A Highland Artform
Presidential Address to the Inverness Field Club 17th March 1999 James D Scarlett, M.B.E., F.S.A.Scot.
Originally published by Inverness Field Club
©James D Scarlett 1999
From the beginning, it has been the Field Club's custom to reward its retiring presidents and give them a chance to get a little of their own back by allowing them to ride their hobby horses in front of a captive audience for half an hour or so. My mount is something of a bucking broncho and widely regarded as nothing more than a circus act but, even though one sometimes has to ride it standing on one's head or facing the wrong way, I hope to show you that it is a Highland thoroughbred with a pedigree of enviable length.
The paper that I am going to read to you tonight explores one branch of that pedigree and reaches a conclusion somewhat at variance with common belief. I have called it A High Land Art-form and must preface it by telling you that, while a sufficiency of old cloth-fragments has survived to enable tentative conclusions to be reached, only the very unwary or unwise regard anything about tartan as being proven.
The popular idea that the Highland clansmen wore tartans chosen by their Chiefs so that they could be recognised in a fight takes a bit of a knock when it is realised that, in several cases, Clans by no means friendly to each other wear tartans so similar as to be not easily distinguished in the cold light of day, let alone in the heat of battle. It does not stand up any better if the various tartans are collated with a map of Clan territories, when it can be seen that patterns tend to travel along the regular lines of communications, river valleys, hill passes and so on, from weaver to weaver, picking up local variations on the way; the implication that it was the weaver, and no-one else, who designed the patterns is a strong one.
It has been suggested that the pictures with which the Picti were supposedly adorned were actually tribal tattoo patterns and it would be in character for such people to demand a formal pattern in their clothing. Whatever the truth of that may be, when the weaver had produced cloth that would serve his customers' requirements in the rigours of the Highland climate, his next task was to devise patterns that would take their fancy. For this, he could dye wool with remarkable permanence and consistency in a limited range of colours and employ it in any design he might fancy so long as it was in the rectangular format enforced by the weaving process. Not by any means all of the weavers were brilliant artists but all possessed and used the ability to make bright and attractive patterns in an almost infinite variety, patterns that have outlasted and outlived fashion to become 'Clan' tartans, the 'right' to wear which is avidly sought, perhaps because of some remote folk-memory of those tribal tattoos that I have already mentioned. An attempt to analyse the art-form reveals it as highly complex, with strong mathematical overtones and dependent on many outside influences. Its attractiveness is a source of disappointment to many who cannot conjure up the mystic 'right' to wear it and has resulted in the generation of large numbers of counterfeit forms. Its durability, both as a style of decoration and as surviving fragments of cloth, is a tribute to the ability and the craftsmanship of the people made it. As a pure art-form it is, in these islands, unique to the Highlands of Scotland and as much a part of the Highland heritage as symbol stones and the Gaelic language.
Setting aside the inventions of the Sobieski brothers (and the few Lowland tartans that they did not invent and were mostly chosen from catalogues) and those tartans of the blue, black and green kind, which are looking increasingly as if they owe their origin to military fashion, we are left, by default, with red, green and blue as the main colours of the vast majority of patterns; old specimens show that red and green were of roughly equal 'grey value' and blue was dark, unlike modern tartans, which have blue and green almost black as in 'modern' colours, or washed-out and scarcely distinguishable from each other, as in 'old', or 'ancient' colours. My knowledge of artistic theory is less than slight, but I am a practical photographer with an inclination to experiment, and the first thing that struck me here was that these were the colours of the light filters used to make the colour-separation negatives that were the first stage in colour- printing before all the electronic miracles came about. Subsequently, I learned that they are also the first colours to appear in all primitive art, so it may be that they fulfil some deep physiological or psychological purpose, but that is a digression and explanations must be left to specialists in those fields, though it does seem that the way is open for somebody to write a large book on the Psychology of Tartan and, probably, for several others to write even larger books to explain the first.
