Publishing Partner: Cambridge University Press CUP Extra Publisher Login

New from Cambridge University Press!


Revitalizing Endangered Languages

Edited by Justyna Olko & Julia Sallabank

Revitalizing Endangered Languages "This guidebook provides ideas and strategies, as well as some background, to help with the effective revitalization of endangered languages. It covers a broad scope of themes including effective planning, benefits, wellbeing, economic aspects, attitudes and ideologies."

New from Wiley!


We Have a New Site!

With the help of your donations we have been making good progress on designing and launching our new website! Check it out at!
***We are still in our beta stages for the new site--if you have any feedback, be sure to let us know at***

Review of  Theoretical Approaches to Disharmonic Word Order

Reviewer: Alexandru Cosmin Nicolae
Book Title: Theoretical Approaches to Disharmonic Word Order
Book Author: Theresa Biberauer Michelle Louise Sheehan
Publisher: Oxford University Press
Linguistic Field(s): Linguistic Theories
Issue Number: 26.3529

Discuss this Review
Help on Posting
Review's Editor: Helen Aristar-Dry


The book is structured in five parts (''On the nature of disharmony'', ''The role of prosody'', ''The question of antisymmetry'', ''Novel alternatives to antisymmetry'', ''The Final-over-Final constraint''), which include 15 chapters, preceded by the introduction to the volume written by the editors, Theresa Biberauer and Michelle Sheehan (henceforth BandS).

The introductory chapter written by the editors, entitled ''Theoretical Approaches to Disharmonic Word Order'' (pp. 1-44), is more than a roadmap of the volume (as introductions to volumes of collected papers generally are); rather, it is a substantive contribution which assesses the place of word order in linguistic theory, with special focus on the problem of harmonic and disharmonic word order. Every chapter of the book is thus placed in a tradition of research, a very useful strategy for the reader. Furthermore, BandS not only present, but critically examine every chapter of the book. After an extremely scholarly survey of word order preoccupations prior to Greenberg's (1963) seminal work, which appear to be very scarce, BandS closely examine the Greenbergian notions of harmony and disharmony, and review and assess the limits of the main approaches to harmonic and disharmonic word order since Greenberg (1963) to the present-day, paying special attention to Kayne's (1994 et seq.) notion of 'antisymmetry'. The antisymmetric approach (via the Linear Correspondence Axiom (= LCA)) imposes a rigid theory of word order, the canonical position of constituents being spec-head-comp; structures diverging from this order (e.g. head-final structures) are derived via movement. While evidence of the non-existence of rightward specifiers has been repeatedly noted in the literature, the same cannot be said for the ban on leftward complements. The final part of the introductory chapter is devoted to the Final-over-Final Constraint (=FOFC), a constraint which rules out head-initial phrases dominated by head-final phrases which are part of the same extended projection. FOFC is taken as a crucial piece of evidence for the fact that the order of heads and complements is manipulated by the LCA.

The second chapter, ''Word-Order Typology: A Change of Perspective'' (pp. 47-73), written by Gugliemo Cinque, is programmatic. Cinque attempts a shift of paradigm, starting from a novel research question, namely ''what precisely the harmonic word-order types are that we can theoretically reconstruct, and to what extent each language (or subset of languages) departs from them'' (p. 49) rather than ''what the predominant correlates of OV and VO orders in actual languages are'' (ibidem). On the basis of Merge (reflecting scope properties) and Move, Cinque identifies two harmonic types, the 'head-initial' type and the 'head-final' type. He illustrates them with clausal (CP-VP) and nominal (DP-NP) structures, and then establishes two generalizations which hold within these two purely harmonic types: (i) whatever precedes the VP/NP reflects the order of Merge; (ii) whatever follows VP/NP is the mirror image of the order of Merge. The most important consequences of this proposal are (i) the possibility of measuring the degree of deviation of actual languages from one or the other ideal type (which allows for a finer-grained typologization); (ii) the possibility of establishing which correlations are most stable and which are more prone to be relaxed; and (iii) the possibility of giving a principled account, dictated by minimal derivational desiderata (e.g. Merge), of certain intra-category generalizations (e.g. FOFC).

