Minims in mediaeval orthography (1)

While thinking about  the sequences of ‘i’ letters in the Voynich manuscript, I came across this interesting account of minims in Latin script from a Harvard website:


Medieval scribes used minims to form letters. A single minim looks like this:


Several minims can make up a single letter, or even a group of letters. In particular, minims are usually used for the following letters:

One minim: “i”, “j”
Two minims: “n”, “u”, “v”
Three minims: “m”, “w”

It is frequently difficult to know what letter or letters a group of minims represents unless you can determine the entire word from context. Look at the word below, and see if you can figure out what it is. Mouse over the image to learn the correct answer.


The leading “a” helps determine what the string of minims in the middle of the word must be. Now, try a harder one (hint: this word is Latin):


Although it looks at first glance like this word is composed entirely of minims, the two center characters are “ll.” As you can see, scribes did not follow our practice of putting uniform spaces between characters, so sometimes minims from separate characters run together, or minims forming the same character will be separate.


In the Voynich manuscript we see many examples of these minims, and in my Feb 2014 paper I suggested that they might be borrowed from Latin and have a similar use and meaning. The account above reminds us that a sequence of two minims might represent ‘n’ or ‘u’ and three might represent ‘w’ or ‘m’.

When reading the Voynich manuscript, in other words, we need to be aware of possible multiple meanings for the same signs , and we need to accept that this was not unique in mediaeval practice.

I am starting to wonder whether the sequence I translated in my paper as KNTAIRN for Centaury, might not in fact be KNTAURN, the U being made up of a single minim plus a second minim with an upward curve signifying the R. That would have implications for other analyses later in the discussion, but it would in fact be a more satisfactory reading of this word, fitting with the Greek ‘aur’ of Centaur.

But time will tell.


  1. Tom O'Neil

    I agree with Stephan Bax in that minims maybe the key to the translation of the Voynich Manuscript.

  2. Derek Vogt

    Is the problem of reading minims the reason why [i] sprouted a dot?

  3. Robert Hicks

    With regard to the transliteration of the VMS minim characters (in EVA, the in, iin, iin, ir, iir, iiir, im, iim, iiim structures):

    The ligatures between the minims MIGHT be significant, but it is far better to count them out of any transcription than count them in. The reason why pertains to loss vs gain. You can LOSE information by ignoring ligatures, but this is a problem you can live with (unless the loss reaches a critical point). Accidentally ADDING information, however, obscures information. To illustrate, let’s take the sentence:

    The quick brown fox jumps over the lazy dog.

    Imagine this is hand-written.
    A hypothetical linguist is trying to transliterate this. He notices that the letters w and m look the same (just different rotations) and that u/n, q/b/d/p, a/e, f/t all might be similarly related. He transcribes thus:

    The quick qrowu fox juwqs over fhe lezy qog.

    Clearly, he has lost information.

    Another linguist notes that the script is composed of vertical, horizontal, diagonal lines and little loops. He sees that no two letters look identical, due to the scribe having made loops and lines of slightly different length. He sees ligatures between some of the letters. He sees light strokes and dark strokes. All of this could be information. He believes all may all be significant, and transcribes thus:

    The quick brown [email protected] j&mps %ve$ t£^ lazy d#g.

    After a while, the first linguist settles on a limited character set which loses partial information. The second, however, has an ever-expanding character set. Each ‘new’ letter needs a new transliteration code, and the end result is a transcription is which all information is utterly diluted.

    • Stephen Bax

      Thanks. We could also look at it this way: if the V scribes wrote ‘ui’ in minims in one word, then ‘in’ in minims in the next, joining them different ways, and we fail to notice and transcribe both as iii, we will get a wrong count of signs overall, and may never reach a proper decoding!

  4. Interrobang

    Funnily enough, back in 2002, I wrote a little essay in my Slashdot journal (now basically defunct) about how someone ought to look at minims vis-a-vis Voynich, and how the plants in Voynich look like the stylised plants from Arabic herbals, and has anybody looked into that yet…and here you are with a background in Arabic and a grounding in the medieval Middle East, talking about minims, and working on the MSS from a linguistics point of view. I can’t tell you how happy your little video made me, and I will read your paper.

    I’m not anybody in particular, just someone with a Master’s in what amounts to applied rhetoric, and an interest in the Middle Ages, medieval calligraphy, and Hebrew.

    Thank you, thank you, thank you!

  5. Dyana

    Could the picture which you think has “cotton’ written near it be some type of cotton thistle?

    • Stephen Bax

      Potentially, but we then also need to identify the matching word in the text. That’s the fun bit!

Leave Your Comment

Your email will not be published or shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>