ساخت نمود کامل در فارسی برپایۀ صرف توزیعی

رویکرد ناواژه‌گرای صرف توزیعی که به کوشش هله و مرنتز (1993, 1994) در اوایل دهۀ نود میلادی پی‌ریزی شد، فاقد واژگان زایای پیشانحوی برای ذخیرۀ واحدهای چندجزئی است و همۀ عناصر صرفی‌نحوی برای درج در هستۀ فرافکن‌های نقش‌نمای مستقل رقابت می‌کنند. مقالۀ حاضر برپایۀ این رویکرد صوری به دستور و بی‌آن‌که گام به ساحت معنا بگذارد، به بررسی ساخت‌هایی می‌پردازد که دارای عنصر دستوری نمود کامل‌اند. این عنصر که در سنت دستور زایشی، وند «en-» خوانده می‌شود، یکی از پرکاربردترین هسته‌های نقش‌نمای زبان فارسی است و فارغ از زمان و وجه، در افعال حال و گذشته، و وجوه اخباری و التزامی حضور می‌یابد و علاوه‌بر بندهای خودایستا و ناخودایستا (یعنی بندهای با یا بدون زمان و شخص‌وشمار)، حتی در سازه‌های کوچک‌تری همچون واژه نیز یافت می‌شود. این پژوهش در دو بخش پیاپی ماهیت این وند و ساختار سازه‌هایی را که میزبان آن هستند، بررسی و تحلیل می‌کند. در بخش نخست می‌کوشیم تا مستقل از نظریۀ صرف توزیعی و تنها با تکیه‌بر شواهد زبانی و استدلال‌های صرفی‌نحوی، نشان دهیم که صورت انتزاعی «de-» گونۀ زیربنایی وند نمود کامل است که در سطح آوایی، با یکی از دو واژگونۀ «‌ده» یا «‌ته» بازنمود می‌یابد و در زبان فارسی، هرگز با وجه التزامی در زمان گذشته همراه نمی‌شود. پیامد این سخن آن‌که اولاً صورت موسوم به صفت مفعولی، متشکل از نشانگر زمان گذشته و پسوند «ه» نیست، بلکه واژگونگی بافتی که نقش مهمی در این پژوهش ایفا می‌کند، شکل خاصی از یک ریشه را در برخی از بافت‌ها، و از جمله در مجاورت وند نمود کامل، عینیت می‌بخشد. ثانیاً اگرچه وند نمود کامل توزیع گسترده‌ای در زبان فارسی دارد، اما این زبان فاقد بندهایی با ساختار گذشتۀ کامل التزامی است. در بخش دوم مقاله، یافته‌های این تحلیل را در چارچوب بنیادهای نظری صرف توزیعی بررسی می‌کنیم و با تمرکز بر ساخت حال کامل اخباری (ماضی نقلی)، یافته‌های خود را با اندک تفاوت‌هایی در فرایندهای پسانحوی، به افعال گذشتۀ کامل اخباری و حال کامل التزامی و صورت سببی این ساخت‌ها بسط می‌دهیم. در این میان، تصریح می‌کنیم که وند نمود کامل در هستۀ گروه نقش‌نمای حرف درج می‌شود که از سوی فرافکن نمود مجوزدهی شده است.

