Source text available here http://www.urbansedlar.com/files/pilnyak/goly_god.txt

Word cloud

Sentence lengths

To get sentences and text fragments, we split the text at the following punctuation:

Repeating sentences

The sentences obtained by splitting the text were grouped and sorted by repetition count. Non-repeating sentences were rejected. Of what was left, 1-character sentences were also rejected.

Remaining sentences are sorted in the descending order of repetition:

Repeating parts of sentences (also considering commas)

The text was again split into sentences and fragments, this time also considering commas. There’s a lot of repetiton in dependent clauses, e.g.:

Obtained parts of sentences were again grouped and sorted by repetition count. All parts of sentences that don’t repeat were again rejected, as well as 1-character entities:

Below are repeating parts of sentences, sorted in descending order of repetition: