FOOTNOTES:
[407] Collectanea Antiqua, vol. i. p. 144.
[408] Roy's Military Antiquities, p. 116.
[409] Archæol. Scot. vol. ii.
[410] Historical Inquiries, p. 41.
[411] Biblo. Topog. Britan. vol. ii. p. 385.
[412] Caledonia Romana, p. 270. I am informed, however, by Sir George Clerk, Bart., of Penicuick, that the author of the Itinerarium Septentrionale was originally a teacher of music at Aberdeen, and according to the traditions of the Penicuick family, was usually known by the name of Galgacus, being no doubt apt to carry his enthusiasm for his favourite hero of Mons Grampius to an extent somewhat amusing, if not troublesome, to friends and patrons.
[413] Roy's Military Antiquities, p. 152.
[414] Caledonia Romana, p. 305.
[415] The preservation of this Scoto-Roman relic is due to the zeal of John Buchanan, Esq., its present possessor, who secured it after it had been in vain offered to the curators of the Hunterian Museum, as an appropriate addition to its Roman collection.
[416] Wordsworth.
[417] C. R. Smith, no mean authority on such a subject, defends the authenticity of Richard of Cirencester in his recent valuable work on "The Antiquities of Richborough, Reculver, and Lime," p. 177. The illustrations of our Northern itinera have led me to an opposite conclusion; but even should the genuineness of the "De Situ Britanniæ" be established, its value to Northern antiquaries must still be open to question.
[418] Roy's Military Antiquities, p. 116.
[419] Caledonia Romana, p. 150.
[421] Ante, p. [171].—I am indebted for much of the information relative to the recent discoveries at Newstead, to the notes and personal observations of Dr. J. A. Smith, and to John Miller, Esq., C.E., in whose possession the spear now is. The dimensions of the skull are given in the table of cranial measurements, p. [166], No. 16.
[422] D. M. Moir, (Delta,) in a communication to the Society of Antiquaries of Scotland.
[423] Gough's Camden, vol. iii. p. 304.
[424] Sibbald's Historical Inquiry, p. 41.
[425] Itinerarium Septentrionale, Appendix, pp. 180-183.
[426] I owe this information to Mr. A. Handyside Ritchie, the well-known sculptor, who examined the Roman ware while in Mr. Sivright's collection. Probably all record of its locality has been lost sight of by its new possessor, if indeed it has been preserved.
[427] In 1846 Mr. Brown presented to the Society of Antiquaries of Scotland "a stone ball, found at the Trinity Hospital, three feet below the surface, and upon a piece causeway." Minutes of Society, 21st Dec. 1846.
[428] Memorials of Edin. vol. ii. p. 176.
[429] Itiner. Septent. p. 117.
[430] Memorials of Edinburgh, vol. ii. p. 50.
[431] "About this time it would appear that Julia, the wife of Severus, and the greatest part of the imperial family, were in the country of Caledonia; for Xephilin, from Dio, mentions a very remarkable occurrence which then happened to the Empress Julia and the wife of Argentocoxus, a Caledonian," &c.—Itiner. Septent. p. 104.
[432] Memorials of Edinburgh, vol. ii. p. 34.
[433] Biblio. Topog. Britan. vol. ii. p. 348.
[434] Itiner. Septent. p. 57.
[435] Sibbald, p. 83; Itiner. Septent. p. 117; Horsley, p. 205; Wood's Cramond, p. 4; Caledon. Romana, p. 163.
[436] Archæologia, vol. xviii. p. 120.
[437] Silurian System, p. 279.
[438] Archæol. Journal, vol. v. p. 226.
[439] Archæol. Journal, vol. vi. p. 181.
[440] A PALO may either refer to the school of the Mediciner, and signify of Palermo, Palatino, or the like; or, as is perhaps more probable, it is the Crocodes of Palermo, or elsewhere. Between sixty and seventy of these medicine stamps are now known, and two specimens of pottery have been found in France impressed with similar prescriptions,—evidently the vessels in which the preparations were preserved.
[441] Vol. xvi. Plate LI.
[442] Vol. xi. p. 105.
[443] Camden, p. 834.
[444] Pennant's Tour, vol. iii. p. 411.
[445] Caledonia Romana, p. 129.
[446] Itiner. Septent. p. 84.
[447] Amphoræ.
[448] Mortaria.
[449] M. A. Lower on the Manufacture of Iron in Britain by the Romans. Journal of the Archæological Association, vol. iv. p. 265.
[450] New Statistical Account, vol. iv., Kirkcudbrightshire, p. 159.
[451] Sinclair's Statistical Account, vol. iii. p. 458.
[452] Logan's Scottish Gael, vol. ii. p. 195.
[453] Stuart's Costume of the Clans, Introd. p. li.
CHAPTER III.
STRONGHOLDS.