Chapter 3, ''Postposition vs Prepositions in Mandarin Chinese: The Articulation of Disharmony'' (pp. 74-105), coauthored by Redouane Djamouri, Waltraud Paul and John Whitman, deals with the highly controversial issue of PPs in Chinese. These are, as the authors observe, both mixed (comprising pre-, post-, and circumpositions) and disharmonic (prepositions occur with head-final NPs, postpositions appear with head-initial VPs). The analysis put forth by the authors is maximally simple and elegant: they show that postpositions are categorially similar to prepositions (hence, they are not nouns) but they denote Place, while prepositions denote Path; the circumpositional structure observes the universal ordering Path > Place. Furthermore, postpositional phrases are shown to pattern with the other phrase-head-final structures of Mandarin, namely NP and CP, whose common property is the inability of their head to assign Case to its complement. Thus, since postpositions are non-Case assigning heads, their complement moves to the specifier of the postpositional projection, and Case assignment is ensured either by (internal/external) merger in an argument position or by a prepositional selecting head. This solution elegantly accounts for the distributional disparities between PrepPs and PostPs while still preserving the same categorial status for pre- and postpositions, and derives disharmony from a hierarchical universal (Path > Place; cf. Svenonius 2007) combined with a language-particular property of Chinese (the inability of phrase-final heads to assign Case).

The fourth chapter, ''The Mixed OV/VO Syntax of Mòcheno Main Clauses: On the Interaction between High and Low Left Periphery'' (pp. 106-135), written by Federica Cognola, represents a fine-grained analysis of main clause word order in Mòcheno, a Tyrolean (Germanic) variety spoken in Northern Italy. Mòcheno is generally characterised as a Germanic variety heavily influenced by Romance as far as constituent-positioning is concerned. The characteristic clause structure (V2) of Continental Germanic is possible, but not obligatory; different parameter settings are assumed to account for the OV/VO variation. Against this background, Cognola presents an extensive analysis of main clause direct object topics and foci, and of the V2 syntax of Mòcheno, and convincingly shows that the OV/VO variation results not from different grammars in competition (cf. Kroch 1989), but rather from a fine-grained interaction between the inner workings of the V2 rule and the information status of the scrambled constituents. The former is argued to apply both to the finite verbs in the higher phase and the past participle in the lower phase. Several interesting results are established (the isomorphy, but not identity, of the CP and vP peripheries; the parallel drawn between FinP and VoiceP, and the identification of a LowForceP), which will certainly be taken up in future research.

Joseph Emonds' chapter, ''Universal Default Right-Headedness and How Stress Determines Word Order'' (pp. 139-161), the fifth in the book, opens the section devoted to the role of prosody in determining word order and headedness. The main objective of the chapter is to put forward a unified account of how the position of heads in morphology and syntax is determined (''morphology, at least as it is usually conceived, has no special principles for head placement that distinguish it from syntax '', p. 141). Thus, Emonds takes issue with Kayne's (1994) universal underlying left headness and proposes, in its place, that the universal default setting is right-headedness, and left-hand heads result only from language-specific deviations. The universal ordering principles advanced by Emonds depend on language-particular stress patterns, and may be summarised on the basis of the following generalizations: (i) ''Universal Default Head-Final Order'': as a default at the syntax-PF interface, heads are right sisters of non-heads (p. 145), and (ii) Nespor and Vogel's (1982) ''Complement Law'': complements rather than heads are preferred locations for stress in all types of domains. Importantly, phonology-driven accounts of word order like the one proposed by Emonds in this study are starting to develop into a separate research strand-- witness Richards' (2014) monograph and the next two chapters (Hinterhölzl; Tokizaki and Kuwana), for example.

