Illuminated Manuscripts in Classical and Mediaeval Times/Chapter 15

The coloured pigments of the illuminators. Though mediaeval manuscripts are splendid and varied in colour to the highest possible degree, yet all this wealth of decorative effect was produced by a very few pigments, and with the simplest of media, such as size made by boiling down shreds of vellum or fish-bones, or else gum-arabic, or occasionally white of egg or a mixture both of the yoke and the white. In the main the technique of manuscript illumination is the same as that of panel pictures executed in distemper (tempera). An oil medium was unsuited to manuscript work because the oil spoilt the beautiful opaque whiteness of the vellum and made the painting show through to the other side.

Blue pigments. The most important blue pigment, both during classical and mediaeval times, was the costly and very beautiful ultramarine (azzurrum transmarinum), which was made from lapis lazuli, a mineral chiefly imported from Persia. This ultramarine blue was the cyanus or coeruleum of Theophrastus and Pliny. It is not only the most magnificent of all blue pigments, but is also the most durable, even when exposed to light for a very long period.

The general principle of the manufacture of ultramarine is very simple; consisting merely in grinding the lapis lazuli to powder, and then separating, by repeated washing, the deep blue particles from the rest of the stone. The process of extracting the blue was made easier if the lapis lazuli was first calcined by heat. This is the modern method, and was occasionally done in mediaeval times, but it injures the depth and brilliance of the pigment, and in the finest manuscripts ultramarine was used which had been prepared by the better though more laborious process without the aid of heat.

The proportion of pure blue in a lump of lapis lazuli is much smaller than it looks; the stone was and is rare and costly, and thus the finest ultramarine of the mediaeval painters was often worth considerably more than double its weight in gold.

Both in classical and mediaeval times it was usual for the patron who had ordered a picture to supply the necessary ultramarine to the artist, who was only expected to provide the less costly pigments in return for the sum for which he had contracted to execute the work.

Pliny (Hist. Nat. XXXIII. 120) tells a story of a trick played by a painter on his employer, who suspiciously watched the artist to see that he did not abstract any of the precious ultramarine which had been doled out to him. At frequent intervals the painter washed his brush, dipped in the ultramarine, in a vessel of water; the heavy pigment sank to the bottom, and at the end of the day the artist poured off the water and secured the mass of powdered ultramarine at the bottom.

It is interesting to note that Vasari, in his life of Perugino, tells precisely this story about Pietro, who was annoyed at the suspicions expressed by a certain Prior for whom he was painting a fresco. The Prior was in despair at the enormous amount of pigment that the thirsty wall sucked in, and he was agreeably surprised when, at the conclusion of the work, Perugino returned to him a large quantity of ultramarine, as a lesson that he should not suspect a gentleman of being a thief.

The library of Corpus Christi College, Cambridge, possesses a manuscript which affords a curious proof of the great value of ultramarine to the mediaeval illuminator. This is a magnificent copy of the Vulgate by a German scribe of the twelfth century, copiously illustrated with miniature pictures, many of which had backgrounds, either partially or wholly, covered with ultramarine. All through the book the ultramarine has in mediaeval times been very carefully and completely scraped off, no doubt for use in another manuscript. This theft has been accomplished with such skill that wonderfully little injury has been done to the beautiful illuminations, except, of course, the loss of splendour caused by the abstraction of the ultramarine.

In illuminated manuscripts ultramarine is very freely used. It is specially noticeable for the thick body (impasto) in which it is applied, so as very often to stand out in visible relief. The reason of this is that this, and some other blue pigments, lose much of their depth of colour if they are ground into very fine powder. Hence both the ultramarine and smalto blues are always applied in comparatively coarse grained powder; and this of course necessitates the application of a thick body of colour.

Smalto blues. Next in importance to the real ultramarine come the artificial smalto or "enamel" blues, which were used largely in Egypt at a very early date under the name of artificial cyanus; see Pliny, ''Hist. Nat.'' XXXVII. 119. Among the Greeks and Romans too this was a pigment of great importance, and when skilfully made is but little inferior in beauty to the natural ultramarine.

Smalto blue is simply a powdered blue glass or vitreous enamel, coloured with an oxide or carbonate of copper. Vitruvius (VII. xi. 1) describes the method of making it by fusing in a crucible the materials for ordinary bottle-glass, mixed with a quantity of copper filings. The alcaline silicate of the glass frit acts upon the copper, which slowly combines with the glass, giving it a deep blue colour. The addition of a little oxide of tin turns it into an opaque blue enamel, which when cold was broken up with a hammer, and then powdered, but not too finely, in a mortar.

Smalto blue is largely used for the simple blue initials which alternate with red ones in an immense number of manuscripts. The glittering particles of the powdered glass can easily be distinguished by a minute examination. Like the ultramarine, the smalto blue is always applied in a thick layer.

The monk Theophilus (II. 12), who wrote during a period of some artistic and technical decadence, the eleventh century, advises the glass-painter who wants a good blue to search among some ancient Roman ruins for the fine coloured tesserae of glass mosaics, which were so largely used by the Romans to decorate their walls and vaults, and then to pound them for use.

Azzurro Tedesco or Azzurro della Magna, German blue, was much used by the illuminators as a cheap substitute for ultramarine. This appears to have been a native compound of carbonate of copper of a brilliant blue colour. It was occasionally used to adulterate the costly ultramarine, but this fraud was easily discovered by heating a small quantity of the pigment on the blade of a knife; it underwent no change if it was pure; but if adulterated with Azzurro della Magna it showed signs of blackening.

Indigo blue. The above-mentioned blues are all of a mineral character, and are durable under almost any circumstances. To some extent however the vegetable indigo blue was also used for manuscript illuminations, both alone and also to make a compound purple colour.

Colours of all kinds prepared from vegetable or animal substances required a special treatment to fit them for use as pigments in solid or tempera painting. Though indigo and other colours of a similar class are the best and simplest of dyes for woven stuffs, yet they are too thin in body to use alone as pigments. Thus both in classical and mediaeval times these dye-pigments were prepared by making a small quantity of white earth, powdered chalk or the like absorb a large quantity of the thin dye, which thus was brought into a concentrated and solid, opaque form, not a mere stain as it would otherwise have been.

These kinds of pigments are described by Pliny, ''Hist. Nat.'' XXXV. 44 and 46; and by Vitruvius, VII. xiv. Eraclius in his work on technique, De artibus Romanorum, calls them colores infectivi, "dyed colours," an accurately expressive phrase.

One method, occasionally used for the cheaper class of manuscripts, was to paint on to the vellum with white lead, and then to colour it by repeated application of a brush dipped in the thin dye-pigment. Many of the colours mentioned below belong to this class.

Green pigments. A fine soft green much used in early manuscripts is a natural earthy pigment called terra verde or green Verona earth. This needs little preparation, except washing, and is of the most durable kind; it is a kind of ochre, coloured, not with iron, but by the natural presence of copper.

A much more brilliant green pigment was made of verdigris (verderame) or carbonate of copper, produced very easily by moistening metallic copper with vinegar or by exposing it to the fumes of acetic acid in a closed earthen vessel; see Theophilus, I. 37.

Verdigris green was much used by manuscript illuminators, especially during the fifteenth century, when a very unpleasant harsh and gaudy green appears to have been popular. When softened by an admixture of white pigment, verdigris gives a pleasanter and softer colour.

A native carbonate of copper, which was called by the Romans chrysocolla, was also used for mediaeval manuscripts. It is, however, harsh in tint if not tempered with white. Both the last-named pigments were specially used with yoke of egg as a medium.

Prasinum, a vegetable green made by staining powdered chalk with the green of the leek, was sometimes used.

Cennino Cennini also recommends a grass green made by mixing orpiment (sulphuret of arsenic) and indigo.

One of the best and most commonly used greens was made by a mixture of smalto blue and yellow ochre; other mixtures were also used.

Red pigments. Red and blue are by far the most important of the colours used in illuminated manuscripts, and it is wonderful to see what variety of effect is often produced by the use of these two colours only.

The chief red pigments used by illuminators are vermilion (cinnabar or sulphuret of mercury) and red lead (minium), from which the words miniator and miniature were derived, as is explained above at page 31.

Both these pigments are very brilliant and durable reds, the more costly vermilion is the more beautiful of the two; it has a slightly orange tint.

