Examples of Irish Knots

I visited Dublin in 2001. The National Museum of Ireland in Kildare Street is splendid. Here are some patterns.

Since these knots were carved on stone, I am showing both the grey and coloured in versions. The pattern on the right took me ages to understand, and even longer to draw on the computer! The four spirals are asymmetric, which is hard to reconcile with the symmetry of the rest of the design.

Celtic knots from the Book of Kells

The Book of Kells is a beautiful illuminated manuscript kept in the Old Library of Trinity College in Dublin. You can buy a CD ROM of the images (or go to Dublin and see the book for yourself) which is the best way of seeing the images. I have tried to copy some of the abstract knots below. I'm afraid that the proportions and angles are not necessarily accurate. They are also not to scale. Some of the original knots are tiny. How they did them without using a magnifying glasses (that hadn't been invented at that point), I'll never know!

I'll start with a simple border. In the Book of Kells, this is bent into a curve. It only has one colour, but I have also given a two colour version so you can trace the strings. The underlying pattern reminds me of a Greek Key.

At first glance, this looks like a plait. Look closer, and you can see that it's a series of knots. It is used as packing inside an illuminated initial letter. As the letter tailed off into a point, the knots got simpler and narrower. Again, the 2-colour is my version to explain the pattern.

Here are some letters. The top colouring is as in the Book of Kells, and below, my own colouring. These knots have a very angular feel to them. The letters are A, V, A and T. The knot in the middle is the middle of the T is a common device. In other parts of the Book of Kells, there are all sorts of people and monsters look very surprised to find they have a knot where their stomachs should be!

This is one of the roundels in the whole page pictures. One interesting thing about it is that for most of the pattern, the lines are tripled. This often happens with the knots on Manx crosses. There, the tripling is interpreted as two strands, which pass over and under over strings together. Here the three lines are interpreted as separate strings, which pass over and under strings separately. This makes a more complicated pattern!

This is a line filler for one of the grand illuminated initial letters. Although you might assume that it had some symmetry, each of the internal knots are different. I couldn't work out whether there were two strings or not until I coloured it in. I wonder if the original designer knew! Or perhaps he made up a knot with real rope, and then copied it.

This should really be three-way symmetry, with lovely swirly spirals on the ends. Here is my attempt on a square grid!

This is a filling for the letter M. The colouring on the left is the original. They coloured each of the knots in contrasting yellow or blue to show them up, as opposed to following one string through with one colour, as I like to do. The strings were followed through with some of the animal patterns, but the abstract patterns didn't, and often only used one colour.

For my version of the colouring, the light and dark green are the same string. The pale blue is another string filling the central stroke, and the red and blue are small loops. For the right hand stroke, the yellow and brown are one string. In the original the knots on the right were spaced out to fill the gaps better. It seems obvious that the right hand side was supposed to match the left. Unfortunately the top was started as a mirror image while the bottom was a copy. By the time they got to the corner, things were in a right muddle and they sorted it out as they could. Perhaps it was a 'work experience' monk who did the right hand side!

Here are some crosses, from one of the full-page illustrations. The Book of Kells colouring seems to show a red string with yellow edges, but if you look carefully, the yellow 'edges' go over and under each other! In my colouring, you can see that there are only two strings, one being dark blue and dark red and the other mid blue and red. In the third cross, the green and yellow/brown are also part of the same two same strings.

These are two similar borders, but in different places in the book. They have two rows of interlinked knots. They are simple, but use more strings than the crosses above. There are 4 independent strings in each border. The one on the right has a more complex unit knot, and so a longer repeat.

This is another entangled double row of knots, but this time it only has two strings. This patterns has the strings in pairs, almost as if they were the edges of a wide ribbon, yet going under and over independent of each other, like the crosses above. But here this isn't high-lighted by the colouring. I've produced my own colouring on the right to show this. The second one changes the edges' colour half way through, so you can see how the ribbon weaves through itself.

This is a more symmetrical pattern, with the four knots pointing into towards a centre. Again, you can look at the pairs of strings as edges of a ribbon, shown on the right. I have only given the middle of this design, as it got a little confused at the ends.

This is a tiny cross which fits inside a small capital letter. Its arms are of different lengths in the original. The original colouring makes part of the pattern appears as a border, yet it is all one knot.

This pattern is repeated four times on the page. The top row is in the book, with the top left being the original colouring. It is nearly symmetrical apart from one part, mid top right. The bottom row shows my symmetrical version. To start with, I assume that this was a mistake, probably in some master design (which explains the same mistake happening four times!) However, the strings are in pairs again, and once I tried colouring them, to trace where they go as a pair, I found that the symmetrical pattern has two pairs of strings, while the original only has one. The change from the symmetrical pattern joins the two pairs into one. Perhaps this was the intention.

This design is slightly different because the knots run across the design rather running with it. The original colouring makes the knot a little hard to follow since they have coloured in between the knots, and at the edges, with the string colour. Also the under and overs are not clearly shown. Still, I think the right hand version explains what it happening.

To start with, this pattern looks as if it has pairs of strings, looking like the edges of a ribbon, but going under and over independently of each other, as in the previous patterns. A closer look shows this is not so. There are two strings (not a pair) which interleave through the whole pattern, shown in grey. The rest of the design is interleaved closed knots. Unlike most abstract patterns in the Book of Kells, there seems to have been an attempt to colour the pair of strings in blue, but this has been varied, presumably for effect.

This is another 'ribbon' pattern, with the edges independent of each other. There is only one ribbon, but I have changed the colouring half way through, to show how the pattern interleaves with itself. The pattern is in a filler in one of the decorated pages. The design is essentially a plait of four ribbons in the middle, which are joined at the ends with some extra knots for decorative effect.

This cross has four ribbons, as shown by the colours. The points are really supposed to meet in the centre! Unlike many ribbon patterns, each edge is outlined, rather than just where it goes under and over.

Although some of the knots above have double strands, they would not be drawn using the double strand method, since the strands are separated rather than touching. They might have used the "felt-tip pen method", although obviously using paint rather than felt-tips! There is a clue that this might be so. Usually the knots only have one colour. One knot has two colours, but colours whole blocks of the design at once, rather than following a single strand. The felt-tip pen method is very hard to follow a strand, but easy to colour a block at a time. One knot does try to follow a strand (mostly) but the original looks as if the darker colour has been added afterwards, since it is rubbed off in several places. Whatever method was used, I am sure that they tried out the design in rough first, and transcribed it carefully onto the finished page. A computer is much easier than vellum for correcting mistakes!

The most splendid knots in the Book of Kells are fantastically entangled creatures, snakes writhing among each other, a man with a beautifully knotted beard, a mermaid with a knot in her stomach. I cannot do justice to their artistic qualities, so have mostly left them well alone. You'll have to see the originals instead!

I have however done a mischievious moving picture of the top of one of the initials, N. The original picture is on the left, showing two men pulling each other's beard. I have allowed their hair and beard to grow and interlace, until, presumably, they can't stand it any more, and grab the other's beard. Then the hair shrinks back to the start.

Book of Kells - men pulling beards

Links to other websites

National Museum of Ireland in Kildare Street

Old Library of Trinity College, Dublin

Gallery of the Medieval Scriptorium showing images of the Book of Kells

Art Imagery showing images of the Book of Kells

Return to Knots index.