Conference report: "VO and OV",
University of Tromsø, 22-23 May 1998
Written for Glot International
by Peter Svenonius
(see the conference Program for a list of the talks)
There has been a recent groundswell of research on basic word order, inspired in large part by Kayne's work (e.g. his 1994 book) in which it is proposed, among other things, that all branching is binary, all complements are to the right, all adjuncts or specifiers are to the left, and all movement is to the left. On this view, there is no headedness parameter; in work like that of Zwart 1993, 1997, Koster 1994, Roberts 1997, it was demonstrated how OV word order in languages like Dutch or Old English could be derived from a basic VO order by wholesale movement of VP-internal material to the left, into functional specifier positions. Thus, a Dutch embedded clause is analyzed much as in (1) (cf. Zwart 1993, chapter IV).
(1)
...dat [AgrSP Jan1 [AgrOP het boek2 [PredP op de tafel3 [VP wil t1 leggen t2 t3 ]]]]
that John the book on the table will put
In English, on this view, the movements to licensing positions such as SpecAgrOP occur covertly. However, such accounts raise various questions. For example, if such movements as those in (1) are triggered by strong features, why are the features generally strong (in consistent OV languages) or generally weak (in consistent VO languages)? In mixed cases, what determines what stays behind (e.g. CP) and what moves (e.g. DP)? If the word order in OV languages is the result of movement, what ensures that the derived order of constituents is identical to the order of those same constituents in a VO language (e.g. indirect object precedes direct object, noun phrase precedes PP)?
Hubert Haider, in a series of articles (Haider 1992, 1995, 1996, 1997a, b, inter alia), has launched an alternative to the Kayne/Zwart view of the difference between VO and OV word order which addresses these questions. On Haider's account, OV order is basic: VP-internal elements are base-generated in the projection of V. The verb starts low and stays there in a strict OV language, but in a VO language, it climbs by head-movement to a higher position. Thus, a sentence like the English one in (4) is derived by head movement, and the VP-internal arguments are in situ, much as in Larson 1988.
(4)
[IP John put1 [VP the book [VP on the table [VP t1 ]]]]
For Haider, the derivation of VO order is forced not by a strong feature, but by the requirements of a parameterized notion of licensing: extended projections must be licensed, and licensing in VO languages is rightward.
Hinterhölzl 1997a, b, bidding for the Kaynean approach of deriving OV from VO, ups the ante, contributing some important developments. First, Hinterhölzl argues that in some cases a larger constituent must move to the left, one from which the verb has been evacuated. Thus one step of Hinterhölzl's derivation for a Dutch clause is sketched in (5) (cf. his 1997a:198/1997b:14; I have simplified the structures, omitting some brackets and traces; Hinterhölzl assumes that the main verb originates to the right of the temporal adverbial morgen, and moves to the left, after the arguments have moved out of the VP à la Zwart; cf. below).
(5)
a. dat Jan [PredP wilde [CP geven [TP Marie het boek morgen]]]
hat John wanted give Mary the book tomorrowb. dat Jan [PredP [TP Marie het boek morgen] wilde [CP geven tTP]]
that John Mary the book tomorrow wanted give
'...that John wanted to give Mary the book tomorrow'
This addresses some of the concerns raised by Haider, for example, it means that relative ordering of VP-internal material (TP-internal, in this analysis) will be preserved in certain cases.
Another important proposal that Hinterhölzl makes is that Verb Raising is the result not of head movement, but remnant VP movement, that is, the VP moves after its arguments have been evacuated from it (actually, what moves is F1P, a functional projection containing the VP). This increases the role of VP-remnant movement in a significant way. An abbreviated version of Hinterhölzl's derivation of a German (embedded) clause with Verb Raising appears in (6). In (6a), the TP has already been fronted (cf. (5b) above; here, TP also contains a trace of F1P, and F1P contains a trace of the object, but I have not indicated these traces). In (6b), F1P moves to the left of the higher verb (Verb Raising). Finally, in (6c), the object scrambles higher.