Illuminators commonly used the two colours mixed. One receipt recommends one-third of red lead combined with two-thirds of vermilion; Jehan le Begue's manuscript, § 177 (Mrs Merrifield's edition, Vol. I. page 141). Vermilion was prepared by slowly heating together metallic mercury with sulphur. Red lead (a protoxide of lead) was made by roasting white lead or else litharge (ordinary lead oxide) till it absorbed a larger proportion of oxygen.

Rubrica or Indian red was a less brilliant pigment, which also was largely used in illuminated manuscripts, especially for headings, notes and the like, which were hence called rubrics. Rubrica is a fine variety of red ochre, an earth naturally coloured by oxide of iron ; another variety was called bole Armeniac. In classical times the rubrica of Sinope was specially valued for its fine colour.

In addition to these mineral and very permanent reds there were some more fugitive vegetable and animal scarlets and reds which were used in illuminated manuscripts.

Murex. One of these, the murex shell-fish, has already been mentioned for its use as a dye for the vellum of the magnificent Byzantine and Carolingian gold-written manuscripts. The murex was also used as a color infectivus by concentrating it on powdered chalk.

Kermes. Another very beautiful and important carmine-red pigment was made from the little kermes beetle (coccus) which lives on the ilex oaks of Syria and the Peloponnese. It is rather like the cochineal beetle of Mexico, but produces a finer and more durable colour, especially when used as a dye. For the woven stuffs of classical and mediaeval times, and in the East even at the present day, the kermes is one of the most beautiful and important of all the colours used for dyeing. The mediaeval name for the kermes red was rubeum de grana; when required for use as a pigment it appears to have been usual, not to extract the colour directly from the beetle, but to get it out of clippings of red cloth which had been dyed with the kermes, by boiling the cloth in a weak solution of alkali and precipitating the red pigment from the water with the help of alum.

The reason for this method is not apparent. Possibly it was first done as a means of utilizing waste clippings of the costly red cloth, and then, when the habit was established, no other method was known to the colour-makers, who in some cases bought pieces of cloth on purpose to cut them up and use in this way. The scarletum blanketum mentioned at page 234 was bought for this purpose.

Madder-red was also used as a pigment by boiling the root of the madder-plant (rubia-tinctorium), and then using the concentrated extract to dye powdered chalk. Various red and purple flowers, such as the violet, were used in the same way as colores infectivi.

Lake-red (lacca or lac) was made and called after a natural gum or resin, the lach of India; see Cennino Cennini, § 44.

This is a beautiful transparent colour, which, in some fine manuscripts of the fifteenth century, is used as a transparent glaze over burnished gold, the effect of which is very magnificent, as the metallic gleam of the gold shines through the deep transparent red of the over-painting. Lake was also used as an opaque, solid pigment by mixing it with white, which at once gave it "body," and destroyed its transparency.

Purple of a very magnificent tint was occasionally made by a mixture of ultramarine with the carmine-red of the kermes beetle; this was specially used by the illuminators of the fourteenth and fifteenth centuries.

Yellow ochre, a fine earth pigment coloured by iron, was the principal yellow of the illuminators.

In late manuscripts orpiment (sulphuret of arsenic), which is a more brilliant lemon-yellow, was occasionally used; see Cennino Cennini, § 47.

Litharge yellow, an oxide of lead, was another important colour, but more especially for the painter in oil, who used it very largely as a drier.

Another fine ochreous earth of a rich brown colour was the terra di Siena or "raw Siena"; the colour of this was made warmer in tint by roasting, thus producing "burnt Siena."

White pigments were perhaps the most important of all to the illuminator, who usually only used pure colours for his deepest shadows; all lights and half tints, both in miniature pictures and in decorative foliage, being painted with a large admixture of white. The use of this system of colouring by Fra Angelico and many painters of the Sienese school has been already referred to; see page 190.

For this reason it was very important to use a pure and durable white pigment which would combine well with other colours.

Bianco di San Giovanni was in all respects one of the best of the whites used by illuminators.

This was simply pure lime-white, made by burning the finest white marble; the lime was then washed in abundance of pure water, then very fine ground and finally dried in cakes of a convenient size; see Cennino Cennini, § 58; and Theophilus, I. 19. The medium used with it was the purest size or gum Arabic of the most colourless kind.