Next to the sepulchral monuments and the temples of remote ages, the fortifications frequently furnish the most durable and characteristic evidences of skill, and of the civilisation of the era to which they belong. In the Great Valley of the Mississippi, after Anglo-Saxon colonists have for upwards of two centuries been effecting settlements on the soil of the Red Indian, and obliterating every trace of him by their more enduring arts, the burial mounds and the forts of a race still older than the Red Indian remain to attest the pre-existence of civilisation in the American continent. Here, too, where for nine centuries at least, we can find authentic records of builders, sculptors, ecclesiastical architects, and military engineers, fashioning rude materials into goodly fabrics, of which traces are still discernible: we also can discover the wrecks of older structures reared in those dim and remote eras, into the secrets of which we long to penetrate. "How cold is all history, how lifeless all imagery, compared to that which the living nation writes, and the uncorrupted marble bears! How many pages of doubtful record might we not often spare, for a few stones left one upon another! The ambition of the old Babel-builders was well directed for this world. There are but two strong conquerors of the forgetfulness of men, Poetry and Architecture; and the latter in some sort includes the former, and is mightier in its reality. It is well to have not only what men have thought and felt, but what their hands have handled, and their strength wrought, and their eyes beheld all the days of their life. The age of Homer is surrounded with darkness, his very personality with doubt. Not so that of Pericles: and the day is coming when we shall confess that we have learned more of Greece out of the crumbled fragments of her sculpture, than even from her sweet singers or soldier historians."[454] The Scottish "Caterthun" is no Athenian Acropolis, and our monolithic temples, though not ineloquent memorials of their builders, must rank with the primeval cyclopean structures of Greece, and not with her Parthenon or Colonna. But the aboriginal strongholds, though mostly of a sufficiently rude and primitive character, must not be overlooked in reviewing those "conquerors of the forgetfulness of men." The construction of offensive and defensive weapons is one of the earliest evidences afforded by man, in a savage state, of that intelligence and design by which he is distinguished from the brutes. Domestic and social relationships follow, from whence spring society, ranks, laws, and all the primary elements of civilisation. Among the first indications of such progress is the union for mutual defence, and the erection of strongholds for the safety of the community and the protection of property when threatened by invading foes. Herein lie the essential rudiments of a commonwealth, when the weal of the community and of its individual members have been recognised as the same.
A very slight review of the more simple class of British hill-forts will suffice, since we fortunately possess, in many of the contemporary records already described, more precise and definite history than they can now yield. It is for this reason that all notice of the aboriginal strongholds has been reserved till now, though it cannot admit of doubt that some of the simplest of them are contemporary with the pit-dwellings of the Stone Period, while others manifest such improvements as seem best to accord with the arts and weapons of the Archaic era which succeeded. Of these we have the circumscribed mote-hill or earthen-mound, steeply escarped, and with the remains of its little vallum of earth surmounted originally by the stronger palisades for which the neighbouring forest supplied abundant material. Nearly akin to these are the small circular forts of earth and loose stones which still crown the summits of so many Scottish hills; their lofty sites having secured them from the inroads of the agriculturist, while his aggressive ploughshare has obliterated all traces of the far more skilfully constructed Roman camp and military road which once occupied the neighbouring valleys. Within the area of some of these, or scattered about their neighbourhood, flint arrows and other primitive weapons have been frequently found, accompanied occasionally by more valuable relics. On removing, in 1830, the rich black mould nearly filling the trenches of three such forts, the remains of which still crown the ridge of a rising ground above the valley of Dalrymple, Ayrshire, human skulls and bones, deer's horns, and a horn-lance or spear-head of primitive type, were discovered. Similar records of the aboriginal fort-builders must no doubt frequently be turned up in the course of agricultural operations: but they can only tell us what is already obvious, that this class of strongholds, or duns, as they are locally termed, pertain to a people whose arts were still in their infancy. Some, however, of the small hill-forts must be regarded as the mere temporary lodge-ments of British outposts, in times of actual open war. Of this class probably are the earth-works on the summit of Birrenswork Hill, in Annandale, while the more extensive intrenchments of the Roman legions occupy the level areas at its base. Similar works are also to be met with in the Western Highlands. At Knoc Scalbert, near Campbelton, Argyleshire, is a fort of larger size and more complicated design, covering an area of about fifty paces in diameter; but the neighbouring heights retain the traces of the smaller outpost stations, indicative, when thus found in combination, of considerable skill and warlike strategy. Such also may be presumed to be the origin of these small hill-forts, where we trace a line of them on a series of successive heights, as may be seen on the Lammermoors and in other Scottish districts, and is especially noteworthy along the southern slopes of the Kilsyth and Campsie hills, immediately to the north of the great Roman wall. These are obviously the outposts of the hardy Caledonian, from whence he watched his opportunity for some sudden foray or midnight surprise of the garrisons occupying the stations along the wall, and which he maintained with such persevering success that the Roman conquerors had at length to give way, and to fix the northern limits of empire on the older line of Hadrian, between the Solway and the Tyne.
The circular British forts or camps surmounting the heights of Galloway and the Lothians, and more or less common in nearly every district of Scotland, generally occupy an area of from three hundred to four hundred feet in diameter, and are inclosed with ramparts of earth and stone, or occasionally entirely of loose heaps of stone, which have lost through time every trace of any definite form of masonry they no doubt once possessed. But the subject has already been treated of with ample details in Chalmers' Caledonia;[455] and little that is worth recording can be added to his careful researches. Roy also includes the most important of these native strongholds in his "Military Antiquities," superadding to his descriptions, plans and sections, by which a very perfect idea can be formed of their original design. These include Wood Castle, a very remarkable circular fort near Lochmaben, in Annandale,[456] which General Roy describes as a Roman post, though it differs in every possible feature from any known example of Roman castrametation. That it is a British stronghold is not now likely to be called in question. It bears, indeed, a singularly close affinity to the circular earth-works which so frequently accompany the Scottish monolithic circles. Others of the supposed Roman forts bear scarcely less conclusive marks of native workmanship, as the intrenched post on Inch Stuthill, near the Tay, (Plate XVIII.;) Liddell-Moat, near the junction of the Liddel with the Esk, (Plate XXIII.;) Castle Over, situated on a high point of land, formed by the junction of the Black and White Esks, (Plate XXVI.) supposed by Roy to be the Roman Uxellum; and Burgh-Head, on the Murray Frith, (Plate XXXIII.) which he unhesitatingly assigns as the Ultima Pteroton of Richard of Cirencester, and the Alata Castra of Ptolemy. All of these bear a curious general resemblance to some of the aboriginal forts of the Mississippi Valley; thus affording, under another aspect, evidence of the mind of man operating in the same way when placed in similar circumstances, and with a force not perhaps greatly differing from the unerring instincts of the lower animals. The last example, that of Burgh-Head, possibly includes some remains of Roman works. The straight wall and rounded angles, so characteristic of the legionary earth-works, are still discernible, and were probably much more obvious when General Roy explored the fort; but its character is that of a British fort, and its site, on a promontory nearly inclosed by the sea, is opposed to the practice of the Romans in the choice of an encampment. The remarkable general correspondence of the Scottish "Deil's Dike," described in the last chapter, to the Scoto and Anglo-Roman walls, proves that the native Britons were not slow to avail themselves of the superior engineering skill of the invaders, displayed in military works of more importance than the mere rectangular vallum.[457] The fortifications here specified are not, however, to be classed with the simple circular hill-forts first noted, wherein we trace the mere rudimentary efforts of a people in the infancy of the arts. They display equal skill in the choice of site, and in the elaborate adaptation of their earth-works to the natural features of the ground. Though undoubtedly of native workmanship many of these are not improbably contemporary erections thrown up by the native Caledonian to withstand the encroachments of the Roman invader.
