Experiments with interpretation of the labels. The label otol.

[Searcher]

 For the simplest example, one may substitute not single letters but, say, bigrams. Take the word "date", if "da" maps to "po" and "te" maps to "lo", then "date" maps to "polo", and there's no 1:1 mapping between single characters (instead, it exists for bigrams). The decipherer would in vain consider the two "o" characters to be the same.

Thanks, Anton! I missed this method for the case. I am sure to try it. And I forgot to mention abjad, as well. I just want to develop a thought about bigrams. Indeed, bigrams can stand not only for another bigram, some bigrams may stand for some combination of three letters. I think, it is possible.
Observations show it is doubtful that the main text is enciphered with simple substitution cipher, although we don't know whether the labels and the text enciphered with the same system.

For a more complex example, there is my idea expressed in another thread about prefixes as relational operators, like "otol" meaning something being made over "tol", for example otol = "with tol", while qotol = "without tol". (There are many labels starting with "o", but very few starting with "qo"). This idea still remains unexplored.

The idea is probably good, but in this case the label words seem to be too short (although, of course, it depends on language). Also there is a question: What with such words as ol, or, ot, etc?