Another white pigment was made of powdered chalk and finely ground egg-shells; this was a less cold white than the bianco di San Giovanni.

White lead (cerusa or biacca) was also used, especially by the later illuminators, but with very unfortunate results, since white lead is liable to turn to a metallic grey or even black if exposed to any impure sulphurous atmosphere.

Many beautiful manuscripts have suffered much owing to the blackening of their high lights which had been touched in with white lead; especially manuscripts exposed to the gas- and smoke-poisoned air of London or other large cities.

The biacca of the mediaeval illuminator was made in exactly the same way that Vitruvius and Pliny describe; see Vitr. VII. xii.; and Pliny, ''Hist. Nat.'' XXXIV. 175.

A roll of lead was placed in a clay dolium or big vase, which had a little vinegar at the bottom. The top was then luted down, and the jar was left in a warm place for a week or so, till the fumes of the acetic acid had converted the surface of the lead into a crust of carbonate. This carbonate of lead was then flaked off and purified by repeated grinding and washing.

In order to keep the white pigments perfectly pure, some illuminators used to keep them under water, so that no dust could reach them.

Black inks. Two inks of quite different kinds were used for the ordinary text of mediaeval manuscripts.

One of these was a pure carbon-black (modern Indian or Chinese ink); this has been described under the classical name atramentum librarium ; see above, page 27. The great advantage of this carbon ink is that it never fades; it is not a dye or stain, but it consists of very minute particles of carbon which rest on the surface of the vellum.

The other variety was like modern black writing ink, only of very superior quality. This acts as a dye, staining the vellum a little below the surface. Unfortunately it is liable to fade, though when kept from the light (as in most manuscripts) it has stood the test of time very well.

Sometimes the mediaeval illuminators distinguished these two kinds of black ink, calling the first atramentum and the second encaustum; but frequently the names are used indifferently for either: see Theophilus, I. 40. The encaustum was made by boiling oak-bark or gall-nuts, which are rich in tannin, in acid wine with some iron filings or vitriol (sulphate of iron). The combination of the iron and the tannin gives the inky black. Both these black inks were used with gum Arabic.

A great part of the beauty of mediaeval manuscripts is quite unconnected with their illumination. The plain portion of the text, from the exquisite forms of its letters and the beautiful glossy black of the ink on the creamy ivory-like vellum page, lighted up here and there by the crisp touch of the rubricator's red, is a thing of extraordinary beauty and charm. This perfection of technique in the writing and beauty of the letters lasted considerably longer than did the illuminator's art. Hence in some of the manuscripts of the period of decadence, executed during the fifteenth century, the plain black and red text is very superior in style to the painted ornament; and one cannot, in some cases, help regretting that the manuscript has not escaped the disfigurement of a coarse or gaudy scheme of illumination.

Red inks were of three chief kinds, namely the vermilion, red lead, and rubrica or red ochre, which have been already mentioned.

Purple ink was used largely, not often for writing, but for the delicate pen ornaments of the initials in certain classes of late Italian and German manuscripts. A vegetable pigment seems to have been used for this; the lines appear to be stained, and do not consist of a body-colour resting on the surface of the vellum.

Gold writing is usually executed with the fluid gold pigment, but in later manuscripts very gorgeous titles and headings are sometimes done with burnished leaf gold applied on the raised mordant, the writing being first done with a pen dipped in the fluid mordant.

The pencils and pens of the Illuminator. Two quite different classes of pencils were used for lightly sketching in the outline of the future floral design or miniature.

One of these was the silver-point or lead-point, very much like the metallic pencil of a modern pocket-book. The use of the silver-point was known in classical times; Pliny (Hist. Nat. XXXIII. 98) remarks as a strange thing that a white metal like silver should make a black line when used to draw with. It is, however, rather a faint grey than a black line that a point of pure silver makes, especially on vellum, and so it was more usual for illuminators to use a softer pencil of mixed lead and tin; Cennino recommends two parts of lead to one of tin for making the lead point, piombino.

Another kind of pencil was made of a soft red stone, which owed its colour to oxide of iron. From its fine blood-red tint the illuminators called it haematita, lapis amatista or amatito, hence an ordinary lead pencil is now called either lapis or matita in Italy. This stone is quite different from the hard haematite which was used in classical times for the early cylinder-signets of Assyria.

