archeological team was ferried to and from the island by Neil Clark from Colintraive. The article below was written by Hugo B. Millar and published by the Scots Magazine in March 1971 under the title “From stone-age man to the Victorians – who would have thought this scrap of land could hold so many secrets?”.
Last month, you heard something of Eilean Dearg’s history, with some tales and stories of its owners, occupiers and others, from the fourteenth century to the seventeenth. The saga of its castle came to a sudden and abrupt end with its destruction by the Navy after Argyll’s invasion of 1685; a vast explosion that blew it completely off its rocky site. All that one could see, nowadays, was a little section of wall and a great deal of luxuriant vegetation on a bare half—acre of rock.
It was, however, just possible that, because of the Norse activities in the area, there might be an earlier basis for the castle development, hence the idea of an archaeological “dig”. It would be interesting, also, to compare the written word of history, with the factual remains on the ground and see how they tallied — if there were any remains to dig . . . Finally, there was the thought of a pleasant holiday in the famous Kyles, and a site that was “far from the madding crowd”.
This, then, was the site which we started to excavate in the summer of 1964 and, almost at once, made an important discovery. This was that it rains in Argyll, like everywhere else, only worse . . . And that rain, rocks, gulls and growth make an unpleasant mixture . . . Our pleasant archaeological holiday turned into one of the most difficult “digs” I, at least, had ever undertaken, and the single season’s work we had contemplated, took five seasons; at the end of that time we had worked as hard as any miner, and with spectacular effect. We reckoned, in fact, that, altogether, we had about dug up half the island and put it on the other half — then shifted most of it back again!
Everywhere we met with the results of the explosion. The whole island was covered with a layer of rubble, in places many feet deep, and the various occupational “layers”, so useful to the archaeologist, had been hopelessly mixed up, like a “dog’s breakfast”. Heavy foundations had shifted considerably, and had even been blown clean off the bedrock in many places. Over everything was a deep-rooted vegetable growth, richly nurtured with the guano of many generations of seabirds.
We started by laying down a grid of pegs, in ten-foot squares, in order to plan the castle as we found it and pin-point, exactly, the position of each artifact as it was discovered. Hopefully, we soaked the island in a strong solution of weedkiller the Easter before digging commenced. This reduced the growth only slightly! In future seasons, we confined ourselves to specified areas, and gave these a thorough drenching of double-strength solution, applied twice. Even then, we had to use cutting implements in places, for such is the growth in Argyll and Eilean Dearg in particular!
Bit by bit, over the seasons, the pattern of the castle emerged — and a surprising pattern it proved to be. Our first major discovery was the site of a chapel, and a largish chapel at that, carefully oriented east-west, standing against the north wall of the castle. We found the altar base, to start with, freestanding beside the east gable and later, the doorway or remains of it, near the west end of the south wall. There was clear evidence that the door had swung on pivots, not on hinges. This puzzled us. Eventually the puzzle became clear, for we found that the chapel had been there long before the castle and was apparently of twelfth century date.
The castle itself had started in the late fourteenth century, as might have been expected from its history, in the shape of a great rectangular tower, built on the western half of the island, and having walls of eight feet in thickness. Colin had built well, indeed! It must have stood at least fifty feet in height.
The tower had been shut off from the rest of the island by a heavy wall, running from north to south, overriding the western gable of the chapel, which now stood outside the small court thus enclosed. The chapel had continued in use, however, for the south wall was “butted” against its new gable with mortar. About a hundred years later, a hall, built over cellar-age, was erected on the south side of the eastern half of the island, and to its eastern end was attached a curtain wall running northwards and containing a great gate between itself and the chapel’s eastern gable.
At this point, the north-south wall, in the area between the chapel and the new hall, was demolished, thus forming one large court to the interior which was then cobbled in its eastern half. The great gate I referred to, was placed opposite the northern landing place; at that point the rock had been scarped to allow for the birlinns, or small galleys of the period, to be dragged clear into the court; the gate had been built over nine feet wide to accommodate them. On the cobbles of the court, underneath all the rubble and turf, were found a number of square-headed clinch nails, some with diamond-shaped rivets still attached, indicating that the old Norse form of boat construction, which used this type of nail, was still in fashion as late as the sixteenth century, when we reckoned that the courtyard had been completed. These nails and the great gate, which is, in fact, a true sea-gate, were the only Norse evidence we found, and due, of course, to influence only.
Some time in the late sixteenth century, no doubt as a result of the Reformation, the chapel had been abandoned as such and re-used as a builder’s yard for castle repair work; we found heaps of stone-masons’ chippings inside. Later still, for possibly as long as fifty years and in the early part of the seventeenth century, the castle had become disused, as a heavy growth of turf, &c., had risen over the courtyard cobbling.
In the re-occupational period that followed, the folk concerned did not know, or had forgotten about, the cobbling, now hidden from sight. They had obviously found the courtyard muddy, as they had built a number of cobbled walkways on top of the new turf growth, to and from internal doorways and other openings, so that they wouldn’t get their feet wet! A little digging would have saved them a lot of trouble.
