View of Vol. 12 No. 1 (2021): Syntax, semantics and acquisition: In honor of Hans Petter Helland

(1)

Oslo Studies in Language 12 (1) / 2021

Christine Meklenborg, Terje Lohndal &

Kathrine Asla Østby (eds.)

Syntax, semantics and acquisition:

In honor of Hans Petter Helland

(2)

Oslo Studies in Language

General editors: Atle Grønn and Dag Haug

Issue editors:

Christine Meklenborg University of Oslo

Terje Lohndal NTNU – Norwegian University of Science and Technology

& UiT The Arctic University of Norway Kathrine Asla Østby University of Oslo

(3)

Oslo Studies in Language 12 (1) / 2021

Christine Meklenborg, Terje Lohndal &

Kathrine Asla Østby (eds.)

Syntax, semantics and acquisition:

In honor of Hans Petter Helland

(4)

Oslo Studies in Language, 12(1), 2021.

Christine Meklenborg, Terje Lohndal & Kathrine Asla Østby (eds.):

Syntax, semantics and acquisition: In honor of Hans Petter Helland Oslo, University of Oslo

ISSN 1890-9639 / ISBN 978-82-91398-12-9

Set in MS Word & L^ATEX fonts Gentium and Linux Libertine by Eirik Hvidsten.

Cover design by UniPub publishing house & Atle Grønn.

Printed by Print House AS from camera-ready copy supplied by the editors.

http://www.journals.uio.no/osla

(5)

introduction

CHRISTINE MEKLENBORG¹, TERJE LOHNDAL^2.3 AND KATHRINE ASLA ØSTBY¹

¹University of Oslo

2Norwegian University of Science and Technology

3The Arctic University of Norway

On June 29, 2021, Hans Petter Helland turns 60. We would like to take this oppor- tunity to celebrate Hans Petter’s accomplishments to date, and we hope this Fest- schrift will provide additional inspiration for him to continue doing research for many years to come.

Hans Petter studied German and French at the University of Oslo from 1982 to 1987, where he graduated with an MA degree in French linguistics. The title of his dissertation was Les constructions passives être + Vp passé et leur double inter- prétation, which got an outstanding grade. In 1987, he also spent a semester in Strasbourg, obtaining a maîtrise de Lettres modernes. He qualified as a teacher in 1988, and he also spent a semester studying Norwegian as a second language.

After teaching French at the University of Oslo, Hans Petter applied for a scholarship from the research council to do his PhD in 1989. The project description is less than one page, which says a lot about how things have changed in academia since then. The application says that he would study central areas re- lating to the syntax–semantics interface, but beyond that, it does not say much about the specific topics. The application was successful, and on January 1, 1990, Hans Petter embarked on his PhD. In late August 1993, he defended his PhD dissertation, and the title was: Une description sémantique et pragmatique de structures temporelles en français moderne. Application particulière au futur simple et au futur pé- riphrastique. His opponents were Professors Georges Kleiber (Strasbourg) and Henning Nølke (Aarhus). Tense and especially the future tense has always been a particularly beloved topic for Hans Petter.

After finishing his PhD, Hans Petter was hired as an Assistant Professor at the University of Trondheim (since 1996 NTNU – Norwegian University of Science and Technology) in 1992. After he defended his PhD, he was promoted to Associ- ate Professor of French linguistics, a position that he held until 1999. On May 1 that year, he was promoted to full Professor of French linguistics. However, soon thereafter, Hans Petter left Trondheim and began as a Professor of French

(8)

[ii] MEKLENBORG, LOHNDAL & ØSTBY

linguistics at the University of Oslo (August 1, 1999). Since then, Oslo has been his main academic home, but he has also had adjunct professorships at NTNU in Trondheim (2010–2016), and at the University of Bergen (2020–2021). He was the director of teaching at l’Office franco-norvégien d’échanges et de coopération (OFNEC) in Caen, France, from 2011 to 2012. He has also been the chair of the Norwegian board for OFNEC from 2008 to 2011. From 2013 to 2020 he was an active member of the department board in Oslo with a keen interest in university politics and the particular challenges that the foreign languages face.

Hans Petter started out with work on passives and then tense, and he has re- turned to both topics several times. For instance, in 2002 he published the book Le passif périphrastique en français moderne. Later he has also done extensive work on secondary predication, possessives, adjuncts, and control. An interest which has become more prominent in recent years, involves the acquisition of second and third languages, including both natural and instructed language acquisition.

All his research is heavily informed by theoretical developments within formal grammar, and he has played a vital role in disseminating insights from this framework to the French academic community in Norway. His textbook, Ny fransk grammatikk: morfologi, syntaks og semantikk, has been widely used. Hans Pet- ter has also edited multiple special issues of journals, and two anthologies: Chal- lenging Clitics (together with Christine Meklenborg Salvesen) published on John Benjamins in 2013, and Affaire(s) de grammaire (also with Christine Meklenborg Salvesen), published on Novus in 2014.

A lot of students have benefitted from Hans Petter’s insights, valuable advice, dedication, and commitment. He has supervised a huge number of MA theses and numerous PhD dissertations. In addition, he has served on a range of assessment committees nationally and internationally.

The contributions in this Festschrift relate in various ways to topics that Hans Petter has worked on. They have all been peer-reviewed. In addition to the editors and contributors, the following scholars have contributed to the peer re- view: Gisle Andersen, Helene N. Andreassen, Camilla Bardel, Kari Kinn, Hanne Korzen and Michael T. Putnam. We are very grateful to all of you for your help.

In addition, we would like to thank Eirik Hvidsten for his terrific editorial assis- tance in the last stages of this project.

On behalf of all contributors and reviewers: Hans Petter, we offer you our warmest congratulations on your 60^th birthday. Please accept this Festschrift as a token of our appreciation.

Oslo and Trondheim, June 14, 2021

Christine Meklenborg Terje Lohndal Kathrine Asla Østby

(9)

Meklenborg, Lohndal & Østby (eds) Syntax, semantics and acquisition: In honor of Hans Petter Helland, Oslo Studies in Language 12(1), 2021 (ISSN 1890-9639/ISBN 978-82-91398-12-9) http://www.journals.uio.no/osla

tabula gratulatoria

Helene N. Andreassen, Tromsø Kjerstin Aukrust, Oslo Bergljot Behrens, Oslo Ingeborg Misje Bergem, Oslo Emerentze Bergsland, Halden

Merete Birkelund, Århus Annlaug Bjørsnøs, Trondheim

Guro Busterud, Oslo Hélène Celdran, Oslo Myriam Coco, Bergen Anne Dahl, Trondheim Kristin Bjørneboe Eide, Oslo

Kristine G. Eide, Oslo Gunn Enli, Oslo

Cathrine Fabricius-Hansen, Oslo Kjersti Fløttum, Bergen Thorstein Fretheim, Trondheim

Rebeca Schumacher Fuão, Oslo Jan Terje Faarlund, Oslo Øyvind Gjerstad, Bergen Anje Müller Gjesdal, Bergen

Karen Gammelgaard, Oslo Atle Grønn, Oslo Karin Gundersen, Oslo Marianne Hobæk Haff, Oslo

(10)

Hilde Hasselgård, Oslo Dag Tryge Truslew Haug, Oslo Marius Warholm Haugen, Trondheim