The harder varieties of the amatista or haematite were used to burnish the gold leaf in manuscripts, small pieces being polished and fixed in a convenient handle. They were also used as a red pigment, the stone being calcined, quenched in water and finely ground; see Cennino Cennini, § 42.

Besides the hard red chalky stone (amatita rossa) used for outlines by the illuminators, a somewhat similar black stone (amatita nera) was also used, but not so commonly as the red.

The pens of illuminators. In early times, throughout, that is, the whole classical period and probably till about the time of Justinian, the sixth century A.D., scribes' pens were mostly made of reeds (calamus or canna); and occasionally silver or bronze pens were used; see above, page 29.

But certainly as early as the eighth century A.D. and probably before that, quill-pens came into use and superseded the blunter and softer reed-pen.

Such exquisitely fine lines as those in many classes of mediaeval manuscripts could only have been made with some very fine and delicate instrument like a skilfully cut crow's quill or other moderately small feather.

The pen was a very important instrument for the illuminator, not only when his pictures were mainly executed in pen outline, like many of those in the later Anglo-Saxon manuscripts, but also in such microscopically delicate miniature work as that in the Anglo-Norman historiated Bibles of the thirteenth century; in these much of the most important drawing, such as the features and the hair of the figures, was executed, not with a brush, but with a quill-pen, which in the illuminator's skilful hand could produce a quality of line which for delicacy and crisp precision of touch has never been surpassed by the artists of any other class or age.

Brushes were, as a rule, made by the illuminators themselves, so as to suit their special needs and system of working. Cennino (§§ 63 to 66) and other writers give directions for the selection of the best hair and the mode of fixing it so as to give a finely pointed brush. Ermine, minever and other animals of that tribe supplied the best hair for the brushes required for very minute work. But a great number of other animals provided useful material to the craftsman who knew the right places to select the hair from, and, a still more important thing, understood how to arrange and fix it in a bundle of the best form.

The implements of scribes and illuminators. The following is a list of the principal tools and materials required by the illuminator of manuscripts, including those which have been already described.

Pens, pencils and chalk of various sorts, as described above.

Brushes made of minever, badger and other kinds of hair.

Grinding-slabs and rubbers of porphyry or other hard stone, and a bronze mortar.

Sharp penknife and palette knife.

Rulers, and a metal ruling-pen.

Dividers to prick out the guiding-lines of the text.

Scissors for shaping the gold leaf.

Burnishers, stamps, and stili for ornamenting the gold.

Small horns to hold black and red ink; see fig. 53 on page 209.

Colour-box, palette, pigments, gold leaf and media of various kinds.

Sponge and pumice-stone for erasures.

Miniatures representing a scribe writing a manuscript are the commonest of all subjects in several classes of illuminated manuscripts. For example the first capital of Saint Jerome's Prologue in the historiated Anglo-Norman Vulgates almost always has a very minute painting of a monastic scribe, seated, writing on a sloping desk, with his pen in one hand and his penknife in the other.

In one respect such scenes are always treated in a conventional way; that is, the scribe is represented writing in a complete and bound book, whereas both the writing of the text and the illuminations were done on loose sheets of vellum, which could be conveniently pinned down flat on the desk or drawing-board.

The processes employed in the execution of an illuminated manuscript of the fourteenth or fifteenth century were the following;

First, if the text were to be in one column, four lines were ruled marking the boundaries of the patch of text and the margin. These four lines usually cross at the angles and are carried to the extreme edge of the vellum.

Next, the scribe, with a pair of dividers or compasses, pricked out at even distances the number of lines which were to be ruled to serve as a guide in writing the text. These pricked holes were, as a rule, set at the extreme edge of the vellum, and were intended to be cut off by the binder, but in many manuscripts they still remain. The scribe then filled the space within the first four marginal lines with parallel ruled lines at the intervals indicated by the pricks at the edge.

In early manuscripts the guiding lines to keep the text even are usually ruled, not with colour or ink, but simply traced with a pointed stilus, which made a sufficiently clear impressed line on the vellum, showing through from one side to the other.

