Only in light of the preternatural majesty of the Book of Kells do the Lindisfarne Gospels appear less than divine. Written and illuminated around the second decade of the eighth century, the volume astounds, its intricate decoration, brilliant hues, and inventive shapes livening page after page. Along with the intricately geometric “carpet pages,” the six major “incipit pages” that open the most crucial sections present the scribe’s genius at its height.
An enormous initial or monogram, filled and interwoven with ornament, dominates each of these pages, and the following letters decrease in scale toward the bottom of the page as they near the half-uncial text’s size — a “diminuendo” effect, as art historians and paleographers describe it by analogy to the musical notation for diminishing in intensity. Many Insular illuminated manuscripts, from the Cathach of Columcille to the Book of Durrow, showcase the technique. Yet no major work has developed this beautifully named term, coined by Carl Nordenfalk in 1947, beyond its initial formulation:
“… [T]he following letter also being designed as an initial, though somewhat smaller … followed as a rule by one or two letters acting as intermediaries between the sizes of the initial letters and that of normal script.”
This definition has remained essentially unmodified through 2003, when Michelle Brown writes of “display script of diminishing scale (‘diminuendo’).”
Nordenfalk and Brown are entirely correct in their portrayals of the respective initials; but by limiting the term to one dimension, size, they ignore the intricacies engendered by the effect in a manuscript such as the Lindisfarne Gospels. Eadfrith, the presumed scribe and illustrator, did not mechanically shrink his display letters line by line as one might do today by changing the point size in a word processing program’s menu. An expert artist, he adapted each set of forms to its height; a letter near the bottom of the page differed in many ways from one at the top. While the Lindisfarne Gospels embrace the freedom inherent in Insular art and thus follow few fixed aesthetic rules, examination of the incipit pages reveals several tendencies as the letters proceed downward. First, the letters diminish in size; second, four to six main levels of hierarchy occur; third, the letters’ stroke widths grow thinner; fourth, the letterforms become narrower and more rectangular; and finally, ornament within the strokes largely disappears and the designs surrounding and filling the negative spaces become simpler.
Implementing the diminuendo effect is therefore not a simple proposition. Although the Lindisfarne Gospels defy a certain level of logical inquiry with their astounding ornateness, one may still ask why the diminuendo exists (aside from it being a Hiberno-Saxon tradition) when it requires such planning and care compared to, for example, a page composed solely of an initial and standard text. In explaining the reason, Nordenfalk’s initial discussion remains entirely accurate: “The decorative effect of the large initial [therefore] is not disconnected from the text, but is graduated down into it in a sort of diminuendo.” The diminuendo integrates the initial on both aesthetic and functional grounds. Visually it prevents a jarring transition, allowing for asymmetry but not imbalance on the page, and it makes the text far easier to read; many lines of text nestled up against an enormous first letter would confuse a reader as to what follows the initial. The Lindisfarne Gospels’ diminuendo, then, is a complex system representative of the complementary form and function in Eadfrith’s masterpiece.
The incipit pages
The six major incipit pages of the Lindisfarne Gospels occur, as per their name, at the beginning of the most important parts of the book. Facing a complementary carpet page, the first (the “Novum opus” page) starts Jerome’s preface to the Vulgate. Four of the remaining five comprise the first words of the four gospels: the “Liber generationis” page at the beginning of Matthew; the “Initium evangelii” page of Mark; the “Quoniam quidem” page of Luke; and the “In principio” page of John. The Chi-rho page is the one outlier, appearing at the first mention of Christ’s name in the Gospel of Matthew. A single elaborate letter or monogram dominates each of the pages; in all but the first (which precedes the less important prefatory material), the initials cover between a third and half of the surface. Eadfrith outlined these Insular-style initials in yellow and black and filled their strokes and counters with nearly every type of Hiberno-Saxon ornament: bird heads, dog heads, peltae, trumpet patterns, spirals, whorls, diagonal key-patterns, abstract interlace, and more. Decorated display letters follow the initials, ranging in style from Insular half-uncials to Roman square capitals. Unlike the text pages, spaces do not separate one word from another, although the “Novum opus” page uses comma-like marks to divide words. Red dots pervade the vellum, two rows tracing the initials’ contours and patterns providing a background for the display capitals. The letters and decoration create a roughly rectangular mass and, in all but the “Novum opus” page, are partially surrounded by a pattern-filled frame. Eadfrith uniquely chose to leave a gap in each frame near the lower-right corner as if to allow the text to continue unrestricted to the next page. Appearing bare and almost an afterthought, a few words in a large half-uncial describe what the page introduces, and Aldred’s tenth-century gloss translates each Latin word into Anglo-Saxon.