But the most remarkable British fort to the north of the Tweed, if not indeed in the whole island, is that which crowns the summit of Caterthun, looking across the valley of Strathmore. Two neighbouring heights are occupied with British forts. The larger of these, called the White Caterthun, from the colour of its walls, is an elaborate, skilfully constructed stronghold, which must have formed a place of great strength when held by a hardy and well-armed native garrison. It is of an oval form, inclosing an inner area of four hundred and thirty-six feet in length, by two hundred feet in breadth. But this only constitutes what may be regarded as the citadel. Beyond it a succession of ramparts and ditches surround the height at lower elevations, including a much larger area, and affording scope for a more numerous body of defenders. The hollow is still visible, though now nearly filled up, which was once the well of the fort, and probably this strength was maintained as a rendezvous and place of temporary retreat for the entire population of the surrounding district. The White Caterthun has been repeatedly engraved, and its construction and details will be best understood by a reference to the plans and sections in Roy's Military Antiquities.[458] The Brown Caterthun, which crowns another hill about a mile to the north, is also a specimen of ingenious native fortification. Its ramparts are nearly circular, and a series of concentric intrenchments extend down the slopes of the height.[459] Both of these native military works have been constructed with immense labour. The astonishing dimensions of the rampart of the former, composed of an accumulation of large loose stones, upwards of a hundred feet thick at the base, and fully twenty-five feet thick at the top, with its extensive lower earth-works and ditches, excite surprise and wonder in the mind of every observer. General Roy remarks after a careful survey of it,—"The vast labour it must have cost to amass so incredible a quantity of stones, and carry them to such a height, surpasses all description."
Another remarkable hill-fort of the same class is at the Barmekyn of Echt, in Aberdeenshire; and at Dundalaiv, on an unusually steep and rugged height in Glenshiora, Badenoch, is one smaller, but perhaps more striking, from the superior masonry of its walls. These are from twelve to fourteen feet in thickness, and being built of thin flat schistose slate, the walls remain in parts fully fourteen feet high, and apparently as perfect as when first erected. The inclosed area of this ancient fortress also contains a well, and considerable ingenuity has been shewn in strengthening the weaker points of the position. Altogether, it is the most perfect relic of a British stronghold of the class that I know of in Scotland.
The so-called "Vitrified Forts" of Scotland which have been the subject of so many ingenious and baseless theories, form another interesting class of native works. Attention was first drawn to them by Mr. John Williams, in his "Account of some remarkable Ancient Ruins, lately discovered in the Highlands and northern parts of Scotland," published in 1777. Mr. Williams had been employed by the trustees of the Scottish estates forfeited in the last rebellion, to superintend some operations in his capacity of a civil engineer, and in the course of this he for the first time investigated the singular remains to which he gave the name of Vitrified Forts. So entirely new was the discovery that it was generally received at first as an extravagant fiction, and no London publisher could be persuaded to undertake the publication of Mr. Williams's Account. His facts, however, proved indisputable, and theorists thereupon undertook to combat his conclusions, and to assign to the supposed forts a volcanic origin. The appearance of some of the most remarkable of these works is well calculated to sustain such a theory. The fortified area on the Top-o-Noth, near the village of Rhynie, Aberdeenshire, for example,—one of the most remarkable specimens of a vitrified fort in Scotland,—could not be more accurately described than by comparing it to the crater of an extinct volcano.
Since the first announcement of Mr. Williams's remarkable discovery there has been no lack of observation or controversy on the subject, though not always with very satisfactory results. In 1825 the Society of Antiquaries of Scotland directed special attention to it, and the results of a series of careful investigations are published in the fourth volume of the Archæologia Scotica, made chiefly under the direction of the late Dr. Samuel Hibbert, one of the secretaries of the Society, and further qualified for the duty as an experienced practical geologist. The fruit of these investigations may be thus stated:—Dr. Hibbert arrives at the conclusion that the vitrification is an incidental and not a designed effect,—having formed no part of the process of erection of the forts or cairns on which it is now traceable, but resulting accidentally from the frequent kindling of beacon-fires as the signals of war or invasion, as well as from bonfires which formed a part of festive or religious rejoicings; and indeed from numerous independent causes, probably no less widely dissimilar in dates than in origination. The nature of the sites, also, where vitrification has been detected proves that it is by no means confined to fortified positions; nor when it does occur on such is it generally found diffused throughout the ramparts of stone, or even restricted to their limits. Dr. Hibbert accordingly rejects the name of vitrified fort for the more comprehensive and untheoretical one of vitrified site, as most descriptive of remains which appear to include small inclosures for the protection of beacon-fires; sites of bonfires periodically lighted at the ancient places of rendezvous for tribes or clans; and hearths of fort-beacons and signal-fires, occasionally occupying not the ramparts but the ditch.