In the twelfth and thirteenth centuries the practice began of ruling the lines with a lead point; and then, from the fourteenth century onwards, they were usually ruled with bright red pigment ; this has a very decorative effect in lighting up the mass of black text, and thus we find in many early printed books these red guiding lines have been ruled in merely for the sake of their ornamental appearance.

This ruling was nearly always done with special metal ruling-pens, very like those now used for architectural drawing; and thus the lines are perfectly even in thickness throughout.

The next stage in the work was the writing of the plain black text. In early times it appears to have been usual, or at least not uncommon, for the same hand to write the text and add the painted illuminations, but when the production of illuminated manuscripts came mostly into secular hands, the trades of the scribe and the illuminator were usually practised by different people; and in late times a further division still took place, and the miniaturist frequently became separated from the decorative illuminator.

Thus we find that in many manuscripts the scribe has introduced in the blank spaces minute guiding letters to tell the illuminator what each initial was to be, and, especially in fifteenth century Italian manuscripts, instructions are added for the miniaturist, telling him what the subject of each picture was to be. These instructions were commonly written on the edge of the page so that they were cut off by the binder, but in many cases they still exist, not obliterated by the subsequent painting.

But to return to the progress of the page, when the scribe had finished the plain text, leaving the necessary blank spaces for the illuminated capitals and miniatures, the work of decoration then began.

As a rule the decorative foliage and the like was finished before the separate miniatures, if there were any, were begun. First the illuminator lightly sketched in outline the design of the ornament, using a lead point. Next, wherever burnished gold was to be introduced, the thick mordant-ground was laid on; the gold leaf was then applied and finished with tooling and burnishing.

The reason why the gold was applied before any of the painting was begun was this; the long rubbing with the burnisher acted not only on the gold leaf, but also naturally rubbed the vellum a little way all round it. This would have smudged the painting round the gold if it had been applied first. Moreover, the burnisher was liable to carry small particles of gold on to the surrounding vellum, which would have given a ragged look to the design, if the adjacent surfaces had not been subsequently covered with pigment. In cases where there is an isolated gold boss there is usually a slight disfigurement from the burnisher rubbing the vellum all round the gold. In these cases the outline of the gold was made clean and definite by the addition of a strong black outline, as is mentioned above.

When the whole of the burnished gold was finished, the painting was then executed. If any fluid gold pigment were used, that was usually added last of all.

In some cases, in the later and cheaper French and Flemish manuscripts, the ornaments in the borders were not specially designed and sketched in for the manuscripts but previously used outline patterns were transferred on to the vellum by a bone stilus and ordinary transfer paper, made by rubbing red chalk all over its surface.

In some of the better class of manuscripts with the "ivy-leaf" border, the illuminator has made the general design of one page serve for the next one in this way; when he had drawn in the main lines of the scroll-pattern on the borders of one page, he held the vellum up to the light and so was able to trace the pattern through from the other side of the leaf. To prevent monotony he varied the design by introducing different little blossoms among the repeated scroll-work which formed the main pattern.

When the scribe, the rubricator, the illuminator and the miniaturist (either as one or as several different people) had completed the manuscript it was ready for the binder. As an indication of the order in which the leaves of the manuscript were to be bound, the scribe usually placed on the lower margin of the last page of each "gathering" of leaves a letter or number.

In the earliest printed books these guiding letters, or signatures as they are called, were added by hand in the same way ; but in a few years the regular and more developed system of printed signatures was introduced.

Scribes' signatures at the end of manuscripts are comparatively rare, but they do occasionally occur in various interesting forms. My friend Mr W. J. Loftie kindly sends me the following:

In a Sarum Missal of the fifteenth century at Alnwick Castle,

"Librum scribendo Jon Whas monachus laborabat, Et mane surgendo multum corpus macerabat."

More commonly manuscripts terminate with a vague phrase invoking a blessing on the scribe, such as this, from a Bible in the Bodleian (No. 50),

"Qui scripsit hunc librum Fiat collectum in paradisum."

Or this, which occurs in several manuscripts,

"Qui scripsit scribat, Semper cum Deo vivat."

In another manuscript Vulgate in the Bodleian (No. 75), the owner of the book, who was named Gerardus, has recorded the fact in this fanciful way,—

"Ge ponatur et rar simul associatur Et dus reddatur cui pertinet ita vocatur."