Roland Hinterhölzl (''(Dis)Harmonic Word Order and Phase-Based Restrictions on Phrasing and Spell-Out'', pp. 162-189) sets as his goal the derivation of (dis)harmonic word orders in a framework which (at least in part) allows for the encoding and manipulation of prosodic information in narrow syntax. Focusing especially on (the ''heaviness'' of) event-related adjuncts and their position with respect to the verb in English and German, Hinterhölzl argues for the existence of two types of transparency, prosodic transparency (Mapping Condition to PF = ''a heavy syntactic constituent must appear on a dominant branch in prosodic phrasing if its containing phase is weight-sensitive'', p. 163) and scope transparency (Mapping Condition to LF = ''if a scopes over b, the Spell-Out copy of a should c-command the Spell-Out copy of b'', p. 163). These conditions are coupled with the distinction between homorganic and non-homorganic phases (homorganic phases are projected by the same phase predicate). Based on these assumptions and on a novel approach to the relation between vPs and event-related adjuncts (viewed as a derived subject-predicate relation), Hinterhölzl successfully manages to encode certain prosodic requirements in narrow syntax and derive word order constraints by interweaving syntax and prosodic structure.

In ''A Stress-Based Theory of Disharmonic Word Orders'' (pp. 190-215), Hisao Tokizaki and Yasutomo Kuwana establish a strong correlation between the position of stress and the availability of complement-head orders (i.e. roll-up, comp-to-spec movement, cf. Kayne 1994) in a given language. Starting from the empirically supported assumption that the ''juncture'' (=the relation between segments in a sequence) is shorter in left-branching structures (i.e. complement-head structures), it is shown that complement-head structures behave like compounds, and hence should have the same stress location as the words and compounds in the language concerned. Hence, complement-to-specifier movement in a language / structure should be allowed only if the resulting stress configuration does not clash with stress pattern of the language in question; otherwise, this instance of movement (taken as being obligatory) is postponed to LF (or resolved via Agree not followed by movement, as suggested in fnt. 5/p.193). Coupled with a fine-grained typology of word-stress locations, this proposal elegantly accommodates the discrete scale of head-finality (e.g. Japanese) to head-initiality (e.g. Romance, Bantu) on the basis of the position of word-stress.

In Chapter 8, ''Why Are There No Directionality Parameters?'' (pp. 219-244), Richard S. Kayne sets out to reinforce one of the basic results from his 1994 antisymmetry theory (Kayne 1994) -- namely, that there are no directionality parameters, and the basic format provided by UG is S(pecifier)-H(ead)-C(omplement) -- in a more derivational approach to syntax, which makes use of the Agree (Probe-Goal) system put forth by Chomsky in the post-2000 papers (Chomsky 2000, 2001, 2008). After reviewing a series of cross-linguistic gaps and asymmetries which indicate that we are not living in a ''symmetric syntactic universe'' -- dislocations and hanging topics, clitics, agreement, relative clauses, serial verbs, coordination, and forward and backward pronominalization --, Kayne carries on by showing that a point which was taken as axiomatic in Kayne (1994) -- i.e. the absence of directionality parameters via the Linear Correspondence Axiom (LCA) -- can actually be derived if precedence is part of narrow syntax. In order to demonstrate this point, Kayne explores an alternative presented (but not pursued) in Chomsky (2008): the merger of two syntactic objects is taken to form an ordered pair <X,Y>, not a set {X,Y}. Coupled with the assumption that the merger of two phrases is unavailable and with the notion of i(mmediate)-precedence, Kayne shows that a head H can be merged with at most two elements, a C and a S, placed on opposite sides of H. Hence, the S-H-C format of UG is derived from more primitive syntactic operations and need not be taken as an axiom; the non-existence of multiple specifiers is also elegantly derived in this account.

Michael Barrie (Chapter 9, ''Antisymmetry and Hixkaryana'', pp. 245-269) also makes a strong case for antisymmetry, thus continuing Kayne's line of thought. Focusing on Hixkaryana and tangentially on Urarina (two genetically unrelated OVS languages), Barrie questions the existence of the Head Parameter, and shows that such an approach is unable to accommodate asymmetries like the one discussed here: SOXV orders are allowed in SOV languages, while OXVS orders are not allowed in OVS languages. The unavailability of OXVS in the languages under scrutiny is elegantly accounted for by the following operations: (i) roll-up movement of O across V in a ''ghost AgrO'P'', (ii) VP-movement of the newly created VP-shell across the subject, thus smuggling O above S (cf. Collins 2005) and circumventing relativized minimality effects. The second operation is inspired by Massam's (2001) work on Niuean, more specifically, by the assumption that EPP in OVS languages is satisfied by VP rather than DP.

