Constraints on Structure and Derivation in Syntax, Phonology and Morphology

| 7 →

Foreword

The present volume is a collection of twelve papers centred around the notion of constraints. The works gathered here address, in particular, the problem of constraints imposed on structure and derivation in syntax, phonology and morphology.

Constraints have played a significant role in syntactic theorising ever since the seminal work by Ross (1967). Although widely adopted and recognised, constraints have not been extensively studied in the literature, with a few exceptions, including Müller and Sternefeld (2000) and Graf (2013). In the literature two highly disputed issues regarding constraints concern the relation between constraints and operations, and the explanatory power of constraints. Recently in the Minimalist Program the question whether constraints are superior to operations or the other way round has been addressed by two rival camps – derivationalists (Epstein et al. 1998; Epstein and Seely 2002; 2006) and representationalists (Brody 1995; 2000; 2002). The former argue for the supremacy of operations over constraints, whereas the latter make representations the focal point of the theory and impose constraints on their shapes. Besides the Minimalist Program, constraints have figured prominently in Head Driven Phrase Structure Grammar, Lexical Functional Grammar and the Optimality Theory (Graf 2013).

The typology of constraints proposed by Müller and Sternefeld (2000) is based on five types of constraints, namely: representational, derivational, global, translocal, and transderivational. Representational constraints apply to single phrase structures, locally bounded or unbounded, whereas derivational constraints affect locally bounded derivations. Global constraints operate in unbounded syntactic derivations. Translocal constraints hold of collections of trees, while transderivational constraints are applicable to collections of derivations. Although the distinctions between constraints that Müller and Sternefeld (2000) put forward are clear cut, it is not an easy task to determine which class a specific constraint belongs to. Locality constraints, including islands, for example, are regarded by Müller and Sternefeld (2000) as derivational, whereas the Empty Category Principle is viewed as an instance of a representational constraint. In turn, the Projection Principle is classed by Müller and Sternefeld (2000) as a global constraint, whereas the Avoid Pronoun Principle is treated as a translocal constraint. Finally, the Shortest Movement Principle of Chomsky (1995) is taken to belong to transderivational constraints. Graf (2013) observes that the borderlines between the particular constraint types are blurry. It is often the case that one type of constraint can be subsumed under another. This happens, for instance, in the case of the Merge-over-Move ← 7 | 8 → preference of Chomsky (1995; 2000), which may be understood as either global or transderivational.

Part I of the book, called “Constraints on structure and derivation in syntax” comprises seven chapters, each of which focuses on some constraint-related theoretical problem. The constraint explicitly analysed in chapter 4 by Aleksandra Gogłoza is the Person Case Constraint, which seems to be a representational constraint. Other papers make less explicit references to constraints. In chapter 1 the constraints imposed by the translator’s native grammar and the function of the text are taken into account in analysing the syntax of two translations of the Psalter. In chapter 2 the focus is laid on the constraints on parsing two negative polarity items within the confines of Dynamic Syntax. Chapter 3 scrutinises the constraints applicable to one type of deverbal nominalisation. Chapter 5 examines the constraints on the event and argument structure of psychological verbs. Finally, chapter 6 investigates the constraints underlying structural alternations available for the load class verbs.

In chapter 1, Magdalena Charzyńska-Wójcik analyses the treatment of syntactic patterns in two medieval renditions of the Psalter in order to determine to what extent the translations have replicated the syntax of the Latin original. A special focus is laid on the similarities and the differences between the two translations of the Psalter – the English one by Richard Rolle and the Polish one by Walanty Wróbel – and the original as regards pronominal subject omission in finite clauses and the treatment of the copula. On the basis of a careful data analysis, the author convincingly argues that syntactic convergence did not serve as a highly rated principle in either text. It is emphasised that neither author rendered the two syntactic phenomena scrutinised in the chapter in a systematic way. Whereas Rolle retained the covert copulas of the Latin text, he used overt pronominal subjects, instead of the covert ones, present in the original. Moreover, Wróbel, who was free to replicate the syntax of the original verbatim on account of the syntactic similarities between Polish and Latin, did not stick to the syntax of the original, but remodelled the (c)overt contexts in either direction. The overall conclusion is that the syntactic patterns of the original are treated in the two translations in an equally unsystematic way.