Frode Helland, Oslo Turid Henriksen, Oslo

Karin Holter, Oslo Eirik Hvidsten, Trondheim

Tabea Ihsane, Genève Karina Kleiva, Bærum

Espen Klævik-Pettersen, Kristiansand Trude Kolderup, Trondheim

Hanne Korzen, København Sonia Lagerwall, Bergen

Pierre Larrivée, Caen Lisbeth Larsen, Oslo

Kjersti Faldet Listhaug, Trondheim Terje Lohndal, Trondheim

Francine Girard Lomheim, Kristiansand Svein Erling Lorås, Oslo

Tina Malen B. Lybekk, Oslo Chantal Lyche, Oslo

Helge Lødrup, Oslo Gro Bjørnerud Mo, Oslo

Christine Meklenborg Nilsen, Eidsvoll Nils Axel Nissen, Oslo

Sigbjørn Næss, Bærum Henning Nølke, Århus Francis Badiang Oloko, Bergen

Anneliese Pitz, Oslo

(11)

OSLa volume 12(1), 2021

Priscilla Ringrose, Trondheim Yves Roberge, Toronto Maximino Jesus Ruiz Rufino, Oslo

Antin Fougner Rydning, Oslo Anne Birgitte Rønning, Oslo

Trond Kruke Salberg, Oslo Kirsti Sellevold, Oslo Kathrine Sinding-Larsen, Oslo

Ingse Skattum, Oslo Mathilde Skoie, Oslo Alexandra Anna Spalek, Oslo

Guri Bordal Steien, Oslo Nelly Foucher Stenkløv, Trondheim

Guri Bordal Steien, Oslo Kjell Johan Sæbø, Oslo Martine Sørensen, Oslo Jean Louis Tarrou, Bærum Solveig Schult Ulriksen, Oslo

Geir Uvsløkk, Oslo Sophie Vauclin, Trondheim Cecilie Weiss-Andersen, Drammen

Eirik Welo, Oslo

Sunniva Whittaker, Kristiansand Anne-Kathrine Woldsnes, Oslo

Helle Waahlberg, Paris Kathrine Asla Østby, Oslo

Tor A. Åfarli, Halsa

(12)

(13)

Meklenborg, Lohndal & Østby (eds) Syntax, semantics and acquisition: In honor of Hans Petter Helland, Oslo Studies in Language 12(1), 2021. 1-20. (ISSN 1890-9639/ISBN 978-82-91398-12-9)

a cluster of changes: norwegian word order

JAN TERJE FAARLUND University of Oslo

abstract

Among the syntactic changes that can be observed in the transition from Old Norwegian to Modern Norwegian are the following word order changes: loss of OV order, object shift from a VP containing a verb (non-finite verb or any verb in a subordinate clause), preverbal preposition, and topicalization of a bare head. The fact that these changes all seem to occur at the same time is not accidental. Old Norwegian was arguably a VO language, like Modern Nor- wegian, but unlike Modern Norwegian, OV order was also possible. It can be shown that it is possible to derive sentences with object shift with a verb in VP, sentences with preverbal prepositions, and with topicalized heads only from an OV structure. Therefore, when the OV order was no longer available, the other three structures could no longer be derived.

[1] introduction

Like biological evolution, grammatical change has no purpose or predetermined direction. Genetic mutation is random, and so is grammatical innovation. Learn- ers and speakers innovate continuously, but which innovations they make, which ones spread in the community, and which ones catch on and become changes in the grammar, depend on various unpredictable circumstances. There is no theory of change predicting future changes. Grammatical change is, however, not totally random. There is an upper bound set by Universal Grammar – understood as the theory of possible grammars; no change may yield a result which is not a possible grammar (Lightfoot 1979, pg. 141). The specification of a possible grammar includes the specification of structural and hierarchical relations within the sentence. Those relations are the preconditions of historical changes taking place in the grammar; a change from the stage La to a later stage Lb of the language L, besides being limited by the human language faculty (Uni- versal Grammar), also depends on the structural properties of La. Thus a grammatical change may be triggered by or prevented by the grammar of the language at the point in time where the change takes place.

In this paper I will show how one particular change in the history of Norwe- gian had important repurcussions in other parts of the grammar. A word order

(14)

[2] FAARLUND

change which took place at the transition from Old Norwegian to Early Modern Norwegian, presumably during the early 16^th century (but like all other grammatical changes, this change did not take place at the same time in all parts of Norway), led with necessity to three other changes.

Old Norwegian is recorded in manuscripts from the 12th through the 15th century, representing the language spoken as far back as the 9^th century. There are no well documented syntactic contrasts between the two major dialects of Old Norse, Old Icelandic and Old Norwegian (Faarlund 2004a, pg. 2). In the 16th century the Norwegian dialects underwent dramatic changes at many levels of the grammar, as well as in the vocabulary, which removed the language signifi- cantly from Old Norwegian, and from later stages of the other West Nordic dialects, Icelandic and Faroese. At the same time, Norway was subjected to the Dan- ish Crown and ruled form the Danish capital Copenhagen, with the consequence that the Norwegian language was replaced by Danish as the official and literary language of the country. The Norwegian dialects survived as a spoken vernacu- lar, and a new standard language was established based on those dialects after the independence from Denmark in the 19^th century.

Modern Norwegian and Old Norwegian share several syntactic patterns and operations, while at the same time there are clear syntactic contrasts between the two stages, as we will see. Most notable perhaps among the shared patterns, are movement of the finite verb via T to C in main clauses, and A’ movement to SpecCP, which in combination create the familiar ‘Verb Second’ pattern at both stages of the language, see the Old Norwegian examples in (1a,b) and Modern Norwegian ones in (1c,d). Another shared pattern is a general head-complement order (VO), as in (1a) and (1d), where a non-finite verb precedes its complement.

(1) a. Hon scal alrigh fa the kona she shall never get that cow-DEF

‘She will never have that cow.’ (DN XI.151)

b. nú gerir maðr langskip í heraði now makes man long-ship in district

‘Now a man makes a long ship in the district.’ (ML 47.10) c. Ho kom ikkje heim

she came not home

‘She did not come home.’

d. No har dei kjøpt ein ny båt now have they bought a new boat ‘Now they have bought a new boat.’

(15)

A CLUSTER OF CHANGES: NORWEGIAN WORD ORDER [3]

OSLa volume 12(1), 2021

On the other hand, Old Norwegian has the character of a ‘free word order’ language, allowing several word order patterns which no longer exist in Modern Norwegian, a ‘fixed word order’ language (Laake 2018). Those word order patterns include, first, the possibility of an object-verb order, as an alternative to the verb-object order in (1a), as in (2a); second, object shift across a non-finite verb in VP, (3a); third, shifting of a preposition from the position in front of its complement, to a position in front of the verb, (4a); and finally, the possibility of moving a head without its complement(s) to the topic position in front of the finite verb, (5a). The corresponding Modern Norwegian sentences in (2-5b) are all ungrammatical. It will be the claim of this paper that these changes did not just coincidentally occur during the same period in the history of Norwegian.

The first change mentioned above, the loss of OV order, can be explained inde- pendently. The other three constructions that were lost, were dependent on the possibility of OV, so when OV disappeared, the constructions in (3-5) could no longer be generated.