(6)
a. er [PredP [TP sie nicht ] wagte [CP [F1P zu küssen ] tTP ]]
he her not dared to kissb. er [PredP [TP sie nicht ] [F1P zu küssen ] wagte [CP tF1P tTP ]]
he her not to kiss daredc. er sie1 [PredP [TP t1 nicht ] [F1P zu küssen ] wagte [CP tF1P tTP ]]
he her not to kiss dared
'...(because) he dared to not kiss her'
The wide-scope reading of the same string is derived with exactly the same movements, but with negation starting in the higher clause, as shown in (7).
(7)
a. er nicht [PredP [TP sie ] wagte [CP [F1P zu küssen ] tTP ]]
he not her dared to kissb. er nicht [PredP [TP sie ] [F1P zu küssen ] wagte [CP tF1P tTP ]]
he not her to kiss daredc. er sie1 nicht [PredP [TP t1 ] [F1P zu küssen ] wagte [CP tTP]]
he her not to kiss dared
'...(because) he did not dare to kiss her'
Kayne has maintained the momentum, postulating further movements, in particular he has argued (Kayne 1998) that in English, there are in fact overt movements in some cases of VP-internal material to the left of the VP, followed by VP-remnant fronting past that material again, resulting in what superficially looks like the original word order. This is similar in some ways to the analysis adopted by Larson (1988) and den Dikken (1995) for Heavy NP Shift phenomena, and brings VO languages much closer to Hinterhölzl's vision for the OV ones. For instance, to account for wide scope of negation in a sentence like 'I will force you to marry noone,' Kayne postulates the derivation in (8), where 'W' and 'Neg' are heads in the extended projection of every verb, but shown only for the superordinate clause. For the narrow scope reading, the derivation in (9) is used, and here the functional heads are only shown for the embedded clause.
(8)
a. I will [WP W [NegP Neg [VP force you to marry noone]]]
b. I will [WP W [NegP noone1 Neg [VP force you to marry t1 ]]]
c. I will [WP [VP force you to marry t1 ] W [NegP noone1 Neg tVP ]]]
(9)
a. I will force you to [WP W [NegP Neg [VP marry noone]]]
b. I will force you to [WP W [NegP noone1 Neg [VP marry t1 ]]]
c. I will force you to [WP [VP marry t1 ] W [NegP noone1 Neg tVP ]]]
In this way, Kayne proposes to eliminate covert movement from the system of grammar. Note that the claim that objects in VO languages like English leave their base positions overtly has been around for a while; for example, Johnson 1991, Borer 1994, and Runner 1995 all take this position. In each of those proposals, however, the verb moves to the left by head movement, as on Haider's account, rather than by VP-remnant movement, as in (8-9).
Other developments have been taking place on more distant shores. Fukui 1993 took the base directionality parameter (the head parameter, in particular) of Chomsky 1981, Stowell 1981, Travis 1984, and Koopman 1984, and showed how it could affect movement as well as base-generation. In more recent work, Saito & Fukui (to appear) extend this view, incorporating a modified version of the head parameter and thereby parameterizing Chomsky's 1995 Merge. A somewhat different perspective on linear order is explored in Fukui & Takano 1998, in which a 'symmetry' view on derivations is proposed: in brief, their symmetry principle requires that the order of phrase structure composition (Merge) and the order of phrase structure decomposition ('Demerge,' followed by the linearization process) form mirror images of each other. One of the major consequences of this principle is that the 'basic' word order is SOV, as in Haider's work.
Inspired by all this theoretical activity and encouraged by their highly successful 1997 workshop on 'Subjects, Expletives, and the EPP,' linguists at the University of Tromsø decided to organize a workshop on VO and OV word order, by the light of the midnight sun. The workshop took place on May 22-23, with moderate snowfall and no reported polar bear attacks. Since no conference fee was collected, the organizers have no exact records of attendance, but it's safe to say that attendance was typical for a conference of similar size. At any rate, the workshop was an unqualified success, with excellent talks and active discussion, and a good time was had by all.