Perfect Construction in Persian within the Framework of Distributed Morphology

Within the nonlexicalist framework of Distributed Morphology (DM), proposed in the early 1990’s by Halle and Marantz, there is no presyntactic generative lexicon in which words are derived and all morphosyntactic elements are subject to competition for Vocabulary Item insertion in the head of functional projections. Based on this formal approach to grammar, the present study investigates the perfect construction and past participle ending in Persian, without resorting to their semantic properties. The past participle ending usually called “en” since the earliest generative work, is one of the most pervasive functional elements in Persian. This suffix not only combines with present and past verbs in indicative and subjunctive moods, but also it appears in both finite and nonfinite clauses (i.e. clauses with or without Tense and Agr) and is even present in smaller constituents such as words. The study explores the nature of participle inflection and perfect constructions in two successive parts. From an empirical perspective and beyond the purely theoretical assumptions, in the first part arguments drawn from language data are used to show that “De” realized as “de” or “te” at PF, is the underlying form of perfect suffix and it never appears with subjunctive verbs in past tense. This analysis has two important consequences. First, the socalled past participle in not derived from the past stem of verbs. But, contextual allomorphy which plays a central role in this research, realizes different forms of a morpheme determined by some items, including perfect suffix, in its context. Second, past perfect subjunctive clauses are not found in Persian, although “en” suffix has a widespread distribution in this language. The second part discusses some views on syntactic word formation of present perfect verbs, by focusing on the assumptions being made within the framework of DM. Such a strategy, it is suggested, can be generalized to all cases of perfect constructions with some subtle differences in postsyntactic operations. Meanwhile, it will be argued that perfect suffix is inserted in the head of Particle Phrase licensed by Aspect Phrase. Keywords: Perfect Aspect, Tense, Past Participle, Allomorphy, Late Insertion IntroductionDistributed Morphology (DM) introduced in Halle and Marantz (1993, 1994) in the early 1990s is a grammatical model that has emerged within the framework of Principles and Parameters. DM which represents a set of hypotheses about the interaction among components of grammar, including Morphology, syntax and phonology claims that the complex structure of a word is created in the same way as is the complex structure of a phrase or sentence. It is important to say that Distributed Morphology is a framework within the Minimalist Program (MP) which rejects the Lexicalist hypothesis and the notion of a generative lexicon (Siddiqi 2009). In this linguistic model, there is only one generative component of the grammar (the syntax) whereas in Lexicalist Minimalism, there are two (the syntax and the lexicon). The four fundamental differences between DM and Lexicalist Minimalism to be mentioned here are categorization, lateinsertion, morphosyntactic decomposition, and underspecification. These key notions are very important for my investigation in the Persian past participle inflection and perfect constructions. Moreover, this article proposes an analysis of  root allomorphy (e.g., ketāb “book”/ kotob “books”) within the framework of DM that showcases the economy constraint minimize exponence. It also accomplishes two other things: First, following Siddiqi (2009), it proposes some of the revisions to the framework of DM in related to the root allomorphy and readjustment rules. Second, it provides an analysis of verbal allomorphy in Persian. It should be emphasized that when roots appear in the derivation, they do not have grammatical category. This is the first difference between DM and MP which is discussed in the next part. Materials and MethodsAs I told above, within the nonlexicalist framework of Distributed Morphology (DM), proposed in the early 1990’s by Halle and Marantz, there is no presyntactic generative lexicon in which words are derived and all morphosyntactic elements are subject to competition for Vocabulary Item insertion in the head of functional projections. Based on this formal approach to grammar, the present study investigates the perfect construction and past participle ending in Persian, without resorting to their semantic properties. The theoretical method of the investigation is mostly based on the following characteristics:(a). CategorizationAn important property of Roots is that they have no grammatical category inherently. This assumption derives from earlier work on derivational morphology. According to the categoryfree theory of Roots, traditional lexical categories like ‘noun’ or ‘verb’ or ‘adjective’ are convenient shorthand labels that refer to syntactic structures in which a Root combines with a categorydefining functional head such as little n or v or a. This is what happens for that a root like √bin (see). During the derivation, it may merge with a little noun head to generate the noun bineš (vision), or it can absorb a little adjective head for producing binande (viewer) and so on.(b). Late insertionTerminologically, theories that allow for morphemes to receive phonological form after they are combined in the syntactic component are said to have late insertion process. In DM, unlike in GB and its Lexicalist derivatives, rather than manipulating fully formed words, the syntax only manipulates abstract formal features to generate syntactic structures. These morphosyntactic features (such as [plural] and [past]) are selected from a fixed list of abstract features (or feature bundles) rather than being selected from the output of a generative lexicon. The late insertion hypothesis (Halle Marantz 1994) holds that the phonology which represents the morphological features manipulated by the syntax is provided at PF rather than being present throughout the derivation. At spell out, syntactic terminals in DM are entirely comprised of interpretable features (including roots). Only once all syntactic processes are finished with the structure is phonological content added. This phonology is provided by a component of the grammar called the Vocabulary. The Vocabulary is a static list of items whose function in the grammar is to provide phonology to realize the interpretable features contained in the terminal nodes of a derivation so that that derivation can be pronounced. Individual items within this list are called Vocabulary Items (or VIs for short).(c). Morphosyntactic decompositionOne of the strengths of the Distributed Morphology framework is the parallel between syntactic structure and morphological structure. Since the grammar of DM manipulates only syntactic features, the complex structure of a word is created in the same way as is the complex structure of a sentence. Spelling out a complex constituent of the syntax as a “phrase” or a “word” depends on the nature of VIs in the structure. In this model not only the verb mibinam (I see) is produced in syntax, but also the noun bineš (vision) is derived in the same component.(d). UnderspecificationDistributed Morphology uses underspecification in the insertion of Vocabulary Items into a terminal node of the syntax. The insertion of a VI is governed by the subset principle which allows for a VI with certain specifications to be inserted into any node that satisfies those specifications, regardless of whether or not it exceeds those specifications. This characteristic of DM is very important for my proposal regarding to past inflection in Persian. Discussion of Results and ConclusionsThe above four characteristics distinguish DM from MP. But before going through them, I should briefly discuss what is meant by Root and Root Allomorphy in DM. These two concepts are continuously referred through this article. In addition to functional morphemes, the grammar contains morphemes that are called Roots.