In chapter 2, Nadežda Christopher examines the complementary distribution between two types of Russian negative polarity items introduced by means of ni- and by means of libo-. Ni-words occur wherever predicate negation is involved, while libo-items appear in all non-veridical contexts, with the exception of predicate negation. Following Pereltsvaig (2006), Christopher calls this distribution pattern the “bagel problem.” The author shows that although the analysis proposed by Pereltsvaig (2006) within the framework of Distributed Morphology accounts for ← 8 | 9 → the majority of cases, it turns out to be problematic whenever the ni-word precedes its licensor, the marker of sentential negation, ne. The paper offers an analysis of the “bagel problem” rooted in the framework of Dynamic Syntax (Cann et al. 2005). The model is based on the dynamics of the left-to-right word-by-word parsing of language within context. It is proposed in the chapter that the two types of negative polarity items in Russian are parsed differently depending on the context. Ni-words are sensitive to the presence of the negative polarity decoration at the root of a clause, whereas the parsing of libo-items is aborted once a negative polarity decoration appears on the root node.

In chapter 3, Judit Farkas and Gábor Alberti focus on a special class of deverbal nominalisations in Hungarian, formed by means of the suffix hatnék. This type of nominalisation shows both verbal and nominal properties, which supports its treatment as a hybrid category. Among the verbal characteristics of HATNÉK-nominalisations, the authors list the possibility of having an accusative case marked argument, the separability of verbal modifiers, adverbial modification, and an internal information structure. The nominal properties of the nominalisation scrutinised in the chapter cover the ability to host a possessive argument, case marking, adjectival modification, as well as definiteness and referentiality. Farkas and Alberti propose a syntactic structure of HATNÉK-nominalisations, rooted in the cartographic model of Giusti (1996). Following Grohmann’s (2003) theory of Prolific Domains, they put forward a tripartite structure for HATNÉK-nominalisations consisting of a thematic domain, an agreement domain and a discourse domain.

In chapter 4, Aleksandra Gogłoza investigates Person Case Constraint (PCC) effects in English, Icelandic and Polish. These effects can be found in English in structures with the expletive there which can have an existential or a list reading. In Icelandic, they are attested in dative subject constructions, and in Polish they surface in copular clauses with the pronominal copula to. The author proposes a unified analysis of these effects in the three languages scrutinised, by relying on Richards’ (2008) account in the case of English and Icelandic and by modifying the analysis of Bondaruk (2012) in the case of Polish. Adopting Richards’ (2008) idea that the defective Goal in Icelandic and English is specified as [3person]_Case, Gogłoza proposes that in Polish copular clauses the pronominal copula to is a defective probe and thus carries features such as [uPerson]_Case. She suggests that the verbal copula być ‘to be’, optionally co-occurring with the pronominal copula to, has a full set of φ-features, but it lacks case. This allows her to account for: the impossibility of 1^st and 2^nd subjects in to-copular clauses in Polish, the nominative case on both DPs surrounding the two copulas, and the φ-feature agreement with the postverbal, rather than the pre-verbal item. ← 9 | 10 →

In chapter 5, Réka Jurth analyses Hungarian object experiencer verbs that have subject experiencer counterparts. Following Alexiadou and Iordăchioaia’s (2014) account of Greek and Romanian psychological predicates, Jurth makes an attempt at testing whether alternating psych verbs in Hungarian involve causation. In the paper causative alternation is understood in the same way as in Levin and Rappaport Hovav (2005), Alexiadou et al. (2006), and Schäfer (2008; 2009). Jurth notes that object experiencer verbs and their subject experiencer variants show the same morphology as the verbs that undergo causative alternation in Hungarian. However, the alternating object experiencer verbs differ from their subject experiencer counterparts as regards aspect, eventivity, event structure, the realisation of the causer with the subject experiencer verbs and resultativity. The main claim of the paper is that the alternating object-subject experiencer verbs that express a change of state represent a subtype of the causative alternation, whereas those alternating object-subject experiencer verbs that do not convey a change of state form a heterogeneous class. Nonetheless, both groups of alternating object-subject experiencer verbs are compatible with causative semantics.