The only argument which tends to throw any doubt on the result of Dr. Hibbert's conclusions is that of Dr. Macculloch,—a shrewd observer, little inclined generally to extend toleration to any antiquarian hobbies but his own,—who affirms, that in situations where the most accessible materials for constructing a stone fort are such as are incapable of being vitrified, suitable materials have been selected and brought with considerable labour from a distance.[460] But the evidence of design in the choice of such materials is by no means apparent. The examples referred to by Dr. Macculloch only confirm the fact, already familiar to the chemist and geologist, that there are very few districts in Scotland where rocks do not occur capable of being more or less vitrified. This subject is fully illustrated by an interesting series of experiments carried on by Sir James Hall, towards the close of last century, with a view to test some of the geological theories in reference to the igneous formation of rocks, which then furnished a fertile theme of controversy between the disciples of Werner and Hutton.[461] All the varieties of trap are so peculiarly susceptible of fusion, that they have been recently selected as the most efficient and economical flux in the smelting of copper ores. I am indebted to Dr. Francis Hay Thomson, the inventor and patentee of the ingenious application of the common rocks to this novel purpose, for communicating to me the results of his experiments. His invention chiefly consists in "the application of what are commonly called whinstones, and of other stones similar to whinstone—such as trap, basalt, sienite, and the like, being fusible silicates, as a flux in the smelting of copper ores." He has found all these materials capable of easy and complete fusion in a reverberatory furnace; but a much more moderate degree of heat would suffice to produce the conglomerated masses usually found on vitrified sites, where the larger stones are merely inclosed and cemented together by the fused matter. In reply to inquiries as to the probable effect of bale-fires kept blazing repeatedly on the same ramparts or heaps of stones, where a gradual accumulation of ashes from the burning pile must fill up the intervening spaces, and supply a flux capable of combining for the ultimate fusion of the whole, Dr. Thomson remarks:—
"Granite is per se very infusible; that of Aberdeen almost entirely so, in consequence of the presence of an overplus of silica. Sandstone is per se quite infusible, being almost entirely silica. Your supposition may, however, be correct, for the addition of the alkali produced from the wood-ashes would much assist the fusion of all kinds of stone that might be used in building these forts. Whinstone contains at least four per cent. of pure soda, fifteen of iron, and from twelve to twenty of lime. All these form a most fusible mixture, and the silica present is only in such proportion as is necessary for vitrification. Limestone is of itself not fusible except at a very high temperature; but the addition of either iron or soda with silica renders it at once vitreous. Although I am not certain as to the exact degrees of heat requisite for the fusion of these materials, I may mention that, in an ordinary reverberatory furnace, I have fused five cwt. of whinstone in one hour and a half, the product being a dark mass similar to bottle-glass; and I have no doubt, were proper precautions taken, that large slabs might easily be moulded for building purposes."
The degree of heat attainable in a reverberatory furnace manifestly greatly exceeds any temperature that could be produced by an exposed fire of wood; but the usual appearance of the vitrified masses found on the sites of forts or beacon-hills is such as proves them to be the product of a more moderate heat. The larger pieces are not fused into a homogeneous mass, but blocks of trap, granite, and sandstone, or occasionally all three in juxtaposition, are enveloped in a vitrified coating of irregular thickness, and bound into a solid piece by this extraneous substance. The alkali supplied by wood-ashes is abundantly sufficient to produce such a result. Carbonate of potash in contact with trap will readily melt at a red heat, and has a power of uniting with the constituents of the trap to form a fusible compound which hardens into glass in cooling. Fire-clay, which is altogether infusible, and less liable to be affected by heat than most of the known natural rocks, is employed on this account in making the chemists' crucibles; but if an alkali is melted in a fire-clay crucible, it forms a vitreous covering on the surface, and where large quantities are used even goes through the crucible. This is a fact familiar to the chemist, and so impossible is it to keep fused alkalis in contact with silicates, that only crucibles of platina or silver can be used for the analysis of silicious minerals. In this way even sandstone, though per se infusible, is perfectly capable of vitrification, and indeed is, under certain circumstances, peculiarly susceptible of it, as its great porosity admits of the ready absorption of the melted alkali.
This susceptibility of the degree of fusion usually observable on vitrified sites, which trap and others of the common rocks of Scotland possess, has long been recognised by able chemists; and when it is taken into consideration along with the very diversified and dissimilar circumstances under which vitrification has been observed, the conclusion seems inevitable, that it is an incidental and not a designed result of the application of fire. But neither the interest nor the importance of this inquiry is exhausted when we have established the undesigned origin of vitrified sites. The question still remains,—Are they peculiar to Scotland? because, even if we reject the idea that the cementing of stone buildings by means of fire is among the artes deperditæ Scotiæ, still the discovery of so many vitrified sites in nearly every district of Scotland, would seem to indicate the practice of peculiar customs and observances during those early centuries in which the primeval forests furnished an unlimited supply of fuel. It is at all times a precarious and unsatisfactory basis of argument which depends chiefly on the absence of contrary evidence. Nevertheless it is worthy of some note, that although upwards of seventy years have elapsed since Mr. Williams published his account of vitrified forts, no single example, so far as I am aware, has been discovered south of the Tweed.[462] This cannot be ascribed to the subject being one of mere local or temporary interest. It has excited much controversy, not only among English antiquaries, but among the students of various kindred sciences; and while the geological features of some districts preclude the possible existence of such structures, it will, I think, involve very important conclusions as to the peculiar customs of the early Caledonians, if it be recognised as an established fact, that neither in the Welsh Highlands nor in the stone districts of England, are any traces of vitrified forts or sites visible. It has been the fashion of late years to give the whole question the go-by in very summary terms as one that has already commanded undue notice. Such, however, is a more convenient than satisfactory mode of dealing with the inquiry. Dr. Hibbert has appended to his "Observations on Vitrified Forts," a list of forty-four sites already noted, extending over twelve Scottish counties, including the most northern and the most southern districts of the kingdom. To these others have since been added, both north and south, in the Orkney Islands, and in the vicinity of Jedburgh, near the English border. It will suffice, meanwhile, to note these facts, in the hope that English Archæologists may, on fitting occasion, seek a reply to the inquiries which they involve:—Were the southern Britons, whether Celtic or Saxon, or the intruding Scandinavians or Gauls, wont to kindle bayle or beacon-fires on cairns, forts, or elevated sites, with such frequency as to leave similar traces to those which are so common in Scotland? Or must we infer that these abundant remains are the result of ancient rites and customs peculiar to the races of the northern kingdom?