(2) OV order

a. Hælldr mindi hann þessa gripina kiosa en allt fee rather would he these treasures.DEF choose than all property annat

other

‘He would rather choose those treasures than all the other property.’ (Olavss.5:218639)

b. *Han ville heller desse skattane velja he would rather these treasures.DEF choose

‘He would rather choose these treasures.’

(3) Object Shift

a. þurfu þæir þat ekki sægia¹ needed they that not say

‘They did not need to say that.’ (DN IV.100) b. *Dei ville det ikkje seia

they would that not say

‘They did not want to say that.’

[1] This sentence starts with the finite verb, since the topic position is empty, another option no longer available in Modern Norwegian (except in questions).

(16)

[4] FAARLUND

(4) Preverbal preposition

a. eigi má ek hug mínum á koma fleiri þá hluti not may I mind.DAT mine on come more those things.ACC

‘I cannot think of more of those things.’ (Kgs 38.15) b. *Eg kan ikkje på komma fleire slike ting

I can not on come more such things ‘I cannot think of more such things.’

(5) Topicalized head

a. en binda ma maðr þiof but bind may man thief

‘But a man may bind a thief.’ (ML 64.14) b. *Binde kan ein mann ein tjuv

bind may a man a thief

‘A man may bind a thief.’

In Section 2 the OV order of Old Norwegian, as illustrated in (2a), will be further examined, and its eventual demise in Modern Norwegian will be accounted for.

In Section 3 I will show that the other three changes, illustrated in (3) – (5), nec- essarily follow from the loss of the OV order.

[2] head final vp

Modern Norwegian is a consistent head initial language: the complement follows the head within VP (6a). No other order is possible in the contemporary language. However, in main clauses a finite verb moves to C via T. In subordinate clauses the verb stays within VP, whether finite or non-finite, (6b).

(6) a. Der vil du sjå ein mann med dress there will you see a man with suit

‘There you will see a man in a suit.’

b. dersom eg aldri møter henne igjen if I never meet her again

‘if I never meet her again.’

The oldest documents in any Scandinavian language (Ancient Nordic, 3^rd to 6^th century) show an OV pattern, as do other older Indo-European languages (Faarlund 2002, 2004b). The change from OV to VO took place during the Old Norwegian period, and it was more or less completed by the time of early Modern Norwegian (16^th century). Thus in Old Norwegian we find both patterns, OV as in

(17)

(7), and VO as in (8). Note that the ‘O’ stands for not only object, but a complement of any category.

(7) a. scall thetta breff maackt haffwa om allder ok eeffwe shall this letter power have in age and eternity

‘This letter shall remain valid for ever.’ (DN II.897) b. hann man þer þriðiungs unna af feno

he will you-DAT third grant of money.DEF

‘He will grant you a third of the money.’ (Olavss. 5:218640) c. meer wilium þydher kungøra ath …

we will you(pl)-DAT notify that …

‘We want to notify you that …’ (DN V.590)

d. Haralldr grænske lætr asto hæim fara til faður sins Harald Grenski let Asta home go to father her ‘Harald Grenski let Asta go home to her father.’ (Olavss.

4:218622)

e. en ælligar man þer til giævo snuazt but otherwise will you.DAT to luck turn-REFL

‘But otherwise it will turn into good luck for you.’ (Olavss.

3:218613)

(8) a. þar mantu sia mann æinn með slicum bunaðe sem ec hævi there will-you see man one with such dress as I have

‘There you will see a man dressed the way I am.’

(Olavss.3:218605)

b. en þess vil ec biðia yðr at þer bloteð mic æigi but that will I ask you(pl) that you worship me not

‘But I will ask you not to worship me.’ (Olavss. 2:218586) c. nu ef þu villt fara or lande

now if you want go out-of country

‘Now if you want to leave the country.’ (Olavss. 1:219265)

Which order was basic and which one was derived at any given point in the history (and dialect geography) of Old Norwegian, is a question of some controversy (Kossuth 1978, Sigurðsson 1988, Rögnvaldsson 1996, Hróarsdóttir 2000, Faarlund 2004a). Sundquist (2006) presents statistics showing the relative frequency of verb final order in Old Norwegian during the period 1250 – 1525, showing that in the earliest period (end of 13^th century), there is 42% OV, and in the early 16^th century this order is down to 6%. So throughout the documented history of Old Norwegian, VO order has been the most frequent one. Therefore, at some point in the history of Scandinavian the basic structure of the VP must have changed

(18)

[6] FAARLUND

from head final (OV) to head initial (VO). Even though structures like those in (7) are easy to find in Old Norwegian texts, there is good reason to posit head initial order as the basic one from early on in Old Norwegian; besides the fact that this is by far the most frequent order (by itself not a proof, though), it is also the least marked one (by any reasonable concept of markedness) (Christoffersen 1993, pg.

305f.). More importantly, even though the verb follows its complement, it may not be absolutely clause final, as can be seen in most of the examples here; parts of the VP may remain behind the verb. For example, in (7b) the direct object is actually þriðiungs af feno, and in (7d) the directional adverbial is hæim til faður sins (on this kind of separation of a head from its complement, see Section 3.2). This would be unexpected if the basic structure was verb final, and it gives us reason to posit an optional rule moving (parts of) the complement of the verb and left- adjoining it to VP. I will refer to this movement as ‘scrambling’. The relevant part of (7a) has the (simplified) structure in (9).²

(9) C’

 C TP

scall  DP T’

thetta breff 

T VP

scall 

DP VP

maackt 

DP V’

thetta breff

V DP

haffwa maackt

Besides the general rules moving the finite verb from T to C, and the subject from SpecVP to SpecTP, the complement of V, maackt, left-adjoines to VP. Eventually, this was seen as a marked order, becoming less and less frequent, until the rule of scrambling was lost, and OV was lost from the grammar of Modern Norwegian.

[3] structures fed by scrambling

The three other changes which make up the cluster of changes from Old to Mod- ern Norwegian, all involve the loss of a specific word order pattern. They are

[2] Actually, the auxiliary scal is not first merged in T; it heads its own V-projection, which takes the VP [thett breff maackt haffwa] as its complement, and from where it is raised to T.

(19)

object shift across non-finite verbs, preverbal prepositions, and movement of naked heads to SpecCP, as illustrated above in (3), (4), and (5), respectively. What those constructions have in common is that they can be derived only on the basis of an OV structure. After the rule of scrambling was lost, they could no longer be derived by the grammar of Norwegian.

[3.1] Object Shift

The Scandinavian languages, including Modern Norwegian, have an operation known as – among other things – ‘object shift’, whereby a pronominal object (direct or indirect) moves across a sentence adverbial to the position immediately preceding that adverbial, following the finite verb or the subject (if it is not topicalized) in main clauses. This pattern exists both in Old Norwegian, as in (13), and in the Modern Scandinavian languages, including Modern Norwegian, (10).³ There are two important conditions for this movement to take place: the pronoun has to be unstressed, and its position has to be immediately following the sentence adverbial before the movement can take place. The second condition means essentially that a sentence adverbial and an unstressed pronoun change places.

(10) a. Eg såg henne ikkje I saw her not

‘I didn’t see her.’

b. Derfor seier dei det ikkje therefore say they it not

‘Therefore they don’t tell.’

c. Eg gav dei ikkje pengar I gave them not money

‘I didn’t give them money.’