Hubert Haider set the tone for the conference with the opening talk, entitled, 'OV is more basic than VO,' continuing the line he had developed in his earlier work. He briefly summarized his theory of VO and OV word order, based on the 'Basic Branching Constraint' (BBC), and contrasted it with Kayne's 1994 theory, based on the 'Linear Correspondence Axiom' (LCA), showing various ways in which the two theories make different predictions, and arguing that the BBC account is empirically superior to the LCA account, for instance with respect to patterns of extraction and crosslinguistically variant patterns of the distribution of adjuncts, secondary predicates, and particles.
During the course of the Workshop, a certain consensus began to emerge. The majority of the talks did in fact derive VO order from OV order, as in Haider's account, either by verb fronting, following Haider, or by VP fronting, in a way similar to that in Kayne's 1998 analysis of English. However, beyond that point significant differences arose. The sharpest theoretical differences were perhaps evident in the two talks based on historical facts, those of Pintzuk and Hróarsdóttir.
Susan Pintzuk argued that the various accounts on the market of OV to VO change in the history of English failed to capture the diachronic facts (discussed in detail in Pintzuk 1991, 1997). For example, she presented statistical evidence suggesting that the relative rate of change from OV to VO order in the history of English was similar for DPs and PPs. No account based on the variability of strength of a feature attracting elements into preverbal positions has been able to predict such a situation, since the features postulated to attract DPs and PPs have been different (and are necessarily so, since in languages like Dutch, DPs obligatorily precede the verb while various kinds of PPs may follow). She showed in addition that the account of Hróarsdóttir 1997 for OV to VO change in the history of Icelandic would not cover the Old English facts. For example, the various optional movements postulated there falsely predict the possibility of having an object between the main verb and the auxiliary in the main verb-auxiliary word order.
(10)
* he thæs habban ece edlean sceal
he thus have eternal reward shall
Pintzuk argued that an analysis based on a variable headedness parameter, with rightward movement of DPs and PPs, accounts for the empirical data as well as any of the more technologically sophisticated approaches, and may support the hypothesis of Kiparsky 1994 that the change from OV to VO is a general move towards uniform head-initial structure.
Thorbjörg Hróarsdóttir moved a step in the direction of solving similar puzzles from the history of Icelandic. She presented data from Middle Icelandic, where VO and OV orders were in variation for a long period before VO became the unmarked order (discussed in Hróarsdóttir 1996). As it turns out, Middle Icelandic, like Old English (and modern German and Dutch) allows main verb-auxiliary order, but never with an object in between, though an object may intervene between an auxiliary (even a non-finite one) and a main verb when the former precedes the latter. Hróarsdóttir argued that several different features of OV order, including VP-internal elements preceding the verbal cluster (non-finite auxiliaries as well as main verb), VP-internal elements preceding the main verb (but not preceding non-finite auxiliaries), and the main verb preceding the auxiliary, all disappeared from Icelandic at the same time, and, echoing Pintzuk, noted that this is not accounted for on most analyses of the derivation of VO order. Hróarsdóttir then proposed that VO order in Modern Icelandic be derived by fronting of the remnant VP, similarly to Kayne's 1998 proposal, an account which successfully links all three of the features and correctly captures the fact that they disappeared simultaneously.
I, Peter Svenonius, continued the discussion of Icelandic, presenting material having to do with negation in English and the Scandinavian languages. In Icelandic, negative objects (equivalent to 'nobody' or 'no books') give rise to OV order, preceding non-finite verbs (as discussed in Rögnvaldsson 1987, Jónsson 1996), i.e. the (b) examples in (8-9) above display in part the overt word order for Icelandic.
(11)
a. Hún var engum neydd til adh giftast.
she was nobody forced to to marry
'Nobody is such that she was forced to marry him'b. Hún var neydd til adh giftast engum.
she was forced to to marry nobody
She was forced to remain unmarried'
The fact that Icelandic has overt negative object movement is one of the motivations in Kayne 1998 for the analysis sketched in (8-9): the idea is that English also fronts negative objects, but then fronts the VP past them.
I argued in my talk that puzzling restrictions on the expression of sentential negation in Scandinavian, including the fronting of negative objects as in (11a), can be made sense of if an account of the expression of sentential negation in the spirit of Ladusaw 1992 is adopted. Specifically, negative expressions like the object in (11a) are licensers of sentential negation, but must appear in a licensing position (SpecNegP, following Jónsson 1996). Ladusaw's account is designed for negative concord languages, which Scandinavian is not. Thus modifications have to be made for multiple negation. However, if negative DPs are taken to be generally quantificational, as in Haegeman & Zanuttini 1997, then false predictions are made. I argued for a hybrid approach, in which negative DPs can only exceptionally be treated as quantificational.
It is interesting, in the light of this data, to compare the specifics of Kayne's 1998 analysis to that of Hróarsdóttir discussed above. Kayne argues on the basis of negative object movement that Icelandic has one fewer movement than English, which is why the negative object remains to the left of the VP. Hróarsdóttir has now argued that Icelandic also has VP fronting. It seems that in order for the two treatments to be reconciled, either Neg fronting in Icelandic is to a higher position than in English, or else VP fronting in Icelandic is to a lower position. It remains to be seen whether additional evidence can be tested against this prediction.
Alison Henry also presented evidence of OV-type structures in VO languages, including object shift phenomena in Belfast English: in imperatives the main verb moves to the left (cf. Henry 1995); in those same contexts, object shift like that seen in Swedish takes place (in Swedish, unlike the other Scandinavian languages, objects may cross subjects under object shift; cf. Holmberg 1986, Josefsson 1993).
(12)
a. Read it you to me.
b. Give it you quickly to them.
In addition, she presented material from Irish, showing that the OV word order in embedded clauses is much harder for English speaking children (in an Irish language immersion school) to acquire than other features which appear to be equally alien to English, for example noun-adjective order. The picture that emerges is that linguistic phenomena like object shift are nearly automatic, at least in a language like English: as soon as the verb is out of the way, the light object moves to the left. However, whatever is responsible for OV word order in embedded clauses in Irish is far from automatic and is in fact quite difficult to acquire. Henry suggested, following Carnie 1995, that the target language (with VSO main clauses and SOV embedded clauses) always has object raising, obscured in main clauses by verb raising, but that in the school environment, children miss the triggers for this object raising, and wind up with embedded VO.
The rest of the talks generally took VO order to be derived, though it would be going too far to say that they all agreed with Haider that OV order is 'basic.' In fact, many of the talks assumed wholesale Kaynean fronting, followed by V or VP fronting. Barbiers assumed V fronting, and Hinterhölzl adopted VP fronting. Several talks were centrally concerned with the issue of V movement versus VP movement, Taraldsen arguing in favor of VP movement for specific phenomena, and Pearson and Holmberg arguing that both VP and V movement are necessarily options. But while Pearson argued for V versus VP movement as a basis for cross-linguistic typology, Holmberg argued for both types of movement as options within a single language (Finnish). Brody argued for a system in which there is no V movement, in fact, no head movement at all, but which mimics its effects, as well as the effects of VP movement.
Matt Pearson, in his contribution, 'Two types of VO languages,' outlined important evidence from Malagasy, Palauan, and Tzotzil that showed that the typological data is more complex than usually assumed. Pearson showed that although VO languages like English and French show the same order of constituents as OV languages like Dutch and Turkish, there is another type of VO language, which shows mirror-image orders, for example the direct object precedes the indirect object in double object constructions, scrambling is to the right, and the relative order of certain adverbials (specifically, the postverbal ones) is reversed (cf. Pearson 1998).
Pearson offered an account of this variation that is consistent with the LCA, given a few critical assumptions. He argued that categorial features on a phrasal node are lost when the head of the phrase is raised; the categorial features of the specifier project to the phrasal node in such cases. Thus, a VP with a DP in its specifier is reanalyzed as a DP when the head V raises to a higher head position. A further critical assumption is that in Malagasy-type languages, there is no verb movement. The derivations then follow: in English-like languages, the VO order is derived by V movement across the various constituents (much as in Haider's approach). In Dutch-like languages, the verb actually moves higher than in English, with the result that the constituent it started from (vP) is reanalyzed as a DP (because of the subject DP in its specifier). This complex element, containing the various arguments, then moves up to the specifier of the nominative-assigning functional head, preceding the verb, preserving the constituent order of the VO languages within the (former) VP. In Malagasy-like languages, verb-movement is unavailable. Thus, when verbal features need to be checked, the remnant VP is forced to move. This leads to successively more complex specifiers, which results in the different observed word orders, because the phrasal nodes in the extended projection are not reanalyzed at each step as they are in the Dutch-type derivations involving V movement.
Anders Holmberg presented data from Finnish showing OV order with a striking constraint: if some focused element occupies a relatively high specifier position, then OV order is licit, sometimes much further down in the clause. To capture the distribution of OV order, Holmberg proposed an account based on a head "New", for 'new information.' Every utterance must have at least one instantiation of New, which can be checked by the verb. Alternatively, a focused element can check New, in which case, the V need not do so; this leads to OV order in a way discussed below. The fact that New appears once per utterance, rather than once per clause, explains the long-distance relation between focus and OV order.
For the specific mechanism by which OV order arises, Holmberg adapted a suggestion of Kayne 1994, developed further in Julien 1997. If V raises to check New, then VO order results. If it does not do so, however, it still must combine with the various inflectional parts of the clause. Holmberg suggested that this can occur either through head-movement, in which case the order in the VP is VO (regardless of the presence of a focus element), or else it can occur through VP movement. However, if VP fronts, the result is necessarily OV order because of the demands of morphology; the verb must be string-adjacent to the functional head that contains the relevant inflectional features. The two possibilities are sketched in (13). In (13a), a pre-Spell-Out structure, the stem of the participle is the head of VP, but the participial affix heads a higher projection, PrtcP. The wh-word in SpecCP is a focus element in the relevant sense and checks New.
(13)
a. Milloin Jussi olisi [PrtcP -nut [VP romaanin kirjoitta]]
when Jussi would.have -en novel writ-b. Milloin Jussi olisi [PrtcP kirjoitta-nut [VP romaanin tV ]]
when Jussi would.have writ -en novelc. Milloin Jussi olisi [PrtcP [VP romaanin kirjoitta] -nut tVP ]
when Jussi would.have novel writ- -en
'When would Jussi have written a novel?'
In (13b), the morphological requirements of the participial head are satisfied by head movement of V to Prtc. This leads to VO order. However, (13c) is also an option. Here, the whole VP has moved to SpecPrtcP. The verb stem is then adjacent to the head Prtc, and the two can form a morphological unit, following Kayne and Julien, giving rise to OV order.
Roland Hinterhölzl, building on his previous work, presented further evidence in support of a Kaynean analysis of OV order in West Germanic languages, deriving OV by movement of various elements to the left. He argued that evidence from infinitivals militates against a Haiderian analysis: for instance, arguments precede the infinitival marker, indicating that they have left the VP (cf. (6-7) above). IPP (infinitivus pro participio) constructions from West Flemish and Afrikaans provide evidence that the infinitival marker is outside VP, not affixed to V as Haider 1993 has argued. Hinterhölzl furthermore extended the analysis of Kayne 1998 in taking English VO order to be the result of further V or VP movement to the left, quite generally.
Tarald Taraldsen presented a new analysis of verb-particle constructions. He showed problems for accounts based on head movement, and took up the complex patterns of variation among the various VO languages (as discussed in Taraldsen 1983, 1991, Svenonius 1994, Haider 1997a). Variation in particle placement within the same language is the result of a lexical ambiguity: the particle can either be the realization of a (lower) lexical head, P, or of a (higher) functional head, Prt. Taraldsen showed that cross-linguistic variation in the VO Germanic languages matches patterns observable in the OV languages, where for example Dutch patterns with Norwegian, and Afrikaans with Swedish. Taraldsen argued that the difference between VO and OV languages is the result of VP fronting in the former. As in Kayne's recent work, there is no LF movement at all; both VO and OV languages involve wholesale leftward evacuation of the various elements in the VP (cf. Hinterhölzl's proposal), but VO languages require an extra step, as in Haider's work.
Sjef Barbiers presented data on nominal and clausal complements, extending his program, developed in Barbiers 1995, for a semantically motivated syntax (cf. the summary and review of that work in GLOT 3.1:10-13). He started from the old observation that the surface distribution of DPs and CPs is different; for example, DPs precede the verb in Dutch, while CPs follow; differences are also observable in English. This has long been taken to be a surface distinction only, the two types of arguments being base-generated in the same position (in each language). Barbiers shows problems with such accounts and argues that in fact the DP and CP complements of verbs like believe are not generated in the same position, and that they have different interpretations. However, he unifies the base word order of English and Dutch, providing an argument along the way for the contention that OV is the basic order and VO is derived, as in Haider's theory.
Finally, Michael Brody presented a sketch of his theory, Perfect Syntax, and the distinctions between it and the Minimalist Program. In the Minimalist Program, syntax is perfect except to the extent that imperfection is forced on it by external considerations; "Move" is such an imperfection. Economy is a way to choose the least imperfect from among a variety of imperfect derivations (or representations). Brody's system, however, hearkens back to the more traditional stance that grammar involves the interaction of 'perfect' subsystems; no constraints are violable, and there is no theory of economy. Furthermore, there are no non-interpretable features, and representations are not duplicated by derivations (cf. Brody 1995, 1997a, to appear a, to appear b).
The theory of phrase structure that Brody develops within this more general program is Mirror Theory (cf. Brody 1997b, c). The two major claims of Mirror Theory are that [i] there are no phrasal projections, and [ii] complementation structure encodes morphology directly; thus, Y is (the head of) a complement of X only if Y and X form a morphological unit. This means that, for example, a main verb cannot be (the head of) the complement of an auxiliary; instead, Brody argues, it is part of the specifier. This has the result that the 'extended word' (cf. the extended projection of Grimshaw) cannot be restricted to a path of right branches, but must be able to continue through specifiers. It also provides a motivation for the kind of VP fronting proposed by Hinterhölzl and Kayne and the various others mentioned above; on the other accounts, a strong feature has to be stipulated to force VP fronting; on Brody's account, the VP is licensed in a specifier position, as a part of the extended word.
The conference was an unqualified success, and a volume is in the works. On behalf of the organizers I would like to thank the participants for making it such a stimulating event. I encourage everyone to keep an eye out for the announcement of another great workshop under the Midnight Sun next year (tentatively, the topic might possibly perhaps maybe be adverbs). It's worth the trip.
References
Barbiers, Sjef. 1995. The Syntax of Interpretation. PhD dissertation, Rijksuniversiteit te Leiden. HIL dissertations 14, Holland Academic Graphics, The Hague.
Borer, Hagit. 1994. 'On the projection of arguments,' in Functional Projections, ed. by Elena Benedicto and Jeff Runner, 19-47. GLSA, University of Massachusetts, Amherst, Ma.
Brody, Michael. 1995. Lexico-Logical Form: A Radically Minimalist Theory. MIT Press, Cambridge, Ma.
