Thank you Emma, very interesting. Yes, the divergence between the frequency of vord-initial [cth] and relative infrequency of vord-initial [ckh] is striking indeed.
I'm glad you also linked to your blog post about the distribution of [k] vs. [t]. I am still digesting all the details of the arguments in this post, but as a basic summary I think I can state that the main point is to try to explain the much greater frequency of [k] than [t], which is particularly pronounced in Currier B. Out of the approximately 4000 more occurrences of [k] than [t], about 1000 of them are due to the much greater frequency of [lk] than of [lt]. About another 1000 of them are due to the strikingly greater frequency of [qoke] than of [qote]. On the other hand, at the start of a line that is not the first line of a paragraph, [t] is much more frequent than [k], although this only covers about 250 occurrences in total. Emma discussed many more details and ideas about possible explanations in her blog post, but those are the main points that I took away from a first reading of it. If I understand the final part of Emma's blog post correctly, she may be suggesting that in some of these environments [k] and [t] may be in a certain kind of what linguists would call a complementary distribution, where one occurs in certain environments and the other occurs in other environments. However, this is at most a partial tendency, and cannot be applied to all occurrences of [k] and [t].
Concerning [qoke] vs. [qote], I would return to Koen's and Marco's verbose cipher analysis again: According to Koen's best version, [qok] and [qot] may each be single units that may be entirely separate and distinct from [k], [t], [ok], and [ot]! In fact, just checking the statistics again now, I observe that in general [qok] occurs over 3100 times, while [qot] only occurs 1130 times! This covers yet another about 1000 "extra" occurrences of [k] compared with [t] (beyond the about 1000 already covered by [qoke] vs. [qote]). Perhaps [qok] is simply a letter/sound/phoneme that is much more frequent than the letter/sound/phoneme represented by [qot]? Just as one hypothetical example, this would make sense if [qok] represented "s" and [qot] represented "z".
[ok] and [ot] without preceding [q] do not have such a great frequency difference, only 2981 vs. 2736 by a simple count from voynichese.com. But then [k] and [t] without preceding [o] have a great frequency difference again, about 1750 more [k]. [lk] vs. [lt] explains about 1000 of these, but there remain about 750 to account for.
Emma, I think you may be interested to read and study You are not allowed to view links.
Register or
Login to view. of the relative frequency of different consonants in a wide variety of global languages in initial vs. non-initial position. While the title refers to Egyptian, this seems misleading as the article is a general study of universal cross-linguistic tendencies. The section titles include "the most frequent consonant", "labials frequent initially", "sonorants rare initially", "r more frequent than l", "stops compared by manner of articulation", "voiced and voiceless stops", "stops compared by place of articulation", "affricates", "voiceless stops in initial vs. non-initial position", "voiced stops in initial vs. non-initial position", "s and z", "voiced fricatives", "š and s", and "č and š". Linguistically-minded analysts of the Voynich ms such as yourself, Emma, may find much of interest in this general cross-linguistic study of consonant types and positional occurrences, I think.
Geoffrey