The second condition, known as ‘Holmberg’s Generalization’ was first discussed in generative terms by Holmberg (1986, ch. 6).⁴ In a later work, he offers the following formulation: ‘Object Shift cannot apply across a phonologically visible category asymmetrically c-commanding the object position except adjuncts’.

(Holmberg 1999, pg. 15). In the spirit of Kayne (1994), ‘asymmetrically c-commanding’ means simply ‘preceding’, and the adjuncts in question are sentence adverbials.

A sentence adverbial is a non-argument phrase which typically modifies the

[3] Object shift may not be obligatory in certain Scandinavian varieties. But it seems to be close to obligatory in the Central and Southern parts of East Norway, and in standard Bokmål, see Faarlund (1977, 2019, pg.

199f) and Munch (2013).

[4] This word order pattern and its conditions have, however, been familiar to traditional Scandinavian grammarians since long before the days of generative grammar.

(20)

[8] FAARLUND

clause, or parts of it, depending on its scope properties. In Norwegian, the sentence adverbial follows the finite verb on the surface in main clauses and precedes it in subordinate clauses. It may precede or follow the subject. The negation behaves syntactically like a sentence adverbial, so for our purpose there is no reason to operate with a separate NEG node (Faarlund 2004a, pg. 225). A sentence adverbial may consist of a simple word, or it may be modified or conjoined.

Object shift takes place regardless of the length or complexity of the adverbial.

The crucial element of Holmberg’s Generalization is the requirement that the object at a certain stage of the derivation be adjacent to the adverbial in order for it to operate. That can only happen after the subject has moved from its base position in SpecVP to SpecTP. Next, the verb has to leave VP for object shift to take place. This happens to finite verbs in main clauses, where they move to C via T. Non-finite verbs remain in VP, thus blocking object shift, as in (11a,b). In subordinate clauses even the finite verb remains within VP, blocking object shift, as in (11c). Consider the contrast between (10) and (11).

(11) a. *Eg har henne ikkje sett I have her not seen ‘I have not seen her.’

b. *Dei treng det ikkje seia.

they need it not say ‘They need not say it.’

c. *dersom du henne ikkje gav pengar if you her not gave money ‘if you did not give her money’

Not only non-finite verbs, but also indirect objects and prepositions block object shift.

(12) a. *Han fortalde det ikkje foreldra sine he told it not parents his ‘He did not tell it to his parents.’

b. *Han snakka det ikkje om.

he talked it not about ‘He did not talk about it.’

The descriptive facts about object shift in Norwegian (and the other Mainland Scandinavian languages) are thus quite clear and can be summed up as follows:

whenever no overt material intervenes between a sentence adverbial and a light object pronoun, the pronoun moves to the left of the adverbial. The theoretical implications of this are, however, rather problematic. It is different from regular

(21)

syntactic operations. First of all, object shift violates strict cyclicity, or the Ex- tension Condition (Chomsky 1993), which basically says that the merge operation can only target the root. This is also known as the ‘no-tampering condition’

on computation, requiring that ‘Merge is invariably “to the edge” ’ (Chomsky 2008, pg. 138). Concretely, this means that once the subject has been internally merged in SpecTP, nothing can be changed in the structure below T. Further- more, object shift has no effect on binding relations (Holmberg 1999, pg. 18).

Holmberg concludes that ‘Object Shift is a stylistic rule, applying in a component of the grammar which I shall call Stylistic Syntax’ (p. 21), which includes phonological features such as [FOCUS]. He does not make it clear what the relationship is between Stylistic Syntax and the phonological component. In any case, it seems that object shift is not in the syntax. It does not obey general syntactic principles, it depends on overt phonetic adjacency without regard for copies of syntactic merge operations, and it is sensitive to prosodic features.

Turning now to Old Norwegian, it seems that it has the same pattern of object shift as Modern Norwegian.

(13) a. þæir rænti hann ækki they robbed him not

‘They did not rob him.’ (DN II.156) b. samde þeim eigi til lykta

suited them.DAT not in end

‘It did not suit them in the end.’ (DN II.99) c. ek lener þik ey hws

I lend you not house

‘I won’t lend you a house.’ (DN I.564)

However, it turns out that a pronoun can also shift out of a VP which contains the governing verb.

(14) a. þurfu þæir þat ekki sægia need they that not say

‘They do not need to say it.’ (DN IV.100) b. getom weer thet æy giort

get we that not done

‘We will not get that done.’ (DN V.8099) c. Sigrid hafde þet goz ekki fenghet⁵ Sigrid had that property not obtained

‘Sigrid had not obtained that property.’ (DN I.201)

[5] As shown by (14c), a definite DP may also undergo object shift, although this is quite rare.

(22)

[10] FAARLUND

This looks like a violation of Holmberg’s Generalization, which, being a ‘generalization’, should work for older Scandinavian as well as for the modern languages.

The question is then whether the sentences in (13) are instances of a different phenomenon, or whether they can be subsumed under the rule of object shift.

Remember that a basic condition on object shift is immediate adjacency of the object to the sentence adverbial. Now, let us assume that the sentences in (14) are not derived from a VO structure, but from the older OV structure, still present in Old Norwegian, see (7). In other words, (14a) is not derived from a structure like (15a), but from (15b), both of which are, as far as we know, grammatical at the relevant stage of Old Norwegian.

(15) a. þurfu þæir ekki sægia þat need they not say that ‘They need not say that’.

b. þurfu þæir ekki þat sægia need they not that say

I will assume that (15b) is derived synchronically from the structure underlying (15a) by a rule of scrambling, moving the complement out of VP and left-adjoining it to VP, thus bringing it in immediate contact with the sentence adverbial.

The two movement operations and their interaction are illustrated in (16) (next page).⁶ Modern Norwegian, lacking OV as a possible order at any stage in the derivation, has no way of shifting the object when the verb is still present within VP, see (11). But in principle, the rule of object shift, or Holmberg’s Generaliza- tion, is the same in Old and Modern Norwegian.

[6] The derivation of the auxiliary þurfu is as that of scal in (9), see footnote 1.

(23)

(16) C’

 

 ^C  ^TP

  ^þurfu

DP T’

þæir 

T VP

þurfu 

DP VP



DP VP

þat 

Adverb VP

ekki 

DP VP

          þat

Object Shift DP V’

þæir  V DP

sægia þat

Scrambling [3.2] Preverbal prepositions

Both Old and Modern Norwegian have head initial prepositional phrases, as ty- pologically expected in VO languages.

(17) a. Hakon konongr hafðe gort brullaup sitt til drotningarennar fru Hakon king had made wedding his to queen-DEF-GEN lady Magrettar

Margarett

‘King Hakon had prepared his wedding to the queen lady Margaret.’

(DN I.8)

b. aller broeðr skyli ser skæmta af þæssare varre minning all brothers should REFL-DAT amuse of this our memory-DAT

‘All the brothers should amuse themselves with our memories.’

(DN I.8)

c. er fallit hefir á hjarta mitt which fallen has on heart-ACC my

‘which has fallen on my heart’ (Barl 6.34)

(24)

[12] FAARLUND

In addition to this regular pattern, the complement may sometimes be extraposed and separated from the preposition. For example, there are cases where the preposition precedes the verb, while the complement of the preposition follows the verb. This pattern is completely out in Modern Norwegian.

(18) a. ok þat hefir mik til rekit svá langrar ferðar and that has me to driven so long journey-GEN

‘And that has driven me to such a long journey.’ (Laxd 58.26) b. eigi má ek hug mínum á koma fleiri þá hluti

not may I mind-DAT my on come more those things.ACC

‘I cannot think of more of those things.’ (Kgs 38.15) c. her er mikit af sagt burtreið þessara manna

here is much of said joust-DAT these men-GEN

‘Much is told here about these men’s joust.’ (FS 35, 54) d. er eigi vildi af láta heiðninni

who not wantedoff let paganism-DEF-DAT

‘who did not want to give up paganism’ (Hkr II.123.13)

Note that the complement in each example is assigned case by the preverbal preposition: genitive by til in (18a), accusative by á in (18b), and dative by af in (18c,d). A similar example is (7d), repeated here as (19). Here heim til faður sins

‘home to her father’ is one phrase, with heim as the head of the phrase in a preverbal position; the complement is another prepositional phrase, til faður sins, following the verb.

(19) Haralldr grænske lætr asto hæim fara til faður sins.

Harald Grenski let Asta home go to father her

‘Harald Grenski let Asta go home to her father.’ (Olavss. 4:218622)

Also other material than the verb may intervene between the preposition and its complement. Thus if the verb is finite and moves to T or C, the preposition and the complement may still be separate.

(20) a. hǫgg þú af tvær alnar hverju stórtré cut.IMP you off two ells-ACC each big.tree.DAT ‘Cut two ells off each big tree.’ (Laxd 220.4)

b. Snorri brá við skjótt orðsending Guðrunar Snorri startled at suddenly message-DAT Gudrun.GEN

‘Snorri got suddenly startled by the message from Gudrun.’ (Laxd 211.1)

(25)

c. þickir monnum þegar mikils um vert vascleik þeirra seems men.DAT at.once much of worth bravery-ACC their ‘The men at once thought much of their bravery.’ (FS 81, 6)

As in (18), the extraposed complement is assigned case by the severed preposition: dative by af in (20a) and by við in (20b), and accusative by um in (20c).

The existence of preverbal prepositions in Old Norwegian, and their disap- pearance in Modern Norwegian, can again be connected to scrambling.⁷By scrambling, a PP as well as a DP may move out of VP and consequently precede the verb. One example is (7e), repeated here as (21a), along with further examples.

(21) a. En ælligar man þer til giævo snuazt but otherwise will you.DAT to luck turn.REFL

‘But otherwise it will turn into good luck for you.’

(Olavss.3:218613)

b. en hann man æigi vilia af þer taka but he may not want of you take

‘But he may not want to take it from you.’ (Olavss. 5:218643) c. þa skal hann a þing foera

then shall him to court lead

‘Then he shall be taken to court.’ (ML 168.21b)

Structures such as these form the derivational basis of the patterns in (18) – (20).

The next step is extraposition of the complement of the preposition. Note that in all the examples in (18) – (20), the complement comes at the very end of the sentence. Generally, extraposition of a focused or complex DP is quite common in Old Norwegian, as can be most clearly seen in the case of extraposed subjects.⁸ (22) a. skal ok standa allar saatmale ok ol skipti

shall also stand all settlement.NOM and all divisions.NOM

‘The whole settlement and all the divisions are confirmed.’ (DN III.421)

[7] Most of the examples in (18) – (20) are from Nygaard (1906, pg. 355, 358). Examples (18a,d) and (20a,b) are Old Icelandic rather than Old Norwegian. They are included here to supplement the Old Norwegian examples. This construction type is quite infrequent in both dialects, and finding more Norwegian examples would require an extremely time consuming manual search.

[8] Theoretically, extraposition is a rather complex syntactic operation, with several formal treatments in the literature. For a recent overview, cf. Corver (2014). In this paper I use the term descriptively to refer to any movement of a DP to the right edge of the sentence, without committing myself to any specifdic formal analysis.

(26)

[14] FAARLUND

b. ok sua hafa fareth fleire þiner frender and so have fared more your friends.NOM

‘And so have several of your friends suffered.’ (DN V.423) c. kom tha fore oss skellig beuising

came then before us clear proof.NOM

‘Then clear proof was presented to us.’ (DN I.841)

The derivation of sentences such as those in (18) requires no other apparatus than what we already have in other parts of the grammar. The necessary starting point is a head final VP, as exemplified in (7) and (21). From its preverbal position, the complement of the preposition is extraposed. This is just another instance of the extraposition rule at work in (22). The derivation of the relevant part of (18d) can then be summarized as in (23).

(23) a. eigi vildi VP[láta af heiðninni]

b. Scrambling:

eigi vildi PP[af heiðninni]iVP[láta _ i ] c. Extraposition:

eigi vildi PP[af _ i ] VP[láta] heiðninnii

Due to the lack of scrambling in Modern Norwegian, structures like those in (18), (20), and (21) are now excluded. On the other hand, in some cases, a preverbal preposition has been reanalyzed as a verbal prefix and/or a verbal particle, while the original complement of the preposition has been reanalyzed as the complement of the verb. This can be illustrated by comparing (20a) to the Norwegian near-equivalent (24a), where greina is the object of hogg, and av is a particle. Such particles may also precede the object, which then still is the complement of the verb, as in (24b). When the verb has the form of a participle, the particle may be prefixed to the verb, (24c).

(24) a. Han hogg greina av he cut branch.DEF off ‘He cut the branch off.’

b. Han hogg av greina he cut off branch.DEF

‘He cut off the branch.’

c. Greina er avhoggen branch.DEF is off-cut

‘The branch has been cut off.’

For a more detailed study of this reanalysis, see Faarlund (1990, pg. 185f and

(27)

1995).

[3.3] Head in SpecCP

In Old Norwegian, in contrast to Modern Norwegian, a transitive non-finite verb may move to SpecCP alone, leaving the complement behind.⁹

(25) a. en binda ma maðr þiof but bind may man thief

‘But a man may bind a thief.’ (ML 64.14) b. styrkia skal ek hann oc hans riki strengthen shall I him and his power

‘I will make him and his power stronger.’ (ML 30.8) c. sia skolu skilrikir menn mark þeirra luta

see shall honest men mark their lot.GEN

‘Honest men shall see the mark of their own lot.’ (ML 94.14)

In (25a-c) non-finite transitive verbs are topicalized, leaving the object behind.

This looks like head movement to SpecCP, an A’ position. However, head movement to an A’ position violates an assumed general principle of generative grammar. In connection with a different kind of verb topicalization,¹⁰ Holmberg (1999, pg. 13) concludes about this possibility that ‘The most controversial aspect of [the] analysis is, probably, the notion that the verb, a head, is moved to a spec- position, namely SpecCP. But this is something that the theory very likely has to allow for anyway’. I think the ‘theory’, or whoever is responsible for it, ought to be extremely reluctant to allow for head movement to a specifier position, since it violates a very fundamental principle of grammar, which is the distinction between head movement, and phrasal movement to specifier positions, whether A or A’ positions. According to this principle, heads can only move to head positions, such as a verb moving from V to T. Therefore, each apparent

[9] On the other hand, topicalization of a full VP including the complement does not seem to be possible in Old Norwegian. The equivalent of this Modern Norwegian sentence does not seem to exist in Old Norwe- gian.