In contrast to the two previous chapters, in Chapter 10 (''Postverbal Constituents in SOV languages'', pp. 270-305) Balkiz Öztürk argues for the existence of rightward movement on the basis of the contrastive analysis of the behaviour of postverbal constituents (PVCs) in two SOV Altaic languages, Khalkha Mongolian and Uyghur Turkic. It is shown that while there is no evidence of a movement-based derivation of PVCs in Khalkha, PVCs in Uyghur display a parallel behaviour to leftward scrambled elements, hence their derivation should involve (rightward) movement. Furthermore, on the basis of cross-Altaic evidence (Japanese, Turkish, Eastern Uzbek, and Osh Kirghiz), a parameterisation of rightward movement on the basis of the presence or absence of EPP-effects is argued for.

In the same line of reasoning, Arantzazu Elordieta (Chapter 11 ''On the Relevance of the Head Parameter in a Mixed OV Language'', pp. 306-329) carefully reviews the evidence which tilts the balance in favour of a modified version of the Head Parameter over an antisymmetric approach, on the basis of an extensive examination of Basque. It is shown that Basque is a non-FOCF-violating disharmonic language (on FOFC, see Biberauer, Holmberg and Roberts 2014, and Part V in the book): the C-domain (i.e. the heads associated with discourse and scope) shows all the signs of head-initiality, while the domain below C is uniformly head-final. The derivation of PP complements and the position of manner adverbs with respect to PPs constitute the crucial pieces of evidence which help tell apart the two competing accounts (other constructions being equally well derivable under a Head Parameter account and an antisymmetric approach). The results are supported by findings from language acquisition and language processing.

In the chapter that opens up the fourth section of the book dedicated to novel alternatives to antisymmetry, Mark de Vos (Chapter 13 ''Afrikaans Mixed Adposition Orders as a PF-Linearization Effect'') puts forth a novel approach to linearisation based on dependency formation, formalised as Dependency Spell-Out. The upshot of Dependency Spell-Out is that the structure passed to the PF-interface is not merely a set of phrase-markers, but a set of functional dependencies; in their turn, the functional dependencies established in narrow syntax play a pivotal role in determining linearization. Setting out to provide an account of the disharmonic word orders (mixed headedness) present in the Afrikaans adpositional system, de Vos shows that disharmony/mixed headedness is only apparent; building on previous literature, it is elegantly shown that a uniformly head-initial system combined with syntactic functional dependency linearisation suffices to account for the data.

The second alternative to antisymmetry put forward is Takashi Toyoshima’s transversal linearisation of bare phrase structure (Chapter 13 ''Traversal Parameter at the PF Interface: Graph-Theoretical Linearization of Bare Phrase Structure'' pp. 358-388). Toyoshima starts by discussing the shortcomings and predicaments of Kayne’s LCA, identifying as weak points both the necessity of postulating functional projections which play no other role than supplying landing-sites, and fact that the original formulation of the LCA does not axiomatically exclude a series of structures (which are -- paradoxically! -- taken by many researchers to be excluded), such as n-ary branching (n>0), adjunction of heads to non-heads and vice versa, multiple specifiers, etc. (see also Guimarães 2008). To overcome these problems, Toyoshima proposes a linearisation procedure alternative to Chomsky’s (1995) Bare Phrase Structure, based on a graph-theoretical approach (see also Kural 2005) which is assumed to apply at the PF interface, as understood in the earlier Y-models.