Dr. Hibbert has already invited the investigations of Scandinavian Archæologists, with a like view, anticipating, from the notices of Olaus Magnus, Snorro, and others, that vitrified sites should be found on the mountain tops in Norwegian provinces. Nineteen years have elapsed since Scottish Antiquaries appealed to those of Scandinavia for a reply. They have not been unmindful of the interests of archæological science in the interval, but still we wait in the uncertain negative which their silence furnishes; casting, meanwhile, some curious thoughts backward on the old Scottish festival of Beltane, and its apparent affinity to the rites of the Assyrian Baal.[463]
To attempt to assign a date for these primitive forts or vitrified sites would be manifest folly, but even to apportion them to one or more of those less definite periods is difficult. Some of them doubtless pertain to the earliest era of combined action, of which they would form one of the first results, while others may belong to a comparatively recent period; and, in particular, such border sites as those of Cowdenknowes and Howden Moor[464] perchance date no farther back than to those eventful times of watch and ward on the Scottish borders, quaintly referred to in the Act of James the Second's Parliament, in 1455, "for bailes making" to warn of the approach of the Southron foe: "Ane baile is warning of their cumming, quhat power that ever they bie of; twa bailes togidder at anis, they are cumming indeed; four bailes, ilk ane beside uther, and all at anis as foure candelles, suithfast knawledge that they ar of great power and meanis far."
The duns and vitrified forts of Scotland have long been the subject of observation and controversy; but there is another class of defensive earth-works observable in various Scottish districts, which, so far as I am aware, have not yet attracted the notice of the antiquary, though sufficiently familiar to rustic observers. These consist of artificial trenches, generally dug in the side of a hill, and obviously designed for the hasty concealment of cattle from predatory bands of marauders, though in some cases tradition associates them with more remarkable events. One, for example, of considerable extent, situated between Kintore and Inverury, in Aberdeenshire, is popularly known as Bruce's Howe, from an old tradition that it had afforded the means of concealment to a party of Robert the Bruce's army before the battle of Inverury. Its depth, like that of most others, is about eight feet, affording effective shelter and concealment both to men and horses. Another of these artificial trenches has been cut out of the side of a hill, near its summit, on the farm of Altyre, parish of Dalry, Kirkcudbrightshire. It is capable of containing about an hundred men, while a person concealed in it can see to a considerable distance, in the two principal directions of approach, without being observed. From the convenient retreat it afforded to the persecuted Covenanters in the time of Charles II., it still bears the name of the Whig Hole. A larger trench of the same kind exists along the side of a steep hill forming one of the range of Scuir-na-fion in Glencoe. This has been constructed with considerable skill, the trench running parallel with the range of hills, and opening at its west end in a gully formed by a small mountain stream, which joins the river Coe somewhat farther down. From a distance, or from any lower part of the Glen, the trench is quite indistinguishable, as the embankment, which in this case has been formed on the side of the hill, is sloped so as completely to coincide with the angle at which the latter rises from the valley. An intelligent correspondent, familiar with this part of the Highlands, informs me that he had frequently visited the Glen without being aware of the existence of the trench, though passing it at no great distance, and his attention was first called to it by observing the fresh colour of the herbage on the upper edge of the embankment, in contrast with the more olive hue of the hill-side beyond,—a phenomenon easily accounted for by the fall of the heavier and coarser debris of the embankment towards its base, thus leaving a finer soil along the ridge. Angus M'Donald, an old and intelligent native of the Glen, at once assigned its origin to troublous times, for the purpose of sheltering the natives and cattle of the Glen when surprised by an invading foe, and stated that it includes ample space for concealing three hundred head of cattle.
Various examples corresponding to those occur in different parts of the Highlands, belonging to no definite period, but indicating the dangers and the resources of a pastoral people, liable to sudden and frequent invasion by powerful warlike foes. A similar state of society, though at a period more advanced in civilisation and the practice of the constructive arts, appears to be indicated in that remarkable class of structures peculiar to Scotland, and generally known as Burghs or Pictish-towers. These, like so many other of our native antiquities, it has been customary to ascribe to a Danish origin; but the increasing interest now manifested by native antiquaries in our northern antiquities, and the frequent communications which have taken place of late years between Scandinavian and British Archæologists, have sufficed to establish the important fact that no such structures are known in the old lands of the Northmen.[465]
The Scottish Burghs are large circular fortresses, or bell-shaped structures, built of unhewn stone, and entirely without cement. The most perfect example of these remarkable edifices is situated upon the island of Mousa, near to the mainland of Zetland; but many remains of them can still be traced, both on the northern and western isles, in Caithness and Sutherland, and on various parts of the north and west coasts of Scotland. They are nearly all formed precisely on the same plan, though differing considerably in size. The form is a truncated cone, occasionally slightly varied, as in that of Mousa, where the wall curves inward till it attains a certain height, and then returns gradually outward again, apparently with the same design as the corbelled battlements of a later date, which enabled the defenders more effectually to annoy any assailant who ventured to approach the base. With this exception the exterior displays no ornamental projections, or any provision for defensive operations, by means of window, loop-hole, or machicolation. The rude but very substantial masonry of the exterior is only broken by a plain narrow doorway, which, from the absence of gate-posts, grooves, or any of the ordinary refinements of more modern architecture, it is not improbable was secured, when danger was imminent, by building it up with a pile of stones. Within the exterior cone a second cylindrical structure is reared, the walls of which are either perpendicular, or constructed at an angle which, leaving a space between the two of about six feet at the base, brings them together at the top. Within this space between the walls a rude staircase, or rather inclined passage, communicates round the whole, and a series of chambers or tiers of interspaces, formed by means of long stones laid across from wall to wall, so as to form flooring and ceiling, are lighted by square apertures looking into the interior area. This central space is open to the sky, and the fact of the only light to the chambers and passages within being derived by means of apertures opening into it, seems to preclude the idea of its ever having been roofed. It is not apparent, however, by what means the occupants could obtain access to the ramparts, so as to resist an assault, and prevent the walls from being scaled, though a sufficiently rude and simple wooden structure may have supplied this very obvious defect.