(i) Binda ein tjuv kan ein gjera bind a thief may one do (cf. (25a))

[10] The structure in question is verb topicalization in sentences such as the Swedish (ii) Kysst har jag henne inte

kissed have I her not

‘I have not kissed her.’

Contrary to Holmberg’s claim, the Modern Norwegian equivalent is ungrammatical, but it would be OK in Old Norwegian, as we have seen.

(28)

[16] FAARLUND

counterexample should be thoroughly scrutinized to see if there are other possible analyses available.

I will now demonstrate that movement to SpecCP in (25) is really phrasal movement. This again presupposes an OV structure, which accounts for the ab- sence of equivalent structures in Modern Norwegian. In Section 2, I described the OV order as the result of scrambling, which means left-adjoining the complement of V to VP, thus leaving the verb as the only element left in the original VP. This verb is then free to topicalize as a phrase by remnant movement.¹¹ The derivation of (25a) can be summarized as in (26).

(26) a. CP[má TP[maðr VP[binda þiof]]

b. Scrambling:

CP[má TP[maðr DP[þiofi] VP[binda _i ]]

c. Topicalization of remnant VP:

CP[VP[bindai] má TP[maðr DP[þiof] VP[_i ]]]]

Even prepositions may be topicalized and thus separated from their comple- ments.

(27) a. af hefir þú mik ráðit brekvísi við þik off have you me advised importunity.DAT with you

‘You have advised me not to be importunate with you.’ (Laxd 98.14) b. a þicki mer vera skugi nockur manninum

on seems me.DAT be shadow some man.DAT

‘There seems to me to be a shadow over the man’ (FS 52.48)

This looks like another instance of head movement to a specifier position. But in this case, too, it is possible to treat it as remnant movement, parallel to the cases with verbal heads in (25). We have already seen in Section 3.2 that the complement of a preposition may be extraposed and thereby separated from its head.

So let us assume that the prepositional phrase is left-adjoined to VP by scrambling, followed by extraposition of the complement. The PP now consists of a preposition alone, which then may be topicalized by remnant movement. The derivation of (27a) can be summarized as in (28).

(28) a. CP[TP[hefir þú VP[ráðit mik PP[af DP[brekvísi við þik]]

b. Scrambling:

CP[TP[hefir þú mik PP[af brekvísi við þik] iVP[ráðit _i]]]

c. Complement of P extraposed:

CP[TP[hefir þú mik PP[af _i ] VP[ráðit] DP[brekvísi við þik]i ]

[11] For similar analyses see Ott (2016) for Old Icelandic, and Salvesen (2011) for Old French.

(29)

d. Topicalization of remnant PP:

CP[PP[afi] TP[hefir þú mik PP[_i] VP[ráðit]]] DP[brekvísi við þik]]

The analysis presented in this section, involving scrambling and remnant movement, makes the assumption of head movement to a specifier position unneces- sary, at least in this type of cases.

[4] conclusion

Syntactic change is slow, and few changes take place from one generation to the next; changes in phonology and in the lexicon are much more frequent. The reason may be that syntax is the most basic and abstract part of the grammar, in the sense that there is no direct interface connection between syntax and the ex- tralinguistic components, such as sound or meaning, which are connected to phonology and semantics, respectively. The basic syntactic structures are part of Universal Grammar, and any changes are either very superficial, or related to features in the morphology or the lexicon. As many as four changes in the course of perhaps less than 500 years may therefore seem unlikely. In this paper I have shown how four syntactic changes have one and the same underlying cause. The difference between the two historical stages of Norwegian is therefore not as radical as it may seem on the surface.

acknowledgements

I want to thank Helge Lødrup, Terje Lohndal, the volume editors, and an anony- mous reviewer for useful comments and suggestions.

sources

Barl – Rindal, M. (ed), Barlaams ok Josaphats saga. Oslo 1981.

DN – Diplomatarium Norvegicum: Oldbreve til Kundskab om Norges indre og ydre Forhold, Sprog, Slægter, Sæder, Lovgivning og Rettergang i Middelalderen.

Christiania/Oslo 1847-1995.

FS – Cederschiöld, G. (ed), Fornsögur Suðrlanda. Lund 1884.

Hkr – Jónsson, F. (ed), Heimskringla: Noregs konunga sögur af Snorri Sturluson I- IV. København 1893-1901.

Kgs – Holm-Olsen, L. (ed), Kongespeilet. Christiania 1848.

Laxd – Kålund, K. (ed), Laxdœla saga. Halle 1986.

(30)

[18] FAARLUND

ML – Magnus lagabøtes landslov. In R. Keyser, P.A. Munch (eds), Norges gamle Love, 2. Christiania 1848.

Olavss – Den legendariske Óláfs saga ins helga (etter Upps DG 8 II). PROIEL Corpus.

references

Chomsky, Noam. 1993. A minimalist program for linguistic theory. In K. Hale &

S. J. Keyser (eds.), The view from Building 20: Essays in Linguistics in Honor of Syl- vain Bromberger, 1-51. Cambridge, MA: MIT Press.

Chomsky, Noam. 2008. On Phases. In R. Freidin et al. (eds.), Foundational Issues in Linguistic Theory, 133-166. Cambridge, MA: MIT Press.

Christoffersen, Marit. 1993. Setning og sammenheng. Syntaktiske studier i Magnus La- gabøters landslov. Kristiansand: Agder distriktshøgskole.

Corver, Norbert. 2014. Freezing effects. Unpublished version.

Faarlund, Jan Terje. 1977. Transformational syntax in dialectology: Scandinavian word order varieties. In T. Fretheim & L. Hellan (eds.), Papers from the Trond- heim Syntax Symposium, 65-83. Department of Linguistics, University of Tron- heim.

Faarlund, Jan Terje. 1990. Syntactic Change. Towards a Theory of Historical Syntax.

Berlin: Mouton de Gruyter.

Faarlund, Jan Terje. 1995. De la préposition au préverbe en nordique. In A. Rous- seau (ed.), Les préverbes dans les langues d'Europe. Introduction à l'étude de la pré- verbation, 61-75. Lille : Presses Universitaires du Septentrion.

Faarlund, Jan Terje. 2002. Syntactic development from Ancient Nordic to Old Nordic. In O. Bandle et al. (eds.), The Nordic Languages. An International Hand- book of the History of the North Germanic Languages, vol. 1, 729-732. Berlin: Mou- ton de Gruyter.

Faarlund, Jan Terje. 2004a. The Syntax of Old Norse. Oxford: Oxford University Press.

Faarlund, Jan Terje. 2004b. Ancient Nordic. In R. D. Woodard (ed.), The Cambridge Encyclopedia of the World's Ancient Languages, 907-921. Cambridge: Cambridge University Press.

(31)

Faarlund, Jan Terje. 2019. The Syntax of Mainland Scandinavian. Oxford: Oxford University Press.

Holmberg, Anders. 1986. Word Order and Syntactic Features in the Scandinavian Lan- guages and English. PhD Dissertation, University of Stockholm.

Holmberg, Anders. 1999. Remarks on Holmberg’s Generalization. Studia Linguis- tica 53. 1-39.

Hróarsdóttir, Thorbjörg. 2000. Word Order Change in Icelandic. From OV to VO. Am- sterdam: John Benjamins.

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

Kossuth, Karen C. 1978. Typological contributions to Old Icelandic word order under particular consideration of the position of the verb. Acta philologica Scandinavica 32. 37-52.

Laake, Signe. 2018. Word order Changes in the History of Norwegian. PhD Disserta- tion, University of Oslo.

Lightfoot, David. 1979. Principles of Diachronic Syntax. Cambridge: Cambridge Uni- versity Press.

Munch, Christine Ø. 2013. North Germanic Negation. A Microcomparative Perspective.

PhD Dissertation, University of Tromsø.

Nygaard, Marius. 1906. Norrøn Syntax. Christiania: Aschehoug.

Ott, Dennis. 2016. Stylistic fronting as remnant movement. Studia Linguistica 72.

1-38.

Rögnvaldsson, Eiríkur. 1996. Word order variation in the VP in Old Ice- landic. Working Papers in Scandinavian Syntax 58. 55-86.

Salvesen, Christine M. 2011. Stylistic fronting and remnant movement in Old French. In J. Berns et al. (eds.), Romance Languages and Linguistic Theory 2009, 323-342. Amsterdam: John Benjamins.

Sigurðsson, Halldór Á. 1988. From OV to VO: Evidence of Old Icelandic. Working Papers in Scandinavian Syntax 34.

Sundquist, John D. 2006. Syntactic Variation and Change in the History of

(32)

[20] FAARLUND

Norwegian: The Decline of XV Word Order. Diachronica 23. 105-141.

contact

Jan Terje Faarlund University of Oslo [email protected]

(33)

Meklenborg, Lohndal & Østby (eds) Syntax, semantics and acquisition: In honor of Hans Petter Helland, Oslo Studies in Language 12(1), 2021. 21-38. (ISSN 1890-9639/ISBN 978-82-91398-12-9)

possessive parasitter eller parasittiske possessiver?

CATHRINE FABRICIUS HANSEN , ANNELIESE PITZ & BERGLJOT BEHRENS

Universitetet i Oslo Sammendrag

I dette bidraget drister vi oss til å forutse vanskene som innlærere med forskjellige morsmål (engelsk, tysk, og norsk) forventes å ha når de skal lære det franske possessivsystemet. Mer konkret dreier det seg om forståelse/tolking av possessivdeterminativene son, sa, ses. De tre valgte språkene har nokså ulike systemer som inviterer i forskjellig grad til transfer og tolk- ningsfeil, og spørsmålet er hvilke faktorer i morsmålets system som gjør det lettere eller vanskeligere å tilegne seg det franske. Det foreligger en rekke undersøkelser når det gjelder norske innlærere (se bl.a. Helland 2017; 2019), men for å kunne svare på hvilken gruppe av morsmålstalere som har den tyngste oppgaven foreslår vi en sammenlignende undersøkelse på grunnlag av en test der informantene blir bedt om å identifisere referenten for possessivet under varierende betingelser. I bidraget beskriver vi kort de forskjellige systemene, våre antagelser om innlæringsproblemer og om transfer fra de respektive L1, samt testoppsettet.

[1] innledning

Relevant forskning synes å vise at ordprosessering og syntaktisk prosessering på et fremmedspråk ikke i samme eller like tydelig omfang er utsatt for påvirkning fra morsmålet (se for eksempel Lago et al. 2020 og videre henvisninger der):

“L1 influence is clearer in L2 word processing than in syntactic processing.

Of course, this does not negate a potential role of L1 influence in syntactic processing: It is possible that reliable evidence will emerge once methodo- logically reliable comparisons are conducted. What is interesting, however, is that this variability is not observed in L2 word processing. For words, learners seem to show reliable evidence of coactivation at the form and meaning level, despite differences in L2 dominance. This suggests that CLI [Cross-Linguistic Influence] may differentially affect lexical and syntactic processing. (Lago et al. 2020, s. 9)

I vår sammenheng er fransk (FR) fremmedspråket mens hhv. norsk (NO), engelsk

(34)

[22] FABRICIUS HANSEN, PITZ & BEHRENS

(EN) og tysk (TY) er L1. Vi kan gå ut fra at norsktalende og tysktalende som lærer fransk har [lært] engelsk som første fremmedspråk, mens fransk muligens er første (og eneste) fremmedspråk for engelsktalende FR-innlærere. Det vil si, vi har å gjøre med fransk som Ln, der n ≥ 2.

Det spørsmål vi er opptatt av er hvordan FR-innlærere med NO, TY eller EN morsmål prosesserer – tolker, forstår – de franske possessive pekerne son, sa og ses under gitte syntaktiske betingelser.

Vanligvis behandles tolkning av 3. persons pronomener i Ln under en rent syntaktisk synsvinkel der spørsmålet er i hvilken grad Ln-innlæreren overholder de syntaktiske restriksjoner som gjelder for pronomener i det aktuelle språket og i hvilken grad han/hun eventuelt er påvirket av sitt morsmål (eller et annet Ln) under prosesseringen (se for eksempel Clahsen & Felser 2006, Felser & Cun- nings 2012, Patterson et al. 2014 og videre henvisninger der). Leksikalsk transfer (’priming’) fra L1 (eller et annet fremmedspråk) har oss bekjent ikke vært noe tema i den sammenheng. Temaet er imidlertid aktuelt også når det gjelder pro- nominalsystemene (i vid forstand) i nært beslektede språk der det kan være mer eller mindre klare tverrspråklige likheter mellom pronominale ’items’; jf. for eksempel NO seg v. TY sich og NO han, de vs. EN he, they.

Formålet med dette bidraget er først å klargjøre hvilke L1 faktorer som på- virker mulige feiltolkninger i tyske, norske og engelske innlæreres forståelse av franske possessiver (avsnittene 2 og 3). På dette grunlaget stiller vi opp noen hypoteser om hvilken effekt de syntaktiske betingelser for tolking av possessivene i det enkelte L1 kan ha på forståelsen (avsnitt 4). Vi skisserer et eksperiment (avsnitt 5) som inviterer den for oss unge og aktive jubilanten til å samarbeide med andre forskere innenfor det tyske, det norske og det engelske språkmiljøet, i god SPROSS-orientert ånd, siden vi, forfatterne av dette bidraget, har gått av med pensjon og ikke lenger har forskningsmidler å bidra med.

[2] s-possessiver og andre possessiver: fransk, tysk, norsk, engelsk

De franske s-possessivene son, sa og ses, i det følgende symbolisert som s*, for- utsetter en possessor – en forutgående eller eventuelt etterfølgende ’ante- cedent’ – i singular, men er uspesifiserte mht. possessors genus/sexus. Dette vil si at både son, sa og ses i (1a) og (1b) må vise tilbake til Anna resp. Pierre; flertalls- nominalene les parents ’foreldrene’ og les enfants ’barna’ kommer ikke på tale som antecedenter. Valget mellom de tre enhetene er bestemt av grammatisk genus og numerus til possessum – det leksikalske hodet i det nominalet som possessivet er determinativ til: son, sa og ses må forbindes med nominaler i hhv. mask.

