The Challenge of Analyzing a Dynamic Text

[JoJo_Jost]

Vocabulary evolves in strict directional chains with intermediate forms (<chol> -> <cheol> -> <cheo> -> <chey> -> <chedy>), and words from early sections persist in late sections but never the reverse. [...]
In Ortloff von Baierland the word "und" may be extremely frequent, but it does not generate dozens of single-edit variants clustering on the same pages.

"Thanks for your reply. Actually, I wasn't concerned with your entire thesis, just that one specific point.

But just for the record, words like this exist in Bavarian texts too. In the Ortloff von Baierland:

‘vnd’ (and) (1,942×) has 20 neighbours at edit distance 1 (vn, vnnd, vns, vno, und, vnde...) and even in one sentence:

Example: Ist aber ter harnn weiß vnd dunn vnnd das kleyn kornlein als d' sand an te potem seind (written with u: und dunn unnd)

sol (443×) → solt (216×) → soltu (43×) (shall). These are not ‘stages of evolution’ — they are simply inflected forms of the same verb: “soll”, ‘sollte/sollst’, ‘sollst du’.

‘kalte’ (cold)  has 7 neighbours: kelte, kalt, kalter, kaltes, kalten, kaltem, kalts — pure inflectional forms.

However, spatial clustering is one of the points where you are actually describing a phenomenon that is less pronounced in normal texts. Whether this excludes a language depends on the encoding – an absorption encoding that merges articles and prepositions with the following word could, of course, create clusters of similar-looking words in the same line, simply because the same grammatical context produces the same prefixes. But that would still have to be proven, of course