A Note on Cyrillic

I tend to stay away from seeking specific languages or scripts to fit the Voynich text, preferring any solution to emerge from our understanding of the text rather than being imposed upon it. But I’ve recently noticed a few interesting points about the Cyrillic script that I wanted to share.

Naturally I don’t claim to be an expert on Cyrillic, and am happy to be corrected where wrong. I also do not want to suggest that the Voynich script is based on the Cyrillic script, but rather that there may be interesting parallels we can learn from. It may be that other researchers have already raised the same points.

The Cyrillic script was invented over a thousand years ago and derives mostly, but not exclusively, from the Greek script. It also has a strong connection to the Glagolitic script, though the nature of that connection is uncertain and debated.

The Cyrillic script has undergone a number of formal and informal reforms over the centuries, most notably the Civil Script reform by Peter I in 1708. All the points below refer to the pre-1708 Cyrillic script, though it is difficult to draw comparisons from when the Voynich text was made in the early 1400s due to informal variation.

Numeral System

Along with inheriting the bulk of Greek letterforms Cyrillic also inherited the Greek numeral system. This system assigned each character a specific value and used an “additive” method. It was one of the common numeral systems (along with Roman numerals) in use in the West before the introduction of the Indian positional system.

To illustrate how it worked for the Cyrillic script: the first nine glyphs were assigned the values 1-9 (in steps of one), the next nine 10-90 (in steps of ten), and the next nine 100-900 (in steps of one hundred). Thus Д = 4, Ѯ = 60, and С = 200, with СѮД = 264.

There are two main lessons here for the Voynich text.

The first is that any numbers written in the text may be indistinguishable from words. They may have odd structures, and generally be short, but given that we do not know how to read any words we would struggle to properly identify them. Thus numbers may exist in the text which we’re unaware of.

The next lesson is more important. In a few places, specifically some of the charts of diagrams, we see Voynich text used in a way which we might suggest has some numerical value. The best example of this is the second ring of the chart on f57v, which has a pattern of individuals glyphs repeated (almost exactly) four times. It has been suggest that each repetition is a quadrant of the circle with the glyphs representing degree/divisions of that quadrant.

Whether this is true or not it’s most important to note that the repeated sequence has glyphs not usually found in the main text. Around a third (maybe 6 out of 17 glyphs, but opinions will differ) are found either only here or in a few places throughout the manuscript. So these glyphs are not used to write the Voynich language.

This has a direct parallel in Cyrillic numerals, as it included several glyphs from Greek which weren’t used for writing language. One such example is Ѯ, given above, used only for the value 60 (and writing Greek loanwords). This provides a hint that the Voynich script too could be descended from an established script and using only a selection/subset of the possible glyphs.

Iotified Vowels

Another interesting feature of the medieval Cyrillic script is that of iotified vowels. A number of vowel glyphs were ligatured with the glyph for /i/ to create symbols which stood for the combined sound /j/ + vowel. Thus І + А = Ꙗ /ja/.

The table below (copied from Wikipedia here) shows the combinations which were possible:

Normal Iotified
Name Shape Sound Name Shape Sound
A А /a/ Iotified A /ja/
E Е /e/ Iotified E Ѥ /je/
Uk ОУ /u/ Iotified Uk Ю /ju/
Little Yus Ѧ /ẽ/ Iotified Little Yus Ѩ /jẽ/
Big Yus Ѫ /õ/ Iotified Big Yus Ѭ /jõ/

Both the form and nature of these ligatures is striking when compared with the Voynich bench glyphs [ch, sh]. We have seen that there is a relationship between benches and the length of e sequences, which is suggestive that bench glyphs are a ligature with an [e] glyph as their second part.

It has also been observed that [c] is also found in combination with the glyphs [o, y] to create [co, cy], and even [ckhh, cthh]. All are rare but indisputable, giving strength to the possibility that [ch, sh] are ligatures and therefore decomposable into separate glyphs.

The Cyrillic model allows us to understand how such a ligature might work and how the resulting glyph could as both as a consonant and a vowel. Indeed, there is a further function of iotified vowels in Cyrillic which has interesting lessons for the Voynich script.


One of the uses of iotified vowels in medieval Cyrillic was to indicate that a preceding consonant is ‘soft’. In Slavic languages this originally meant that the consonant is palatized, as it does in Russian today, but expression of ‘soft’ can differ by language. The important point to note is that a single consonant glyph can have two related realizations dependent on the following glyph and that this can be phonemic.

The potential for the Voynich script is obvious. Bench characters [ch, sh] often occur after certain glyphs, creating a very common sequence. Yet the entropy of the text is too low. The possibility that two glyphs (or a glyph and a ligature) can combine to create a different sound will help alleviate both problems.

What this means in practice is that [k, kch, ksh] might be three different sounds, and this would be true for all (or most?) the glyph which are commonly followed by benches. As up to eight glyphs [k, t, f, p, d s, r, l] might be involved in such a process, the growth in the number of sounds is quite substantial. Even were this process only happening for some of those, it would still make a difference.

We can see the potential combinations and their frequencies in the table below.

Glyphs Count %age Glyphs Count &age Glyphs Count %age
[k] 8640 87% [kch] 1093 11% [ksh] 216 2%
[t] 4769 80% [tch] 992 17% [tsh] 180 3%
[f] 210 50% [fch] 193 45% [fsh] 20 5%
[p] 575 41% [pch] 752 54% [psh] 72 5%
[d] 12400 96% [dch] 341 3% [dsh] 164 1%
[s] 2768 96% [sch] 86 3% [ssh] 25 1%
[r] 7238 97% [rch] 137 2% [rsh] 46 1%
[l] 9473 90% [lch] 729 7% [lsh] 283 3%

The frequencies are quite interesting, as while some combinations would be very rare and only one hundredth of the non-combined glyph, others would be very common. The question is whether such relationships between any related series of sounds (such as palatized and non-palatized) is realistic.

Sadly I don’t have the right statistics at hand to amply evidence one way or the other. I can say that for voicing, in English /t/ is only about a third more common than /d/, whereas in Finnish /t/ is about ten times more common than /d/.

It remains an interesting idea (and not original to me) which deserves more investigation. Indeed, I hope this note on Cyrillic is interesting overall for further discussion.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s