f and p appear predominantly in the first lines of paragraphs

[MarcoP]

Indeed Marco and Jkp. I think that a visual look at the layout of these glyphs tends to make you think that they can't be letters, but must instead by a sense marker of some sort - a pilcrow or system of per cola et commata.

Thank you for mentioning "per cola et commata" - an expression I hadn't read before.
As I wrote, I don't think that the observed phenomenon necessarily points to the glyphs not being letters: they could be special letters that (due to the systematic organization of the paragraphs) tend to appear in specific positions. I find it harder to think of pilcrows clustering in this way than of some special characters possibly connected with "foreign" languages.

I will try too look for something similar in actual texts: the clustering of any kind of graphic symbol in specific paragraph positions.