Cordiner and Pennant have each given a very full account of Dun Dornadil, a Burgh or Pictish tower in Glenelg, and one of the largest of this singular class of military structures.[466] Gordon furnishes descriptions and engravings of Castle Tellve and Castle Troddan, two other examples which he examined;[467] and Dr. Macculloch also supplies a minute account and measurements of one of those, in Glenelg.[468] "The masonry," he remarks, "is without lime, but remarkably well laid, and the lines of the curvature are beautifully preserved throughout. The floors of the galleries consist of single flags, and the window apertures are, in a similar manner, divided by transoms of stone."
One necessary consequence of the plan on which all these buildings are constructed is, that while the lower galleries are roomy, and admit of free passage, the space narrows so rapidly that the upper ones are too straitened even to admit a child. This is particularly observable in the Burgh of Mousa, which, though more perfect, is considerably smaller than that of Dun Dornadil, and consequently a much greater proportion of the internal galleries must have been totally unavailable, either for occupation or the storing of property. No great difficulty, however, need be made about this, even where windows are found made in the inner wall, equally for the wide and the most straitened tiers of galleries. One model, and that a very simple one, supplied the design for all; and it would not be difficult to find corresponding examples in modern masonry where the same unreasoning fidelity to the original is shewn, as in the latest structures of the Tudor style, where unperforated gargoyles project from solid walls, and flying buttresses are thrown where there is nothing to support.
The most remarkable deviation from the common arrangement of these singular structures is where, as in the Burgh of Achir-na-Kyle, Sutherland, regular conical chambers are constructed in the solid wall. This is a manifest refinement upon the original design, and may be regarded as the first progressive step in the art of military architecture. Cordiner remarks of this example:—
"It is situated with peculiar taste on the top of a lofty rock, opposite to some pleasant woods, and near excellent pasture; and round the precipice which overhangs the Brora, the river tumbles over its rocky channel in a number of irregular cascades. This building would have doubtless merited a very particular description, had it not corresponded with your account of those in Glenelg.[469] I must except the apartments within the walls, which are of an oval form, distinct and entire, about eight feet long, six high, and four wide. Those on the ground-floor are still a retreat from the storm for the goats that feed on the neighbouring hills. The stairs from the first to the second row of chambers are regular and commodiously made out. The apartments are carefully lighted by windows from within, a strong evidence that the area within these towers had never been closed above, or entirely covered. The door looks over the precipice towards the river, and is full six feet high.... One chamber had several plans of a level entry to it, and measured nine feet in height; this had been probably intended for the chieftain. The whole structure seems to me so well contrived that it is not easy to conceive a people who could not work in wood or iron could have been more conveniently accommodated in places of defence."[470]
Considerable skill and ingenuity is frequently shewn, both in the choice of a site for these defences and in turning it to the best account. They most frequently occupy capes, headlands, or small islands, either in a lake or on the open sea. Sir Walter Scott describes a curious device which he observed employed for guarding one of those in Shetland against the approach of strangers. "I remember," he remarks, "the remains of one upon an island in a small lake near Lerwick, which at high tide communicates with the sea, the access to which is very ingenious by means of a causeway or dyke, about three or four inches under the surface of the water. This causeway makes a sharp angle in its approach to the Burgh. The inhabitants, doubtless, were well acquainted with this, but strangers, who might approach in a hostile manner, and were ignorant of the curve of the causeway, would probably plunge into the lake, which is six or seven feet deep at the least. This must have been the device of some Vauban or Cohorn of those early times."
These remarkable buildings can hardly be viewed with too great an interest by the Scottish archæologist. They are the earliest native architectural remains which we possess, the cromlechs and stone circles being at best only rudimentary and symbolic or representative forms of architecture. They constitute, therefore, a most important element in our national history, supplying very definite facts relating to an ancient era of which we have received no other information in any degree so trustworthy. The first point accordingly is to ascertain, with such accuracy and minuteness as may now be possible, the precise nature of these facts. Careful investigations have accordingly been carried on of late years, accompanied in several instances with excavations around the buildings and within the inclosed area, the results of which are worthy of note. In more than one instance human remains have been found on removing the accumulated rubbish and debris from these ancient ruins, suggesting the possibility of their correspondence to the Nuraghes of Sardinia, which they somewhat resemble in outward form. It is altogether inconceivable, however, to ascribe a sepulchral origin to these chambered towers; while the same excavations which have discovered the remains of the dead have also in most cases furnished no less conclusive evidence of the former presence of the living. But, it has been already observed, the archæologist finds both "knowledge and understanding" in the grave, and esteems it a conceivable source of valuable insight to have even these dead to question on the subject. Dr. Macculloch mentions the discovery of human bones in the Burgh of Glenelg, but without entering into details; but the results of a careful examination of another of these towers, near Dunrobin, in the summer of 1849, elicited more definite information. On removing the rubbish from the chambers and galleries, a human skeleton was found in one of them, while excavations within the open area disclosed abundant traces of a fire in the centre, and also discovered several stone quernes or hand-mills. The skeleton here appeared obviously to belong to a later period than the quernes and the central fire; but no accompanying relics of the deceased were found to tell how long the fire of the old garrison had been extinguished ere the chamber of their fort was made a receptacle for the dead. More satisfactory results attended the examination of the Burgh of Burghar, another of these singular towers in the parish of Evie, Orkney. It is described by Mr. A. Peterkin, in a letter addressed to Dr. Hibbert in 1825, as the most perfect though not the largest of several in the neighbourhood. Several barrows occur in the vicinity, some of which have been opened and found to contain urns. The central area of the Burgh of Burghar was nearly filled up with the accumulated ruins and rubbish of centuries, and resisted more than one effort to explore it; but the son of the clergyman of the parish renewed the attempt in the spring of 1825, and succeeded in partially investigating the contents of the ruined area. On digging out the earth and rubbish, he found a human skeleton, beside which lay part of a deer's horn, and the rude bone comb represented here, about one-third the size of the original, which is now deposited in the Museum of the Scottish Antiquaries. Mr. Peterkin appears also to have forwarded the skull to Dr. Hibbert, though it has not been preserved. His description of it has already been noticed in a former chapter.[471]
