Single Leg Gallows

[Jorge_Stolfi]

For me, the most interesting thing about this is that it is not so much a rule, but that it is 'just' a very strong trend. There are too many exceptions. [...] 

Over the years, I have developed an opinion, which would explain a lot of the unusual things we are seeing. This is that the text has been encoded somehow, with rules that allow for some flexibility or some choice by the encoder.

I agree with the last statement, but I would change "encoded" to simply "written".  The flexibility with regard to one-leg or two-leg gallows seems to be similar to that which other Medieval scribes would have when writing Latin.  Namely, they would be free to use abbreviations or write words in full, to use ornate letter forms on the first line and first word of a paragraph, draw or omit serifs, etc.  

A while ago I looked at the frequency of the glyphs immediately following each gallows glyph, and concluded that t and k were very similar, and p and f were very similar, but the two pairs were quite different.  On the other hand, the frequencies of the immediately preceding glyphs were fairly similar for all four gallows.  So indeed it would indeed seem that p and f are shorthands for combinations like te and ke; but I have no idea which is which.

These days I have been looking at the differences between the "hooked" and "unhooked" variants of p and f.  I am still collecting the data, but I suspect that the hook (unlike the left loop that distinguishes p from f) is a purely aesthetic feature, which may be added when there is space for it (such as  at the start of a word) and omitted otherwise.  For one thing, there seems to be a continuous spectrum of shapes between the two variants.  If p is indeed an abbreviation for te or ke, the hook may be just an optional visual reminder of the e.  

By the way, I believe that the Scribe who put the text onto vellum was copying from a draft written by the actual Author on paper.  In that case, it is possible that the line breaks of running text would fall on different places in the two versions.  That could explain the occurrence of p and f in the middle of paragraphs: namely, in the draft those gallows were on parag-leading lines, but Scribe did not notice the parag breaks and wrote those lines as continuations of the previous parags.  Because I also believe that the Scribe knew the alphabet, but did not understand a single word of the language.