The purpose of this dissertation is to build a description model of Word Formation System (WFS) of Modern Persian by means of formalized morphology. A study of the word formation mechanisms is a necessary step in a holistic understanding of a language structure. The complexity of the word formation system is caused by its close connections to the other levels of a language, mobility and lack of a clear distinction between its potency and actual implementation, between derivation and inflection, as well as by the diversity and communicative significance of its functions. Formalized morphological approach includes a number of methodological guidelines: completeness, simplicity, uniformity and formalization. Completeness of formalized morphological description is achieved by the involvement in the initial stage of all linguistic phenomena of the studied area without exception that are recorded in an authoritative source or group of sources. Simplicity of the morphological description is provided by selection of the economical metalanguage, as well as commitment to transparency of the analysis procedure available for being consistently repeated, checked or verified. It is important to remember that the metalanguage of the morphological description should not be more complicated than the described linguistic object itself. Formalization of the study is assured by neat use of the laconic symbol system of the morphological categories (for example, N ? root morpheme of a noun, Aj - root morpheme of an adjective, V? present tense root morpheme of a verb, Pron - root morpheme of a pronoun; pl - plural, etc.). The principle of uniformity is aimed to avoid mechanical and eclectic mix of the mutually incommensurable tools and techniques of the research worked out on other methodological grounds. The notions of metalanguage, inventory of word-formative models, morphological classes, productivity and valence, initiale, mediale and finale, hapax, segments of unique and recurrent word-formative formulas form the core of the theoretical structure of the formalized morphological approach. In this research metalanguage is regarded as universal formalized semiotic code (or "second-level" language) used in the morphological description of the modern Persian language ("first-level language") which allows further classification of the obtained material, establishing various kinds of relations between its components and making calculations in order to reveal statistical regularity in WFS. A set of word-formative formulas acquired by performing the morphological analysis on all dictionary lexemes was coined by the term word-formative inventory a study of which make it possible to formulate a number of linguistic laws in the Persian WFS. Word formation inventory is considered to be a kind of a model of final states (morphological grammar of word formation) that enables presenting a morphological description of the first level. This research has been conducted on the Dictionary of the Persian Language and Literature compiled by Mohammad Qharib (29162 words). Inventory analysis has revealed quantitative correlations of morphological classes in the Persian language, as well as of direct and indirect loanwords from other languages. According to the performed research the morphological classes in the examined lexicon are distributed as follows: 55.44% of the total number of words account for nouns, 23.6% ? for adjectives and 12.91%, 3.72%, 2.26% and 1.3% account respectively for verbs, adverbs, nouns-adjectives and adjectives-adverbs. The quantitative characteristics of the morphological classes' correlations are estimated separately for simple and compound words. The total number of loanwords (direct and indirect) in the examined lexicon is 52.4%. The largest share of lexical borrowings is from Arabic (46.4% of the examined lexicon); 2.8% of loanwords account for French, 1.04% ? for Turkish, 0.2% ? for English. The largest number of indirect loanwords belongs to the Greek words that were borrowed into Persian via Arabic (0.23%). It was shown that word-formative inventory consists of seven main morphological classes (noun, adjective, verb, adverb, numeral, pronoun, verboid) and six auxiliary morphological classes (preposition, article, conjunction, interjection, interrogative words, and particle). Some words of the lexicon which under certain conditions change their grammatical role and convert from one part of speech to the other were ascribed to the conversive morphological classes (most of such words were borrowed from Arabic); overall we distinguished 17 conversive morphological classes in word-formative inventory. The morphological class of nouns is formed by 407 word-formative formulas, adjectives - by 241, adverbs - 71, verbs - 15. The total number of word-formative formulas in the conversive morphological classes is 80. Thus the Persian Native World Formation System as a whole includes 814 word-formative models. The composition of the word-formative units of the inventory in percentage is distributed as follows: 50% of the inventory account for nouns, 29.61% ? for adjectives, 8.73% ? for adverbs, 1.84% ? for verbs and 9.82% ? for conversive morphological classes. The WFS of the specific morphological tools in Modern Persian was comprehended as a dynamic analytical model which has a multi-faceted structure with distinct internal regularities and relations. The components of this model are organized on a basis of nature of initiales and finales - whether they are formed by affixes or the morphological classes (division of the system into four morphological subsystems), productivity or non-productivity of the word-formative models (division of inventory into segments of the unique and recurrent word-formative formulas; differentiation of three Bradford zones in the recurrent segment of word-formative formulas), as well as on a basis of attributing a word-formative model to a certain part of speech (distribution of word-formative formulas in the morphological classes). Shown tendencies in the Persian WFS concern both its inner and outer structure (regularities of the initiales' and finales' organization in word-formative formulas; Estoup-Zipf law 1, 2, 3).