When red light is superimposed on green, something yellowish results; add blue light and a muddy green appears. It is evident that a cloth patterned in these colours will merge with the landscape very well indeed when seen from such a distance that the pattern is no longer clearly defined. From the point of view of art, red and green provide colour contrast and red and dark blue, or green and dark blue, provide grey-scale contrast and the prime task of the artist- weaver was to arrange blocks of these colours in such quantities and proportions as would suit the purpose for which the cloth was intended and please the eye of the customer. Although there are many beautiful patterns made up from just these three colours, it is usual to break them up into 'eyeful-size' areas and to embellish them with overchecks of one of these colours or of others from the rather limited range available. This range normally extended from the aforementioned red, green and blue to black and white, light blue and yellow, but the limitations are more apparent than real, and this is where the arithmetic comes in.
A tartan pattern consists of stripes in both directions of the cloth and each stripe in the weft therefore crosses every stripe in the warp. When a stripe crosses another of its own colour a block of plain colour results and when a stripe crosses one of another colour the result is an equal mixture of the two; given that suitable shades are used, a distinctly midway colour will emerge but, all too often in modern tartans, one colour swamps the other and the effect is completely spoiled. The number of mixtures increases in rapid disproportion to the number of 'starter' colours, two starter colours giving one mixture, four starters giving four mixtures (eight shades in all), six starters giving fifteen mixtures (21 shades) and seven, the normal maximum, twenty- one mixtures (twenty-eight shades). Blocks of plain colour can meet only corner-to-corner, on the diagonal; laterally and longitudinally they are separated by blocks of mixtures which lead the eye from one solid colour to the next. It follows that, the more starter colours and the more stripes the softer and more diffuse will be the finished pattern, with consequent enhancement of it camouflage value; tartan is nothing if not paradoxical.
There are two ways of making woollen cloth weatherproof. One is to make it from a softish, woollen-spun yarn in plain weave and then shrink it and felt it, making it thick and heavy; this was the broadcloth upon which England's textile fortunes were founded and it is totally unsuitable for skipping about on mountains. The other way is to weave it from fine, hard, worsted-spun yarn closely woven in the twill weave that makes a lightweight cloth about 50% denser than plain weave. For obvious reasons, whether consciously or not, the Highlanders chose the latter and produced a gaberdine-like cloth so tough that two centuries of mastication by moth makes little impression.
In a plain-woven cloth, the crossings of colour mingle in pepper-and-salt fashion, but twill produces diagonal ribbing in the cloth which gives an effect like pencil shading, with fine lines of colour alternating; along the diagonal, the colour has a subtly different appearance from the cross-diagonal view, giving some life to the pattern, even as one walks past it in the shop window.
Now we come to what is to my mind one of the most interesting questions, "Why tartan, anyway?". Some of you, I am sure, will answer "Easy, it was brought from the East by the Celts who came through Brittany, Cornwall, Wales and the Isle of Man to Ireland and the Scots brought it to our mainland from there". A good story, especially if you want to convince the Bretons, Cornish, Welsh, Manx and Irish that they simply must buy your products and, incidentally, tap the huge market of emigrant Lowland Scots, but it does not explain why no residue of developed tartan-type patterns was left along the way, or why the ancient languages have no word which expresses what we mean by tartan. The Romans could say 'chequered' but usually went no further than 'striped' and, despite attempts to find some Gaelic derivation for the word, 'tartan' just means a kind of cloth, as is evidenced by a merchant in Edinburgh who, as recently as November 1828, ordered a quantity of 'Plain dark green coloured tartan, no pattern on it.'
Another matter requiring explanation is how the Irish 'Leine Chroich', a cassock-like garment of linen, pleated and, as one account has it, 'manifoldy sewed' and dyed yellow, metamorphosed into the large sheet of chequer- patterned woollen cloth that was the belted plaid. If the archaeologists have got it right, the regularly repeating check pattern and the twill weave were known in Scotland before the end of the third century, before the Scots came, and it is my guess that when these visitors from outer space arrived on our shores, cold, wet, and predictably seasick after a voyage across the Irish Sea _ no millpond _ in a boat made of hides stretched on a light wicker frame, and encountered the natives, warm and comfortable in several layers of wool, their first demand would have been "Take me to your tailor".