In chapter 6, Katarzyna Mroczyńska discusses the problem of the antipassive construction in Polish. The author characterises the use of the marker się, which does not have to indicate the missing patientive object and the antipassive structure in all of the cases. This effect is independent of the semantic class of the verb, but is closely connected with the development of the antipassive and reflexive structures involving the polysemous marker się. The author observes that certain verbs with się allow an antipassive and a reflexive interpretation, whereas others only allow reflexive readings. Verbs with the marker się do not exhibit an identical behaviour due to both the polysemous nature of się and the multiple structural functions it may perform, such as the reflexive/reciprocal, anticausative, middle voice, passive, antipassive or involuntary state się constructions.

In chapter 7, Katarzyna Sówka-Pietraszewska studies the syntax of load verbs in Old English. The verbs from this group license the so-called locative alternation, whereby they can appear in the locative frame and in the with-frame. In the locative frame, there is a Theme direct object, followed by a PP location, while in the with-frame a Location direct object appears, accompanied by a Theme PP. In present-day English, the two patterns differ in meaning and event structure. In Old English, the number of structures with the with-frame was significantly higher than the number of sentences with the locative frame. The author argues that the with-frames in Old English signalled events with an endpoint, in contradistinction to present-day German, where they do not have a resultative interpretation. Few structures with the locative frame that were present in Old English are not direct ← 10 | 11 → equivalents of the present-day English locative frame, as they do not represent genuine locations. Sówka-Pietraszewska concludes that the locative frame must have developed later than in Old English, and that the two patterns that load verbs show are not derivationally related.

The second part of the present volume, called “Constraints on structure and derivation in phonology and morphology,” comprises five chapters which focus on the operation of constraints in phonology and morphology. It is noteworthy that recently phonological theory “has witnessed a move away from rule-based analyses of structural differences among and within languages” (Harris 1994, xi). Present-day studies are more likely to perceive linguistic variation as deriving from the language-specific parameters of universal principles. Beyond doubt, Government Phonology constitutes a conspicuous example of this type of theorising. Here, the notion of phonological representation lies at the heart of phonological analysis. It is viewed as the source of all phonological events. Phonological processes always apply wherever the structural conditions for their operation are satisfied. The model is defined as non-derivational in the sense of eliminating any rule component that would be responsible for mapping the lexical representations onto phonetic forms. The derivational aspect has thus been reduced to a very limited set of primitive operations, as opposed to the previous rule-based accounts favoured by Generative Phonology. The framework advocates the view that phonological phenomena are amenable to principle-based explanations. It is thus proposed that phonological representations are constrained by a finite set of universal laws and all phonological events ultimately stem from structural settings. The approach to derivation delineated above remains in sharp contrast both to the one formulated by Declarative Phonology (Coleman 1995), where no derivations are allowed, and the model proposed by Bromberger and Halle (1989), in which free interplay between phonological processes takes place (Kaye 1995).

The notion of constraint, crucial to phonological theory, can be approached from two different angles. On the one hand, it can be applied to a particular model and in this way betray either its unconstrained or constrained nature. Viewed from this perspective, phonological frameworks formulate criticisms with respect to one another. More specifically, Government Phonology tends to be criticised for being insufficiently constrained, while Declarative Phonology is perceived as being too heavily constrained. On the other hand, we have different definitions of phonological constraints. The one advocated by Government Phonology views constraints as universal restrictions imposed on lexical structure. Examples of such limitations are the Projection Principle (Harris and Kaye 1990), the Complexity Condition (Kaye, Lowenstamm and Vergnaud 1990) or the Empty Category Principle (Lowenstamm ← 11 | 12 → and Kaye 1986). Within the model of Optimality Theory (Prince and Smolensky 1991), by contrast, phonological analysis focuses on the interplay of constraints, producing different language-specific rankings of them. More precisely, it establishes a dichotomy between the operational component of grammar, called GEN, and the constraint component. The former constructs a set of candidate output forms that in many ways differ from the input. The constraint component (EVAL) has an evaluating character and its function is to select a number of forms to be the actual output of the grammar. Whatever definition of a phonological constraint is adopted, all contemporary models subscribe to the view that they perform a major role in phonological processing and consequently in phonological analysis. The five contributions to this volume that address language-specific effects caused by the operation of phonological constraints, attempt to define the existing derivational mechanisms and try to establish the structural configurations underlying the linguistic phenomena.