After the awe-inspiring size and elaborate nature of the initials, the diminuendo effect is perhaps the next most prominent feature of the incipit pages. As one’s eyes move from the top of the sheet to the bottom, one sees the letters grow smaller. The range of the height difference over a page varies, with the average size of the letters on the last line of the page being about twenty-five times smaller than the page-spanning initial. Except for the letters on the first line following the larger initial, Eadfrith consistently employed a minimum unit of a line of text. While two or even three consecutive lines would often have the same height, he never scaled letters within a line aside from the shrunken forms used to condense a line to the allotted space. The final line on the incipit pages remains between two and ten times larger than the body text size, except for the “Novum opus” page, which contains a final line of the same size and half-uncial style as the body text. In this manner each of these pages transitions the eye from a full page initial to a decorative but manageable introduction to the text.
Grouping by height, style of decoration, and, typically, by line, each of the incipit pages has between four and six hierarchical levels. The first level is the decorated initial: N in “Novum,” LIB in “Liber,” XPI in the Chi-rho page, ini in “Initium,” Q in “Quoniam,” and INP in “In principio.” The second level concludes the first line in the gospels, but the Chi-rho page’s horizontal initial forces the second level onto the second line and the “Novum opus” page fits both the second and third levels on the first line. The intermediate two or three levels tend toward conservatism, often shrinking gracefully with few decorative or structural changes to the letterforms. In most of the incipit pages, the final line is also the last hierarchical level and the smallest on the page.
As the height of the letters on the pages decreases, their strokes narrow. The initials exhibit a relatively thick average stroke width, especially in the case of the INI and INP ligatures that join two vertical strokes to make a double-width stroke. Although some of the monograms have a uniform stroke width across two or three letters, several demonstrate a marked thinning; the b in “Liber” and the PI in “XPI” possess a stroke about half the thickness of the preceding letter. Intriguingly, Eadfrith’s technique of using thicker strokes for the biggest letters runs contrary to the modern typographic practice of making display typefaces more delicate than the corresponding text face. If one sees both styles, however, as increasing the detail of the largest letters — one with decoration inside the stroke, the other using the stroke itself for nuance — then perhaps they do not contradict. The group of letters following the initials thins again by about half the previous width, but the following set of letters usually attenuates only a bit. For the majority of incipit pages, the stroke width remains the same after this point, though in a few pages the stroke width of the last line diminishes with its height. Through these changes, the artist ensures that each letter has enough space within and around its strokes to allow the embellishment it requires. If the stroke widths all stayed the same, the page would appear overly even and the smaller letters would become more prominent than they should be if a visual diminishing is the goal.
The Lindisfarne Gospels’ diminuendo also trends toward narrower, more rectangular letterforms as one proceeds down the page. The initials all exhibit Insular half-uncial-forms: rounded and originating in the minuscule (lowercase) rather than the majuscule (uppercase). Eadfrith manipulates the traditional shapes of each letter to combine them and create an engaging composition; the L in “Liber” curls to have its arm terminate in the B’s counter, the X in “XPI” stretches across the page, and the three N’s feature different styles for the middle arm. While the largest of the other display letters maintain this Insular character, they become more condensed, most likely to save space since the great roundness of the initials would severely limit the number of words per page if it applied to all the capitals. As the letters become narrower in the middle lines, they adopt forms more typical of the Roman square capital than the half-uncial, swapping a semi-circular E for a square-stemmed one and a sinuous, tailed U for a straight-sided version. This switch is inconsistent and incomplete, however, and condensed rounded half-uncial letters continue to exist in the middle and bottom lines, as do fascinating flat-sided expressions of characteristic half-uncial forms. This trend reverses suddenly in three of the incipit pages (the “Novum opus,” Chi-rho, and “Initium evangelii” pages), in which the letters in the last lines appear remarkably close — and in one case, identical to — the half-uncial text, with the otherwise unseen circular A’s and E’s with closed upper counters. This return to roundness perhaps amounts to an attempt to integrate the incipit pages with the text even more closely.
