The first part of the project: https://doi.org/10.5281/zenodo.18247099 The third part of the project: https://doi.org/10.5281/zenodo.18247371 This study presents an operational interpretation of the Voynich Manuscript language based on the analysis of the Operational Token Lexicon and the formal rules governing word-form construction. It is demonstrated that the manuscript text is organized neither as a natural language nor as a substitution cipher, but as a formal system of procedural tokens operating under a strict grammar of actions, states, and transitions. The concept of the operational token is introduced as the minimal meaningful unit encoding not an object but an operation, mode, boundary, or result. The mechanisms of word-form formation are described through the composition of atomic EVA units, modal and durational markers, as well as fixation and boundary morphemes. Semantic modification is shown to occur not through lexical expansion, but through allomorphy, positional variants, and combinatorial use of a stable and limited set of elements. Special attention is given to the formation of composites and series, in which token order reflects sequences of operations rather than syntactic relations typical of natural languages. Fixation, permission, and terminal elements play a system-forming role, ensuring formal closure of stages and procedural reversibility. Repetition, cycling, and intensification are demonstrated to be strictly regulated through specialized markers and never occur freely. A neutral glyph code layer, independent of visual calligraphy, is proposed, enabling reversible transliteration between EVA sequences and abstract glyph units. This separation of graphical form and operational semantics allows the text to be treated as a formally describable system of states. The token dictionary is therefore interpreted as a functional registry of operators rather than a lexicon of objects or concepts. As a result, a coherent model emerges in which the manuscript is read as a procedural protocol governed by a unified grammar of tokens, allomorphs, and composites. This approach provides a foundation for corpus-level statistical verification, cross-comparison with the manuscript’s graphical sections, and the further reconstruction of reproducible algorithms encoded within the text.
Nadezhda-Victoria Vinyukova Vinyukova (Wed,) studied this question.