As we dug down into the castle site in the first two seasons, we kept finding odd chunks of vitrified rock, which puzzled us. Vitrified rock is stone which has fused and run in semi-liquid form as a result of great heat. It was not slag, as from old iron workings, but true vitrification and it began to dawn on us that our chapel and castle were not the first buildings on the island. Eventually we found masses of this rock, some with mortar adhering, as they had been re-used in the castle walls. It became clear that there had existed, in the early Dark Ages, a vitrified fort on the site of the castle, many centuries before. This type of fort used stone-walling laced with timber; when the walling was fired, the heat was so great that the stone fused and compacted together. When this happened to sandstone, a peculiar kind of glassy surface was created, quite unmistakable to the archaeologist. We found many chunks like this. It is generally thought, nowadays, that such vitrification was caused by accident, possibly when a fort was attacked and set on fire.
When excavating in the older courtyard area, west of the north-south wall, we came on a flint of a strange shape and, later, a number of flint flakes scattered over a widish area. These were taken to Kelvingrove Museum, Glasgow, where the large flint was pronounced to be an excellent example of a flint scraper, much to our surprise. This, together with the flakes, indicated that our site was older still than the Dark Age period and was, indeed, Neolithic, datable to possibly 2000 B.C. This meant that our island had been occupied by the earliest man in Scotland, who arrived here after the last Ice Age. Unfortunately, no other trace of this, or the vitrified period, were found; each occupation had obviously gone a long way towards destroying the traces of the previous one, and our great explosion of 1685 had completed the task, mixing all the periods up.
The last “layer” of all to be described, and one that afforded us some amusement, was found almost immediately during the first season. In the turf of the topsoil, some broken and rotted timbering was discovered with a few iron “holding-down” bolts of a recent type. It transpired that this was the remains of a wooden platform, erected towards the end of the Victorian period when the Clyde steamer traffic for tourists was in its heyday, and all kinds of “gimmicks” were being tried by the competing companies for the trade. One of these was a “dancing floor”, constructed on the remains of the castle, where competitions were held in Highland dancing, to the music of the pipes, for the amusement and entertainment of the steamer, passengers. Not only dancing, apparently, for, on a level area on the site of the tower, we found a few Victorian pennies. Some people had been indulging in a “Pitch and Toss” school as well!
Outside the sea-gate and just west of the west curtain wall, we found the well. We knew there had to be one, of course; no one, throughout the ages would inhabit an island site without a source of fresh water. We had expected to find it inside the court, but eventually tracked it down to this damp and heavily overgrown area outside the gate. It was not a well in the true sense of the word, but a rock-cut tank of irregular shape, trapping a small spring which had a flow of approximately half a gallon every two minutes. The tank had the remains — only a few stones — of a rough perimeter wall, and was completely choked with earth and very evil-smelling mud. Every cubic foot of this was dug out and washed at the shore in case any artifacts escaped us. The result, for all our very dirty labours, was disappointing as only a few odds and ends were raked up of no great interest.
All over the castle area were found the artifacts of its history. “Green Glaze” medieval pottery sherds came up in profusion, going back in date to the fourteenth century. Masses of rusted ironwork of all kinds were uncovered, some pieces so far gone as to be unidentifiable. Others were once door bars or fragments thereof, hinges and hinge pins, knives, both table and pocket, and nails in profusion — roofing, slate and general purpose — all hand-hammered. Only wrought iron, of course, will survive for long in the open, unless in abnormal conditions. Some time after we had left the island, more ironwork was discovered. A passing yachtsman landed with his family, curious to see the results of our work. His children noticed something sticking out of the side of one of our deliberately collapsed trenches, washed, clear by the subsequent rain. After a great deal of tugging and scraping, they triumphantly pulled out a tangled mass of very rusty iron bars! The yachtsman located my address and was good enough to pass the find over for examination. After cleaning and straightening, it turned out to be the remains of a seventeenth century fire-basket, blown clean out of its setting, where no doubt it had once toasted the toes of its worthy Campbell owners.
Pieces of window glass, and the strips of lead that held them in position, were found in various places around the site. Glass was scarce in these days, and windows were glazed in the top half only as a rule, the lower half being closed with a shutter. Most of the glass found was coloured.
One bronze object discovered near the hall was a small coiled-metal ring, identified as the type used for connecting panels of chain mail together. Another bronze artifact, near the first, was the pin from an annular brooch, such as was used to attach the folds of a light cloak or plaid to the shoulder.
In a trench beside the north gable of the tower we came on a find of great interest. Among the masses of fallen masonry pulled out, we recognised the stonework of what had been a little wall—cupboard of the type called an aumbry. Its contents were scattered around and consisted of about 40 gaming counters, made of slate, quite round and flat and varying in size from quarter of an inch in diameter to an inch and a quarter. Some had incised markings on them. Many were, unfortunately, broken, but a number of whole ones were recovered. A few such counters have been found in the past, singly or in twos, on different archaeological sites. It had been conjectured that they were gaming counters, but nobody had been sure. Now we knew for certain. They obviously belonged to the last occupational period — that of Argyll’s garrison — and show us that soldiers everywhere in any age, will indulge in games of chance !
The best area of all for finds proved to be just south of the south gable of the tower. Here, at one time, had been a small open court, carefully cobbled and drained, squeezed in between the tower and the south curtain wall. It had been walled-off from the rest of the castle and had a private door of access from the tower basement — we found a side of the wall passage still in place. No doubt it had been intended to form a private pleasaunce for the laird where he could sun himself in seclusion! Later, the court had been roofed-over with possibly more than one floor erected on top. We found the ash of the timber concerned all over the cobbles. Mixed up with this ash were the artifacts from the floor or floors above. They indicated that the area had been used as an armourer’s shop, with maybe a store on the floor above. First of all there were two pieces of a bronze platter, with turning marks on its base. The platter had apparently been broken in the explosion. This was a surprise, as no one had any idea up till now that such platters were being made in bronze in the late seventeenth century, so it was an important find domestically.
Next came a little whorl of steatite, beautifully made, with four tiny holes at the cardinal points, arranged round the usual central hole. There has been much speculation as to the purpose of these whorls, which have been recovered from “digs” of all periods. Whatever its original purpose, this one had apparently been used later as a decorative charm, with possibly little jewels or coloured stones set into the additional holes.
Afterwards came two pieces of corroded metal which, at first sight, looked like pairs of scissors. Some careful cleaning on one pair the same evening at Port-an-Eilein, our “Dig headquarters”, revealed that it was made of brass, and could be dividers rather than scissors. If so, this was a find indeed, for such instruments are extremely rare and our knowledge of them comes mainly from illustrations on old Dutch charts. They were carefully taken to the Royal Scottish Museum in Edinburgh for expert cleaning and assessment. The museum confirmed our view that they were indeed two pairs of early Dutch marine dividers, in good condition under the verdigris. The curator concerned was greatly impressed by them and wrote a special paper on the subject. No doubt they had come across from Holland with Argyll’s little squadron.
The next discoveries to be mentioned also came from Holland and proved the most interesting of all; not only for their own sake, but also for what they mean, historically speaking. One by one, from all over the cobbled floor, we picked up musket flints; more were found among the fallen rubble down the rock slope. In the south, just outside this enclosed area, where they had been blown clear. In all, we collected over 100. With them were found a dozen musket balls. These latter were moulded in two halves; the dividing line was clearly seen in each case, running round the circumference of each ball. Clearly, great care had been taken in their manufacture. Their diameter was three-quarters of an inch. Musket flints are in themselves a rarity, and this quantity represents, as far as I know, the largest find recorded from any site. But they meant much more than that. Musket flints — and moulded balls — from as early as 1685 — were a rarity indeed. At this time, flint-lock muskets were something new, a revolutionary idea which was just coming on the market. The usual type of musket was the match-lock, which required a slow-burning fuse to fire it; a cumbersome process for troops in the field, and almost impossible to use in wet weather. The flint-lock produced its own spark and was a great advance. In 1685 such weapons were rare and mainly in private ownership; they were expensive items and highly prized. The Privy Council records indicate that a few parcels of 200 to 300 had been bought for the forces, as they became available, since 1680. The Dutch, of course, were the leading manufacturers in this field and it was only from them that such quantities could be obtained. With the accession of William of Orange and his consort, Mary, to the throne, the Low Country armouries were open to Britain and the flint-lock became the standard weapon in the forces for 100 years—we know it as the famous “Brown Bess”. These, then, were the muskets “of new invention” on which the naval observer had remarked when the arms were being transferred to Captain Hamilton’s ships before the final explosion ; 3000 stand were removed in this way, and at that time, Argyll had about 2000 troops in the field. This represents a very large purchase indeed; the Dutch armouries must have been stripped for the purpose and the whole country scoured to make up the total. The authorities, in fact, must have co-operated fully in the enterprise, under the direct instructions of William himself. The historical mystery had been cleared up at last, the answer hidden for all these years under the turf and rubble of the “Red lsland”.
There is a saying in archaeological circles that you never know what you are going to find on a dig until you start digging. This was certainly true of Eilean Dearg; the last find alone justified the whole operation.
We had set out originally to find a Norse castle, but that misty dream slowly evaporated, leaving us instead with a residue of many other things which were quite unexpected. Each year had brought its own surprises and, at the end of it all, what we had discovered was a microcosm of Scottish history, from Stone-Age Man to the Gay Victorians. Who would have thought that one little island could have held so many secrets ?
The seagulls are once more in undisputed possession and, in a year or so, they, and Mother Nature between them will have completely healed the scars which we left behind. Yachtsmen landing on the place will find little to interest them . . .