In Chapter 14 (''Disharmonic Word Orders from a Processing-Efficiency Perspective'', pp. 391-406), John A. Hawkins puts forward an alternative account of disharmonic word orders based on language processing and efficiency considerations. The starting point of the discussion is the typological observation that while consistent (head-initial or head-final) types account for the majority of languages, mixed types (both FOFC-violating, and non-FOFC-violating) are generally dispreferred and occasionally unattested. The upshot of Hawkins’ proposal is that ''grammars have conventionalised syntactic structures in proportion to their degree of preference in performance'' (p. 406). In the present contribution, Hawkins reviews a series of parsing principles (Mother Node Construction; Immediate Constituent Attachment; Minimize Domains), subsumed to the Performance-Grammar Correspondence Hypothesis, which have been put forward and extensively developed in his own work (Hawkins 1994, 2004, 2014), taken to account for (dis)harmony from a processing perspective.

Michelle Sheehan (Chapter 15 ''Explaining the Final-over-Final Constraint: Formal and Functional Approaches'', pp. 407-444) examines two competing accounts of FOFC: a functional approach grounded in the work of Hawkins (1994, 2004, present volume), and a formal explanation developed in the author's own work, i.e. a PF-interface account, different in certain respects from the formal account provided in Biberauer, Holmberg and Roberts (2014) (and previous work). An important part of the paper is devoted to the review of the empirical evidence for FOFC (inflected auxiliary placement and the verb phrase: the cross-linguistic absence of V-O-Aux orders; polarity question particles and complementizers; clausal complements), and of the counterexamples to FOFC (some of which are apparent, some of which at the very least have an unclear status, e.g. particles). Thus, while both the functional and the formal approaches fare well in accounting for a large amount of FOFC effects and violations, several empirical considerations, as well as the fact that performance based-approaches entail that there is a relation between cross-categorial harmony and FOFC (contrary to fact), tip the scales in favour of the formal account of FOFC, a variant of which is put forward in this chapter.

In the final chapter of the book, Brian Hok-Shing Chan (Chapter 16 ''Sentence-Final Particles, Complementizers Antisymmetry, and the Final-over-Final Constraint'', pp. 445-468) approaches the thorny problem of sentence-final particles (SFPs) in different Chinese varieties. Chinese SFPs are known to be a notorious counterexample to FOFC, given their traditional analysis as C-heads. However, by systematically comparing the behaviour of true complementizers and SFPs, the author shows that SFPs receive a more adequate analysis if they are treated as a category of their own, different from complementizers. Thus, SFPs are analysed as affixes attached to a focused constituent, which moves to the specifier of a phonetically null Focus head, a solution which avoids FOFC-violations.


The book reviewed is impressive from many points of view. First and foremost, from an empirical perspective: the material discussed in the chapters of the book is drawn from a large number of genealogically unrelated, typologically distinct and geographically diverse languages, some of them familiar to the general public, some of them rarely discussed in the literature. A second important strength of the volume is the range of perspectives from which the issue of harmony and disharmony is analysed: virtually every level of the faculty of language (narrow syntax, interfaces, phonological structure, etc.) plays an important part in the identification and explanation of (dis)harmony. It goes without saying that the book is a 'must-read' not only for researchers interested in the issue of word order, but also for typologists, generativists, theoreticians, and descriptive linguists.

Besides this general characterisation, I would also like to point out a few other aspects, which (at least to me) seem to hold true across a large number of chapters in the book. To begin with, a recurrent (beneficial) theme is that disharmony appears to be derived from universal constraints and language specific characteristics (parameter settings), a result which offers a principled account of disharmony. Strong evidence for the antisymmetric nature of natural language syntax is another important result, central to Kayne’s and Barrie’s chapters (ch. 8 and 9), but also visible in other parts of the book (e.g. in the discussions surrounding FOFC). Another interesting (and somewhat controversial) issue taken up at various points in the book is the problem of ''precedence'' and its relation to narrow syntax and the interfaces (and hence word order): contrary to the general assumption that the merger of two syntactic objects forms a set, Kayne (ch. 8) claims that precedence is part of narrow syntax and derives from the fact that the result of merger is an ordered pair; de Vos (ch. 12) claims that precedence is established in narrow syntax in the form of a dependency relation, which is then transferred to the PF-interface, playing a pivotal role in linearisation.