The size of the letters also correspond to the amount of decoration they receive. The initials, being the thickest and the largest, are the most fully adorned. Abstract and zoomorphic interlace run throughout their strokes and, in most cases, within the negative spaces surrounding them. Spirals, whorls, trumpet patterns, peltae, and animal heads project from the letters, creating appendages that further emphasize their importance. Such complex ornament is on display that the sea of spirals emanating from the XPI of the Chi-rho page blankets a third of the area and panelling partitions many of the initials into compartments of separate, elaborate designs. If the second or third letters of the monograms differ at all from the first, it is that they possess only slightly less sophisticated decorations. In all but the “Novum opus” page, though, the letters following the initials lack much of the extravagance of their predecessors. For three of the five, these letters remain visually connected to the initials through similar colors or textures either within the strokes or in the interstices; the other two, the “Quoniam quidem” and “Initium evangelii” pages, show little apparent relation. Empty vellum fills the strokes of two of the letterforms in this group and one in the next. Brown posits that the roughness of the visible drawings indicates that Eadfrith never finished these letters. She suggests that he might have darkened the outlines and filled them with a pale color (as in the “Initium evangelii” page) or gilded them. Although these second and third sets of decorated letters have lost some of the brilliance of the initials, they regain a measure of intricacy with the expansion of the small red dots from a two-dot contour to thick and complicated patterns.
By the middle lines, simple black stroke fills and colored washes for certain counters have replaced the previous busy, individual designs. Far fewer types of ornament exist to differentiate each letter; only bird heads, spirals, a single row of red dots, small triangles that wedge into color-filled counters, and sporadic instances of simple interlace remain. The red dot patterns continue to liven the spaces between letters toward the bottom of the page, but by the final line most pages have exchanged diaper or interlace patterns for a headline and baseline of double dots and a single row outlining the letters, bringing the final words closer to the simple style of the text.
These five trends combine to produce the diminuendo effect, fulfilling the aim of allowing the initial and the text to coexist in the same volume despite their many differences. The care that Eadfrith takes in altering each letter from the top to the bottom of the page suggests the aesthetic importance of the technique. A benefit of the symmetry common to much Classical art is that it is easy to implement. The Hiberno-Saxon asymmetry, however, requires a strong sense of line, plane, and rhythm; only by carefully adjusting the letterforms can the artist create a dominant element that attracts the viewer but does not overwhelm the page such that nothing else is visible. The color-filled counters, the red dot patterns, the clear levels of hierarchy — these all serve to balance the composition, creating pages that capture one’s attention with contrast and complexity while avoiding a sense of disproportion that would be off-putting to the reader.
The Lindisfarne Gospels’ diminuendo also produces dividends for the reader on a functional level. Without it, approximately thirty lines of the half-uncial main text would abut the initial. The initials, spanning the height of the page, carry the eyes far downward. Making the jump from near the bottom of the page back to the top would therefore be disconcerting and confusing, which is exactly what a designer would try to avoid when creating a book for solitary reading. Eadfrith did not design the Lindisfarne Gospels for immediate accessibility — the sophisticated ornament and creative letterforms attest to this — but the careful spacing of the text pages, the use of per cola et commata punctuation to clarify the meaning of a Latin sentence with the length of the line, and the careful placement of major and minor initials to show divisions in the text suggest a concern for readability. The diminuendo as implemented guides the reader down each incipit page without any unmanageable transitions in size or style and thus allows the book to unite powerful design and textual accessibility.
Not merely a matter of decreasing each successive line’s height, the diminuendo in the incipit pages of the Lindisfarne Gospels consists of adjustments in multiple dimensions: size, width, roundness, letter style, color, and pattern. With this highly sophisticated technique, Eadfrith demonstrates his artistic genius, simultaneously elevating the word of God and bringing it to His faithful followers. This mix of mundane and holy qualities perhaps would suggest that if the Book of Kells is “the work of an angel and not of a man,” the Lindisfarne Gospels is the work of a divinely-inspired man who created it for both Heaven and Earth.
This paper was written for Professor Robert Nelson’s 2009 class Art of Christian Empires: From Constantine to Charlemagne at Yale University. I have revised it slightly. I have also experimented with adding links for unusual terms and moving the original in-line citations to Medium’s paragraph-level notes.
Bischoff, Bernhard. Latin palaeography: Antiquity and the Middle Ages. Trans. Daíbhí Ó Cróinín and David Ganz. Cambridge: Cambridge University Press, 2006.
Brown, Michelle. The Lindisfarne Gospels: Society, spirituality and the scribe. London: The British Library, 2003.
Codex Lindisfarnensis = Kendrick, T.D., et al. Evangeliorum quattuor codex Lindisfarnensis. Oltun: Urs Graf, 1956–1960.
Nordenfalk, Carl. “Before the Book of Durrow.” Acta archaeologica 43. 1951: 141–174.