Hindsight sees more clearly then foresight and it is always easier to explain why something happened than to deduce that it will happen, but this is my interpretation of the circumstances that caused the tartan pattern to develop.
People like a bit of colour in their lives and in the days before bright lights and in a country where the sun can disappear behind a mountain in October and not reappear until March this must have been especially so. I once saw a very beautiful plaid in a tartan pattern made from several natural shades of undyed wool but, in general, natural materials are not sufficiently colourful. Colour therefore has to be added by dyeing and there are several ways in which dyes may be used. The Highlanders of old lived in small, isolated and self-sufficient communities. Their manufactures took place on the spot and if there was anything in the nature of a dye works it was on a pretty small scale using domestic-size dye pots. The Batik technique can be used to introduce a jolly look to cloth but is limited in scope and not suited to mass-production; good enough for use in the sort of climate where a handkerchief is big enough to give protection against the elements but too cumbersome to use for our own belted plaid. Tie-and-dye is another way of applying an irregular individual pattern and piece dyeing can apply a single over-all colour, but all these need big pots. If at least half of the yarn can be dyed in one batch, cloth can be woven coloured but small dye- batches and the consequent inevitable slight mismatches will cause streaks in the colour. A way round this is to dye the fleece in batches and blend these to give even colour, but the dyeing removes the natural grease from the fleece and this has to be replaced before the yarn can be spun, so a ready source of suitable oil is needed and this was not always available.
The human eye is very good at picking out minute errors in matching in the body of cloth but much less so if two slightly differing shades are separated by a band of a contrasting colour, so the simplest way of achieving a planned pattern is to make the cloth striped. Several colours can be used and the widths of the stripes can be varied, but the result is still not very exciting and a plain weft tends to degrade the colours of the stripes in the warp. Also, for best results an even quality of weft is required and this was not easy, even if possible, to manage. According to I.F.Grant, the Saxony type of spinning wheel was rare in the Highlands as late as the eighteenth century and most spinning would therefore have been done with the spindle. The spindle can spin a very fine and even yarn and its slow rate of production would not have mattered when there was a plentiful supply of spinners; however, lots of spinners would have spun lots of different thicknesses of yarn and the indications, from old specimens, are that it was the habit of Highland weavers to use the finest and best yarns for warps, which take the tension of weaving, and the rest as they came for the weft. There are specimens of old tartans in which the thickness of the weft yarn varies from one colour to another.
Despite the difficulties, though, stripes look to be a good bet and somebody must have been very gratified to find that most of problems were removed if the pattern was made striped in both directions. The rest is a matter of refinement of the basic idea. If the weft stripes are made the same as the warp, the pattern becomes symmetrical, and the cloth can be used any way round. In weaving, the weft pattern can be read from the warp, and all the weaver has to do is to work the pedals and throw the appropriate shuttle back and forth until the particular square is complete, then change colour and do the next one. That part of the job could be left to an apprentice, while the real weaver, whose skill was required to design the pattern and set up the loom, went off to start the next piece or to cut his peats or do anything else that he had on hand. It was only natural common sense to make the unit of pattern, the sett, of manageable size and to divide it into mirror image half- setts to repeat back and forth along and across the web. Both made the pattern easier to follow and to remember in an age when nothing was written down.
That is my hypothesis of how natural circumstances could have made the development of the tartan type of pattern almost inevitable in isolated, self- sufficient and wool-producing communities and I see some encouragement in the fact that Bhutan produces tartan-type patterns (But not tartan-type ballyhoo. Only in Scotland has that reached the status of an art-form.) and in a recent report the discovery of mummies clad in what appears to be tartan at an archaeological site in a mountainous district of southern China. It seems a tenable supposition that tartan is an outcome of environment rather than race and a peculiarity of people of the High Lands. Periodically, one hears of pockets of 'lost people', living among mountains, speaking a strange language unlike any known locally and wearing 'tartan'; the urge to dismiss such stories as travellers tales and romantic nonsense is compelling but perhaps there is something common to the countries called Alba, Alps, Albania and the like.