(35)

POSSESSIVE PARASITTER ELLER PARASITTISKE POSSESSIVER? [23]

sing. (som dessert), fem. sing. (som crêpe) og plural (som sandwiches).¹ (1) a.

son dessert-M.SG Tandis que les parents Pierre / Anna mange sa crêpe-F.SG

commandent du vin, ses sandwiches-PL

(’Mens foreldrene bestiller vin, spiser Anna / Pierre desserten sin/ crêpen sin/ sandwichene sine’)

b.

son dessert-M.SG Tandis que Pierre / Anna les enfants mangent sa crêpe-F.SG

commande du vin, ses sandwiches-PL

(’Mens Anna / Pierre bestiller vin, spiser barna desserten hennes/

hans// crêpen hennes/ hans// sandwichene hennes/ hans’)

Norsk og tysk har også adnominale s-possessiver: norsk sin, si, sitt og sine (i det følgende representert som si*) og tysk sein* (med bøyningsformene sein, seine, sei- nen, seinem, seiner, seines).² I begge språkene er de forskjellige formene possessum-relaterte som for fransk son, sa, ses: De kongruerer med det nominale hodet mht. grammatisk genus og numerus, i tysk også kasus. Når det gjelder possessor- relaterte trekk er det imidlertid vesentlige forskjeller mellom s-possessivene i de tre språkene, eller mer generelt: mellom possessivsystemene i sin helhet (se Fabricius-Hansen et al 2017 for en oversikt).

De norske s-possessiver er inherent refleksive, dvs. sin, sitt, sine i (2) må vise til subjektet i egen setning (såkalt lokal binding, cf. Chomsky 1981 ); subjektet i den forutgående undersetningen – Linda, Andreas eller vennene – er strukturelt utelukket som antecedent/possessor. Fransk s* kan derimot brukes både refleksivt som i (1a) og ikke-refleksivt som i (1b). På den andre siden er norsk si*

uspesifisert mht. possessor-genus og -numerus – jf. (2) – mens FR s* som nevnt har trekket [possessor-sg.]. (I det følgende markerer vi strukturelt mulige pos- sessorer med √ og strukturelt utelukkede med #, i tillegg til hhv. blå og rød

[1] Men merk at son – som mon ’min’ og ton ’din’ – benyttes isteden for sa (ma, ta) i kombinasjon med et feminint substantiv som begynner på en vokal, f. eks. son amie ’veninne’. I disse tilfellene kan vi si at sa bare tar formen son, dvs. at son ikke signaliserer maskulinum singular. For å si det med jubilantens ord (Helland 2006: 155): “I femininum 1p-3p finner vi to varianter av det samme morfemet avhengig av om nomenet introduseres med konsonant (ma, ta, sa) eller vokal (mon, ton, son). De siste formene sam- menfaller med maskulinum 1p-3p.” I det følgende ser vi bort fra denne komplikasjonen.

[2] For enkelthetens skyld ser vi bort fra den possessum-relaterte hunkjønnsformen si*.

(36)

[24] FABRICIUS HANSEN, PITZ & BEHRENS

skrift.)

(2) Mens #Linda/ #Andreas/ #vennene bestiller konjakk, spiser √Anna/ √Petter/

√barna crêpen sin/ sukkertøyet sitt/ pannekakene sine.

Det tyske s-possessivet sein* likner det franske s-possessivet på to punkter: det er uspesifisert for refleksivitet og spesifisert for possessor-entall. Men i motsetning til det franske s*-possessivet, som er neutralt mht. possessors kjønn, forlanger sein* en antecedent i hankjønn, dvs. det har trekkene [possessor-sg., possessor-mask.].³ Det betyr for eksempel at sein i (3a) kan referere (refleksivt) til det lokale subjektet Peter eller (ikke-refleksivt) til Andreas, men ikke til Anna eller Linda, mens antecedenten for son i (3b) og (3c) kan være Anna eller Pierre.

(3) a. Während #Linda/ √Andreas Cognac bestellt, isst √Peter/ #Anna sein Dessert.

‘Mens Andreas bestiller konjakk, spiser Peter desserten sin/hans.’

b. Tandis que √Anna commande du cognac, √Pierre mange son dessert.

‘Mens Anna bestiller konjakk, spiser Pierre desserten sin/hennes.’

c. Tandis que √Pierre commande du cognac, √Anna mange son dessert.

‘Mens Pierre bestiller konjakk, spiser Anna desserten sin/hans,’

Forskjellene mellom s-possessivene i fransk, tysk og norsk er sammenfattet i tabell 1.

Possessor-relaterte trekk

Possessiv Numerus Genus (hvis sg.)

Refleksivitet

FR son/sa/ses sing. – –

TY sein* sing. mask. –

NO sin/sitt/sine – – refleksiv

tabell 1. s-possessiver i fransk, norsk og tysk

Når betingelsene for å anvende s-possessivet ikke er oppfylt, må det i alle tre språk brukes et annet possessiv(leksem):

[3] I denne artikkelen har vi utelukkende å gjøre med antecedenter/ possessor i hankjønn, hunkjønn (eller flertall). Vi ser derfor bort fra at tyske substantiver også kan være neutrum.

(37)

POSSESSIVE PARASITTER ELLER PARASITTISKE POSSESSIVER? [25]

• FR leur*, som er spesifisert for possessor-flertall og har formene leur (possessum-sg.) og leurs (possessum-pl.); jf. (4a).

• TY ihr*, som forlanger en antecedent i flertall eller hunkjønn entall og bøyes som sein* i samsvar med possessum; jf. (4b).

• NO hans, hennes eller deres avhengig av om possessor er hankjønn entall, hun- kjønn entall eller flertall; disse possessivene er inherent irrefleksive, dvs. de kan ikke referere til subjektet i egen setning – ikke være lokalt bundet – og bøyes ikke (ytterligere).⁴⁵

(4) a. Tandis que √les parents commandent du vin, #Anna/#Pierre mange leur dessert/ leurs sandwiches.

‘Mens foreldrene bestiller vin, spiser Anna/Pierre desserten deres/

sandwichene deres.’

b. Während √Linda Cognac bestellt, isst #Peter ihr Eis.

Während √die Eltern Cognac bestellen, isst #Peter ihr Eis.

‘Mens Linda/foreldrene bestiller konjakk, spiser Petter isen hen- nes/deres.

Engelsk har i motsetning til fransk, norsk og tysk ikke noe s-possessiv. Når det gjelder possessor-kjønn og -tall, svarer his, her og their til NO hans, hennes og deres, men i motsetning til disse er de engelske possessivene neutrale mht. refleksivitet; sml. for eksempel (5a) and (5b).

(5) a. While √Linda is ordering cognac, √Ann eats her dessert.

b. Mens √Linda bestiller konjakk, spiser #Anna desserten hennes.

Tabell 2 gir en oversikt over possessivene uten s- i de fire aktuelle språkene.

[4] Vår fremstilling av bindings-/refleksivitetsrestriksjonen på si* er svært forenklet; se for eksempel Fab- ricius-Hansen et al. (2017, sek. 3) og referanser samme sted.

[5] Historisk sett dreier det seg om genitivformer av 3. persons personlige pronomener han, hun og dere.

View of Vol. 12 No. 1 (2021): Syntax, semantics and acquisition: In honor of Hans Petter Helland

Christine Meklenborg, Terje Lohndal &

Kathrine Asla Østby (eds.)

Syntax, semantics and acquisition:

In honor of Hans Petter Helland

Christine Meklenborg, Terje Lohndal &

Kathrine Asla Østby (eds.)

Syntax, semantics and acquisition:

In honor of Hans Petter Helland

Contents

introduction

tabula gratulatoria

a cluster of changes: norwegian word order

possessive parasitter eller parasittiske possessiver?