Brody, Michael. 1997a. 'Towards perfect chains,' in Handbook of Syntax, ed. by Liliane Haegeman. Kluwer, Dordrecht.
Brody, Michael. 1997b. 'Mirror theory.' UCL Working Papers in Linguistics 9, University College London.
Brody, Michael. 1997c. 'Mirror Theory.' (revised version) Ms. University College London.
Brody, Michael.To appear a. Review of the Minimalist Program. Mind and Language.
Brody, Michael. To appear b. 'Phrase structure and projection.' Linguistic Inquiry.
Carnie, Andrew. 1995. Non-Verbal Predication and Head-Movement. PhD dissertation, MIT, Cambridge, Ma.
Chomsky, Noam. 1981. Lectures on Government and Binding. Foris, Dordrecht.
Chomsky, Noam. 1995. A Minimalist Program for Linguistic Theory. MIT Press, Cambridge, Ma.
Dikken, Marcel den. 1995. 'Extraposition as intraposition, and the syntax of English tag questions.' Ms. Vrije Universiteit Amsterdam/HIL.
Fukui, Naoki. 1993. 'Parameters and optionality.' Linguistic Inquiry 24.3:399-420.
Fukui, Naoki, and Yuji Takano. 1998. 'Symmetry in syntax: Merge and Demerge.' Journal of East Asian Linguistics 7:27-86.
Haegeman, Liliane, and Raffaella Zanuttini. 1996. Parameters and Functional Heads: Essays in Comparative Syntax, ed. by Adriana Belletti and Luigi Rizzi, 117-180. Oxford University Press, Oxford.
Haider, Hubert. 1992. 'Branching & discharge,' Arbeitspapiere des SFB 340 23, University of Stuttgart; revised version in Lexical Structure, ed. by Peter Coopmans and Jane Grimshaw.
Haider, Hubert. 1993. Deutsche Syntax, generativ. Vorstudien zur Theorien einer projektiven Grammatik. Narr, Tübingen.
Haider, Hubert. 1996. 'Extraposition,' in Henk van Riemsdijk and Norbert Corver (eds.), Proceedings of the 1995 Tilburg Conference on Rightward Movement. De Gruyter, Berlin.
Haider, Hubert. 1995. 'The Basic Branching Conjecture,' in Studies on Phrase Structure and Economy, Working Papers of the SFB 340 70, University of Stuttgart.
Haider, Hubert. 1997a. 'Precedence among predicates.' The Journal of Comparative Germanic Linguistics 1:3-41.
Haider, Hubert. 1997b. 'Projective economy,' in German: Syntactic Problems, Problematic Syntax, ed. by Werner Abraham and Elly van Gelderen, 83-103. Max Niemeyer, Tübingen.
Henry, Alison. 1995. Belfast English and Standard English. Oxford University Press, Oxford.
Hinterhölzl, Roland. 1997a. 'A VO-based approach to verb raising,' in Proceedings of the North East Linguistic Society 27, ed. by Kiyomi Kusumoto, 187-202. GLSA, University of Massachusetts, Amherst, Ma.
Hinterhölzl, Roland. 1997b. 'An XP-movement account of restructuring.' Ms. University of Southern California.
Holmberg, Anders. 1986. Word order and syntactic features. PhD dissertation, University of Stockholm.
Hróarsdóttir, Thorbjörg. 1996. 'The decline of OV word order in the Icelandic VP: A diachronic study.' Working Papers in Scandinavian Syntax 57:92-141.
Hróarsdóttir, Thorbjörg. 1997. 'Argument movements in the Older Icelandic VP.' Ms. University of Tromsø.
Johnson, Kyle. 1991. 'Object positions.' Natural Language & Linguistic Theory 9.4:577-636.
Jónsson, Jóhannes Gísli. 1996: Clausal Architecture and Case in Icelandic. PhD dissertation, University of Massachusetts, Amherst, Ma.