More extensive excavations were made within the ruined burgh at a subsequent period, and led to the discovery of some very valuable relics, including two fine gold armillæ, now in the possession of the Earl of Zetland. In this example also there can be little hesitation in assuming that the deposition of the dead body did not take place till the abandonment of the burgh, perhaps not till it had been long in ruins, as it does not appear from the description to have been below the level of the original floor, but within the accumulated soil which encumbered the area. This, however, is open to doubt, as the letter is not quite explicit; and if the interment was at some depth below the floor, it might have taken place while the burgh was occupied, and an assailing force precluded access to the neighbouring downs on which the aboriginal sepulchral tumuli are still visible. It may be doubted whether the gold relics were placed there as a sepulchral deposit, or only for security or concealment. They belong possibly to a period long subsequent to that of the first interment with its simple and rude accompaniment of the bone comb. The latter discovery, indeed, seems to furnish some approximation to the period of those buildings. It shews, what we might expect, that they are the work of a people whose arts were extremely rude, and so far as any general reasoning may be built on a solitary instance, it seems to point to the erection of the burghs at a period long prior to the earliest recorded traces of Scandinavian invasion. The discovery, however, is not altogether without precedent. Another bone comb in the Museum of the Scottish Antiquaries, of even ruder construction, though nearly resembling the one found at Burghar in its general form, was found, in 1782, in the ruins of another burgh in Caithness, and a third discovered under similar circumstances is in the museum at Kirkwall. There appears, indeed, to have been a predilection for burying among the debris of the ruined forts, which must be supposed to have been formed long subsequent to their abandonment as strongholds and places of abode, and therefore adds to the evidence of their great antiquity. Some of these interments are undoubtedly traceable to the period of the Scandinavian occupation of the Orkney Isles.
How of Hoxay.
Mr. W. H. Fotheringham of Kirkwall, Orkney, has communicated to me an account of the recent exploration of another ruin of a circular fort, apparently belonging to the same class of buildings as those previously described. It occupies as usual an isolated promontory, called the How of Hoxay, in South Ronaldshay, immediately opposite to the Bay of Scupa. Rising abruptly from the small Bay of Hoxay is the How or Height of Hoxay, on the top of which are the remains of a circular building. Until brought to light in the course of recent excavations it was entirely buried beneath the accumulated soil, and presented only the appearance of an earthen tumulus. It has now been completely exposed externally, and the inclosed area excavated to the surface of the rock, so that the work of exploration has been most effectually performed. The external wall measures fourteen feet in thickness, and about eight feet in greatest height, and incloses an area of about thirty feet in diameter. The construction of the wall is singular, the exterior and interior facings appearing to have been carefully built with unhewn stones fitted together with great nicety, and the intervening space filled up with stones thrown in with little care or design. No cement had been used, but the wall is still strong and without any displacement in the facings, though so much ruined that no certain idea can now be formed of its original height. The great quantity of stones which lay both within and about it served, however, to shew that only a small portion of the original fabric remains. The accompanying view of the most perfect side of the interior will convey a better idea of the general appearance and details than any description could do. The two upright stones about half way up the wall on the left of the drawing appear to be the side-posts either of a door or outlook, to which the projecting step below was probably designed to give access; but it was found built up like the other parts of the walls, and the proprietor having since, in a misdirected zeal for the preservation of the ruin, had the whole pointed with lime, it is no longer possible to detect the additions of later builders. Round the inner circumference of the wall upright flag-stones project at intervals of six feet apart. Only six of these now remain, but the fragments of others were discovered among the debris. In the recesses formed by these projecting stones there were found several stone quernes, a shallow stone mortar and pestle, or corn-crusher, of the rudest and most primitive construction, and also two smaller circular stone vessels, the one seven and the other five inches in diameter, and both about four and a half inches deep. The remains of the doorway in the eastern and most ruined part of the wall appear to have been of an unusually intricate construction, but these also have unfortunately been obliterated by later repairs, the whole wall having been raised to a uniform height, and a platform and flagstaff superadded in very questionable taste. The proprietor was actuated in his labours by a sincere desire for the preservation of this venerable ruin, and antiquaries must respect his motives, though he has not effected it exactly in the way they would have wished. I am favoured by Mr. Fotheringham with the following description and sketch:—"As to the door on the east side, the information I have got is that it was contracted by means of slates thus; and that at the side of the door was a chamber in the thickness of the wall leading from the interior, from which there was an aperture or slit to the widest part of the doorway, either for the purpose of outlook, or for projecting a weapon against a hostile intruder." This arrangement in the construction of the doorway more nearly approaches the plans for outlook and defence with which we are familiar in medieval military architecture. It is greatly to be regretted that no opportunity was afforded for more accurate observation.