On a final note, the editors of the volume should be warmly congratulated, not only for managing to harmonise the contents of the book by selecting complementary analyses which neatly piece together the puzzle of word order harmony and disharmony, but also for the excellent introductory chapter, a well-informed, up-to-date discussion of the multiple issues raised by harmony and disharmony.


Biberauer, T., Holmberg, A., Roberts, I. 2014. A Syntactic Universal and its Consequences. In Linguistic Inquiry 45: 169–225.

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

Chomsky, N. 2000. Minimalist inquiries: the framework. In: R. Freidin, D. Michaels, J. Uriagereka (eds.), Step by Step: Minimalist Essays in Honor of Howard Lasnik. Cambridge, MA: MIT Press, 89-155.

Chomsky, N. 2001. Derivation by phase. In M. Kenstowicz (ed.), Ken Hale: A life in language. Cambridge, MA: MIT Press, 1-52.

Chomsky, N. 2008. On phases. In: R. Freidin, C. P. Otero, M. L. Zubizarreta (eds.), Foundational Issues in Linguistic Theory. Essays in Honor of Jean-Roger Vergnaud. Cambridge, MA: MIT Press, 133-166.

Collins, C. 2005. A smuggling approach to the passive in English. In: Syntax 8, 2: 81-120.

Greenberg, J. 1963. Some universals of grammar with particular reference to the order of meaningful elements. In: J. Greenberg (ed.), Universal of Language. Cambridge, MA: MIT Press, 73-113.

Guimarães, M. 2008. A note on the strong generative capacity of standard Antisymmetry theory. In Snippets 18: 5-7.

Hawkins, J. A. 1994. A Performance Theory of Order and Constituency. Cambridge/New York: Cambridge University Press.

Hawkins, J. A. 2004. Efficiency and Complexity in Grammar. Oxford: Oxford University Press.

Hawkins, J. A. 2014. Cross-linguistic Variation and Efficiency. Oxford: Oxford University Press.

Kayne, R. S. 1994. The Antisymmetry of Syntax. Cambridge, MA: MIT Press.

Kroch, A. 1989. Reflexes of grammar in patterns of language change. In: Language Variation and Change 1, 199-244.

Kural, M. 2005. Tree traversal and word order. In Linguistic Inquiry 36: 367-388.

Massam, D. 2001. On predication and the status of subjects in Niuean. In: W. Davies, S. Dubinsky (eds.), Objects and Other Subjects. Amsterdam: Kluwer, 225-246.

Nespor, M., Vogel, I. 1982. Prosodic domains of external sandhi rules. In: H. van der Hulst, N. Smith (eds.), The Structure of Phonological Representations. Dordrecht: Foris, 225-56.

Richards, N. 2014. Contiguity Theory.

Svenonius, P. 2007. Adpositions, particles and the arguments they introduce. In: E. Reuland, T. Bhattacharya, G. Spathas (eds.), Argument Structure. Amsterdam: John Benjamins, 71-110.
Alexandru Nicolae is a researcher at “Iorgu Iordan - Al. Rosetti” Institute of Linguistics, Bucharest, and a teaching assistant at the Department of Linguistics, University of Bucharest. PhD dissertation (2013): ''Types of Ellipsis in Romanian'', University of Bucharest (& University of Cambridge, cotutelle). He is currently working on word order and configurationality in (old) Romanian. His research interests include: minimalist syntax, diachronic syntax, and the syntax of Romanian. He has co-authored the latest academic grammars of Romanian (“Gramatica de Bază a Limbii Române”, ed. Gabriela Pană Dindelegan, 2010; “The Grammar of Romanian”, Oxford University Press, ed. Gabriela Pană Dindelegan, 2013), and has been working in the past five years with Alexandra Cornilescu on the syntax of the Romanian nominal phrase. Visiting PhD Student (2012) and Visiting Researcher (2015) at the University of Cambridge.

Format: Hardback
ISBN-13: 9780199684359
Pages: 560
Prices: U.K. £ 75.00