Josefsson, Gunlög. 1992. 'Object shift and weak pronominals in Swedish.' Working Papers in Scandinavian Syntax 49:59-94.
Julien, Marit. 1997. 'On syntax and morphological structure.' Ms. University of Tromsø.
Kayne, Richard. 1994. The Antisymmetry of Syntax. MIT Press, Cambridge, Ma.
Kayne, Richard. 1998. 'Overt vs. covert movement.' To appear in Syntax 1.2.
Kiparsky, Paul. 1994. 'The shift to head-initial VP in Germanic.' Ms. Stanford University.
Koopman, Hilda. 1984. The Syntax of Verbs. Foris, Dordrecht.
Koster, Jan. 1994. 'Predication incorporation and the word order of Dutch,' in Paths Toward Universal Grammar: Studies in Honor of Richard S. Kayne, ed. by Guglielmo Cinque, Jan Koster, Jean-Yves Pollock, Luigi Rizzi, and Raffaella Zanuttini, 255-76. Georgetown University Press, Washington, D.C.
Ladusaw, William. 1992. 'Expressing negation,' in SALT II: Proceedings of the conference on Semantics and Linguistic Theory 2, ed. by C. Barker and D. Dowty, The Ohio State University, 237-59.
Larson, Richard. 1988. 'On the double object construction.' Linguistic Inquiry 19.3:335-391.
Pearson, Matt. 1998. 'Rightward object shift and the syntax of adverbs,' in The Structure of Malagasy vol II., ed. by I. Paul (UCLA Occasional Papers in Linguistics 20). UCLA, Los Angeles, Ca.
Pintzuk, Susan. 1991. Phrase Structures in Competition: Variation and Change in Old English Word Order. PhD dissertation, University of Pennsylvania.
Pintzuk, Susan. 1997. 'From OV to VO in the history of English,' to appear in Dufresne, Monique, and Fernande Dupuis, eds. Papers from the Fourth Diachronic Generative Syntax Conference, Montreal, October 1995.
Roberts, Ian. 1997. 'Directionality and word order change in the history of English,' in Parameters of Morphosyntactic Change, ed. by Ans van Kemenade and Nigel Vincent, 397-426. Cambridge University Press, Cambridge.
Runner, Jeffrey. 1995. Noun Phrase Licensing and Interpretation. PhD dissertation, University of Massachusetts, Amherst, Ma.
Rögnvaldsson, Eiríkur. 1987: 'On word order in Icelandic,' in Proceedings of the Seventh Biennial Conference of Teachers of Scandinavian Studies in Great Britain and Northern Ireland, 33-49, University College London.
Saito, Mamoru, and Naoki Fukui. To appear. 'Order in phrase structure and movement.' Linguistic Inquiry 29.3.
Stowell, Tim. 1981. Origins of Phrase Structure. PhD dissertation, MIT. Cambridge, Ma.
Svenonius, Peter. 1994: Dependent Nexus: Subordinate Predication Structures in English and the Scandinavian Languages. PhD dissertation, UCSC, Santa Cruz, Ca.
Taraldsen, Knut Tarald. 1983. Parametric Variation in Phrase Structure: A Case Study. PhD dissertation, University of Tromsø.
Taraldsen, Knut Tarald. 1991. 'A directionality parameter for subject-object linking,' in Principles and Parameters in Comparative Grammar, ed. by Robert Freidin, 219-268. MIT Press, Cambridge, Ma.
Travis, Lisa. 1984. Parameters and Effects of Word Order Variation. PhD dissertation, MIT. Cambridge, Ma.
Zwart, C. Jan-Wouter. 1993. Dutch Syntax: A Minimalist Approach. PhD dissertation, University of Groningen. Grodil, Groningen.
Zwart, C. Jan-Wouter. 1997. Morphosyntax of Verb Movement. A Minimalist Approach to the Syntax of Dutch. Kluwer, Dordrecht.
Return to Projects page
Return to Peter Svenonius' Home Page