Tartan is a fascinating study which impinges upon a very wide range of disciplines, practical and theoretical, but the perceived history of tartan is based on little more than wild romanticism. blatant commercialism and uninformed blind guesswork. The search for the truth about tartan is very like one of those old- fashioned 'magic lantern' shows at which the operator examines a slide by a stray ray of light, twisting it this way and that, finally inserting it in the machine when, with any luck, the picture will come on the screen the right way up and the right way round. Unfortunately, the general public has been conditioned to prefer the upside-down, back-to-front versions and an exposition of the truth is generally met with disbelief, if not actual hostility.
The number of mixture colours (M) in a tartan is given by the formula M = «[N2 - N] where N is the number of 'starter' colours; the total number of shades (i.e. colours and mixtures) = M + N. However, the figures are more meaningful if shown separately.
James Logan (The Scottish Gael, 1832) was the first to publish tartan patterns which he did in the form of tables of measurements taken from samples of cloth obtained from the Bannockburn weavers, William Wilson & Son. Later, Logan passed these samples on to the artist McIan, for use in illustrating their joint work The Clans of the Scottish Highlands. Wilsons' records of the transaction, and their pattern books, have survived and so a full comparison can be made of Logan's and McIan's works with the originals; generally speaking, McIan has a poor reputation for accuracy, but this comparison shows him in much better light and Logan in rather worse.
In 1842, the brothers who were by that time calling themselves John Sobieski Stuart and Charles Edward Stuart, published their Vestiarium Scoticum, based on a manuscript that they claimed to possess, which was supposed to date from the late sixteenth century. D.C.Stewart and J.C. Thompson made a very detailed analytical study of the Vestiarium which was published as Scotland's Forged Tartans in 1980, three years after the Stewart's death, and showed the Vestiarium to be a complete forgery. Further study, particularly of the tartan patterns shown in the Vestiarium has confirmed their diagnosis.
In 1850, the brothers William and Andrew Smith, makers of Mauchline ware, and Thomas Smibert, set out to discover and illustrate the authentic tartans of their day. They did not find it easy, but their books, Authentic Tartans of the Clans and Families of Scotland and The Clans of the Highlands of Scotland set the standard for future generations of tartan picture books. The Smith book was notable for its illustrations, which were produced by a machine that drew closely-spaced parallel lines in opaque ink on black paper and gave a remarkable imitation of woven cloth.
D.W.Stewart made the first analytical attack on the history of tartan with Old and Rare Scottish Tartans (1892). 'O&R' is another that is remarkable for its illustrations, in this case hand-woven in silk.
In 1886, James Grant's The Tartans of the Clans of Scotland was published, a massive tome that was the prototype of the tartan picture book that has supplied the tourist market ever since with potted Clan histories and pictures, of varying quality, of their tartans. It is rare for one of these to have anything original to say about tartan or even to draw upon new research, but they can be useful to the student in providing a comprehensive collection of illustrations that can be of value to the student.
It was one of these tourist-market productions that spurred D.C.Stewart, son of 'D.W.' son, to write his book, The Setts of the Scottish Tartans (1950 and 1974), arguably the most important book ever written on the subject of tartan for it resulted in the formation of the Scottish Tartans Society of which he became the first archivist. Until his death in 1977, the Society under his guidance did valuable pioneering and research work and, although this fell into abeyance after his death and that of the Society's founder, Stuart Davidson, the nucleus and the foundation have survived. It is no exaggeration to say that Stewart laid the foundations of modern tartan research.
From time to time I have made efforts to record my own work and, mostly without success, to persuade those who should need no persuasion that tartan has some importance in the Highland context. For the generally-interested, the most useful of these are The Tartan Weaver's Guide and Understanding Tartan.
A steady stream of tartan picture-books continues to flow on to the market. The quality of the illustrations varies widely and only rarely does one have anything new to say.