In chapter 8, Anna Bloch-Rozmej considers the problem of geminates in Polish. The phenomenon is viewed from the perspective of Government Phonology. She discusses the available approaches to the representation of consonantal geminates and points to the evidence that testifies to their existence in a given linguistic system. The representation of such segments in Polish, as argued in the analysis, is closely connected with the operation of inter-onset government whose working is regulated by Polish-specific constraints. In this language a distinction between true and fake geminates has to be recognised. The contraction of inter-onset governing domains, in turn, is heavily dependent on the licensing properties of nuclei. Interestingly, the two geminate types exhibit different phonological behaviours.

In chapter 9, Rita Manzini and Leonardo M. Savoia explore the issue of the neuter in Italian and Albanian varieties. The discussion is couched within a generative, minimalist framework. The analysis aims at proving that morphological externalisation patterns are sensitive to some other deeper patterns. This results in dedicated lexicalisations for mass singulars and syncretisms between mass singulars and count plurals. Their analysis is compatible with the assumptions of Distributed Morphology in showing that words should not be defined as primitives but as complex syntactic objects, which indicates that their internal structural organisation is similar to that of phrases. The authors subscribe to the view that the lexicon is the basis for the projection of syntactic structures. They maintain that N/Class morphology possesses semantic content, which testifies to the existence of the syntax – semantics interface. In conclusion, it is proposed that the neuter reveals that the mass/count distinction is coded by the N/Class morphology. ← 12 | 13 →

In chapter 10, Leonardo M. Savoia and Benedetta Baldi examine the nature of vowel – sonorant interactions and the processes effected thereby. The author analyses the relevant phenomena attested in Italo-Romance and Romansh languages. The analysis proposes language-specific phonological representations of forms where such effects can be observed. It is claimed that a relationship exists between particular segmental content and its prosodic manifestation. Another significant question that is addressed is the relationship between structure and melodic content as well as its effect on the phonetic manifestation of forms. The analysis relies on the application of phonological constraints such as the Projection and Non-Arbitrariness Principles. The data presented and analysed reveal the cavity and duration properties that stressed nuclei possess when occurring before sonorants and in contexts involving the acoustic space associated with long nuclei. The colour and length of stressed vowels, as proposed in the analysis, affects the licensing capacity of the corresponding nucleus.

In chapter 11, Steven Schaefer discusses the structure and phonetic manifestation of “combining forms,” which he defines as compound words deriving from Latin and Greek bound stems, such as /scope/ and /photo/. These items become attached either to free morphemes or to other bound morphemes in morphologically complex forms following the patterns typical of bi-morphemic compounding. These are argued to form the basis for building larger, multi-morpheme sequences. In the pronunciation of such forms, morphological factors seem to play a significant role, for example, by determining stress placement. In fact, they interact with the phonological factors involved in the behaviour of word-final syllables for instance. The author proposes that the structure of certain roots determines their stress-attracting character, in other cases they are also responsible for a shift of stress to the antepenultimate syllable of the compound. The analysis culminates in the formulation of the morpho-phonological rules that drive the process of compounding of the bimorphemic forms that are examined.

In chapter 12, Sławomir Zdziebko focuses on the structure of resultative adjectives in Polish and proposes an account of the particular vowel alternations that they exhibit. More specifically, a lack of alternation is blamed on cyclic-spell-out where the feature [-masculine-personal] is absent at the stage where the resultative adjective structure undergoes Vocabulary Insertion. The discussion also involves the morpho-syntactic properties of Polish verbs and their division into classes. The author outlines Cetnarowska’s (2000) approach to resultative adjectives and presents both their syntactic and semantic characteristics and exponence. In conclusion, a syntactic approach to the derivation of resultative adjectives is proposed. ← 13 | 14 →

As briefly delineated above, the concept of lexical structure is crucial to both the analyses of linguistic phenomena addressed by the individual authors and our general understanding of the organisation of linguistic systems. The accounts of language-specific developments reveal the existence of both universal and language-specific constraints that regulate and impose limitations on the operation of processes and also establish the bounds of well-formed language structures. We hope that the present volume will contribute interesting insights into the ongoing debate of the abovementioned issues.