The result of these investigations is highly satisfactory and encouraging, giving promise of further information from the labours of future explorers. Meanwhile some important conclusions may be arrived at. It is not necessary that we should follow Cordiner in his learned arguments concerning King Dornadil, a successor of Fergus I., who ascended the throne A.D. 263, and signalized his reign by erecting the Burgh of Dun Dornadil on the north-west coast of Inverness-shire. With precise dates the archæologist can rarely, if ever, have aught to do while treating of primitive antiquities; but this at least seems established, that they are native erections, and belong for the most part to a period long prior to the era of Scandinavian invasion. Where the Teutonic and Scandinavian races ultimately prevailed they bear the name of Burghs; where the older Celtic race and language survive they retain the name of Duns: and Sir Walter Scott has pointed out, in an ingenious note appended to Ivanhoe, that the venerable Saxon stronghold of Conigsburgh is only a refinement on the older model of the Scottish burghs. This has been illustrated by drawings and sections in the Abbotsford edition of the novels, and the resemblance is certainly sufficient to carry much probability with it, though at the same time the complicated arrangements, and the provisions for aggressive operations against assailants in the burgh of the southern Saxon, cannot but add to the conviction that the Scottish strongholds of this class belong to a much earlier period. They are manifestly the work of an ingenious and patient race, who aimed far more at defence than aggression. Strongholds they undoubtedly are, but they retain no trace of features strictly adapting them to military posts. The Saxon burghs of England were rapidly superseded by the more efficient keep of their Norman conquerors; yet when we institute a comparison between Conigsburgh and Mousa or Dun Dornadil, it seems to present a contrast not unlike that which distinguishes the defensive operations of the wild-cat and the hedgehog!—a contrast which either marks a very great change on the character of the hardy tribes that withstood the Roman legions, or indicates a marked difference between the races which occupied the northern and southern regions of Caledonia.
Dr. Macculloch remarks of these Scottish burghs,—"From the expensive nature of their construction, or the power of hands that must have been employed on them, it might be supposed that they were the palaces or castles of the chiefs or kings of the days in which they were erected. But it seems an insuperable objection to this notion, that four should have existed within so small a distance from each other in Glenelg, or that so many should be found in Sutherland and in Shetland not far asunder. The limits of territory that surround any one are too narrow for any chief; and where all chiefs were in a state of general and constant hostility, it is not likely that they should have chosen to build so near to each other. It is equally impossible that they should have been the dwellings of the inhabitants in general, as the expense of erection bears no proportion to the limited accommodation they could afford." The expense of erection is, in other words, the labour, time being of small value in a primitive state of society; and when their number is taken into consideration along with their limited accommodation, it is difficult to evade the conclusion that they were the temporary places of shelter of a people liable to sudden inroads from powerful foes, like the palisaded log-house or fort which the first settlers in the backwood frontiers of America were wont to erect as a place of safe retreat on any attack of the treacherous aborigines. There is no period that we know of in early Scottish history to which this description so aptly applies as to that immediately preceding the conquest of the Orkneys by Harold Harfager, about the year 880. Prior to this the rude Norse Vikings were wont to make sudden descents on these islands, as well as along the whole Scottish coast, spoiling and slaying with the most remorseless cruelty. At such a period, therefore, we can most readily conceive the natives of a district combining to build a burgh, whither they could retreat so soon as the fleet of the northmen was espied in the offing, and driving thither their cattle, and carrying with them all their most valuable moveables, they could lie secure till the spoilers set sail again in quest of some less watchful prey. Experience would teach the necessary improvements requisite for rendering these structures effectual against such foes; while the improbability of the northmen abandoning their ships and attempting a regular siege of one of these burghs, may account for the absence of the very distinct provisions for offensive operations against assailants which are so characteristic of the Anglo-Saxon burgh.
The Burgh of Mousa, which is still the most perfect of these ancient strongholds, is the only one of which we have any distinct historical notice. Torfæus tells us that Erland, the son of Harold the Fairspoken, carried off the mother of Harold, a Norwegian jarl, who was famed for her beauty, and took shelter with his prize in the Castle of Mousa. Earl Harold followed and laid siege to the place, endeavouring first to take it by assault, and afterwards to reduce it by famine. But both means proved equally ineffectual, and the wrathful Jarl was forced at length to agree to terms by which his mother became the wife of her ravisher. This burgh is not only the most perfect, but also the best adapted for defence of any that now exist; and it is not improbable that it owes its projecting parapet, as well as the more effective repair which has secured its preservation, to its later Norwegian occupants.
Still it does not necessarily follow from the correspondence of the state of society in the north of Scotland in the ninth century, as a weak people, constantly liable to sudden inroads by powerful and merciless invaders, with the apparent indications of these strongholds, that we must therefore assume the origin of all of them to that period. The conquest of the Orkneys, and the occupation of the northern districts of Scotland by the northmen in the ninth century, marks the close of a period which is still involved in almost total darkness. How long before this the natives had learned to watch the horizon for the dreaded fleets of the northmen, or in what form the earliest migration of the Cruithne to the north took place, we have yet to learn; but the very fact of the frequent descents of the former on our coasts must be viewed as affording some evidence that the arts of civilisation had advanced far beyond the rude state indicated by such primitive relics as those which were discovered in the How of Hoxay. The "exactors of rings" could have found little to tempt a second visit to the barbarian Orcades of the Stone Period; for the wandering Vikings knew not the pride of conquest which could tempt a Cæsar to guide his legions to the Ultima Thule, that he might return to the proud honours of a Roman triumph. The disappointed Viking might indeed be not inaptly supposed to apostrophize the outlooking Orcadian, as the English poet Cowper has done the "Gentle Savage" of Tahiti:—
"Expect it not. We found no bait
To tempt us in thy country.
We travel far, 'tis true, but not for nought;
And must be bribed to compass earth again
By other hopes and richer fruits than yours."
Without attempting to deduce from such evidence as is now attainable, more than it seems fairly to warrant, it is obvious that we have followed down the unwritten history of our island from that remote and imperfectly defined era in which we catch the first glimpses of its occupation by wanderers from the eastern home of our common race, to the period when definite history begins, and written records supply to some extent the information heretofore painfully sought amid the relics of older times. There still remains, however, some few pages more of these archæological annals to be deciphered before we attempt to sift the perplexing mixture of truth and fable which makes up our earlier written history.