We owe a debt of gratitude to the many people who have contributed to the creation of this volume. Special thanks go to the Authors of the individual chapters for their critical and insightful analyses as well as their wholehearted co-operation. We address our thanks to Professor Jolanta Szpyra-Kozłowska, the editor of the series SOUNDS – MEANING – COMMUNICATION. Landmarks in Phonetics, Phonology and Cognitive Linguistics, for supporting our proposal for this book, and for her invaluable guidance and suggestions. We gratefully acknowledge the help of Professor Ángel Jiménez Fernández, who reviewed the entire volume and recommended it for publication. We also would like to express our thanks to Elżbieta Sielanko-Byford and Nigel Byford for having proofread the entire volume and to Ms Monika Gozdór for extensive administrative and secretarial support.

References

Alexiadou, Artemis, Elena Anagnostopoulou, and Florian Schäfer. 2006. “The Properties of Anticausatives Cross-Linguistically.” In Phases of Interpretation, edited by Mara Frascarelli, 187–212. Berlin: Mouton de Gruyter.

Alexiadou, Artemis, and Gianina Iordăchioaia. 2014. “The Psych Causative Alternation.” Lingua 148 (September):53–79.

Bondaruk, Anna. 2012. “Person Case Constraint Effects in Polish Copular Constructions.” Acta Linguistica Hungarica, 59(1–2):49–84.

Brody, Michael. 1995. Lexico-Logical Form: A Radically Minimalist Theory. Cambridge, MA: MIT Press.

Brody, Michael. 2000. “Mirror Theory: Syntactic Representation in Perfect Syntax.” Linguistic Inquiry 31:29–56.

Brody, Michael. 2002. “On the Status of Representations and Derivations.” In Derivation and Explanation in the Minimalist Theory, edited by Samuel D. Epstein and Daniel T. Seely, 19–41. Oxford: Blackwell. ← 14 | 15 →

Bromberger, Sylvain and Morris Halle. 1989. “Why Phonology Is Different.” Linguistic Inquiry 20:51–70.

Cann, Ronnie, Ruth Kempson, and Lutz Marten. 2005. The Dynamics of Language: An Introduction. Amsterdam: Elsevier.

Cetnarowska, Bożena. 2000. “The Unergative/Unaccusative Split and the Derivation of Resultative Adjectives in Polish.” In Formal Approaches to Slavic Linguistics: The Philadelphia Meeting 1999, 78–96, edited by Tracy Holloway King, and Irina A. Sekerina. Ann Arbor, MI: Michigan Slavic Publications.

Chomsky, Noam. 1995. The Minimalist Program. Cambridge, MA: MIT Press.

Chomsky, Noam. 2000. “Minimalist Inquiries: The Framework.” In Step by Step, edited by Roger Martin, David Michaels, and Juan Uriagereka, 89–155. Cambridge, MA: MIT Press.

Coleman, John 1995. “Declarative Lexical Phonology.” In Frontiers of Phonology. Atoms, Structures, Derivations, edited by Jacques Durand and Francis Katamba, 333–383. London and New York: Longman.

Epstein, Samuel D., and Daniel T. Seely. 2002. “Rule Applications as Cycles in a Level-Free Syntax.” In Derivation and Explanation in the Minimalist Program, edited by Samuel D. Epstein and Daniel T. Seely, 65–89. Oxford: Blackwell.

Epstein, Samuel D., and Daniel T. Seely. 2006. Derivations in Minimalism. Cambridge, MA: Cambridge University Press.

Epstein, Samuel D., Erich M. Groat, Ruriko Kawashima, and Hisatsugu Kitahara. 1998. A Derivational Approach to Syntactic Relations. Oxford: Oxford University Press.