1. technology and computing
2. programming languages
3. javascript

Nordic Atlas of Language Structures (NALS) Journal, Vol. 1, 506­511
C opyright © Björn Lundquist 2014 Licensed under a C reative C ommons Attribution 3.0 License
Long distance binding 2: relative clauses
Björn Lundquist
UiT The Arctic University of Norway
1. Introduction
Long distanc e reflexives are in general more ac c epted in the insular North Germanic languages than in the mainland North Germanic languages (see e.g.
Holmberg and Platzac k 1995 and Thráinsson 2007 and referenc es therein). However, binding into relative c lauses seems to be more ac c eptable in the
mainland varieties than in the insular varieties. As shown in (1), from Lødrup (2009), binding into a relative c lause is sometimes possible in Norwegian:
(1) Det kan og
se
ut
til at
noen av informantenei
er
preget
av særinteresser
som gjelder for sinei
institutter.
(Bokmål.)
it
may also look out to that some of informant.PL.DEF. are influenc e.PASS.PART by spec ial.interests that apply for RFLX.POSS.PL department.PL
‘It may also seem as if some of the informantsi are influenc ed by the spec ial interests that apply to theiri departments.’
As shown in (2), from Strahan (2011), binding into a relative c lause is not possible in Ic elandic :
(2) *Það getur litið út
fyrir að
nokkrir heimildarmannanai séu mótaðir
af sérhagsmunum
sem varða
sínari
deildir.
(Ic e.)
it
c an look like suc h that some informant.PL
are influenc e.PASS.PART by spec ial.interest DEF REL c onc ern.PRES RFLX.POSS.PL department.PL
‘It may also seem as if some of the informantsi are influenc ed by the spec ial interests that apply to theiri departments.’
Lødrup (2009) shows that binding into relative c lauses in Norwegian is restric ted to c ontexts where the subjec t of the relative c lause is inanimate, or
non­referential. This restric tion does not apply to long distanc e binding into c omplement c lauses in Ic elandic , or so­c alled mid­distanc e binding into
infinitival c lauses in Mainland and Insular Sc andinavian. Lødrup also shows that long distanc e reflexives c an be c omplex in relative c lauses, in c ontrast to
the long distanc e reflexives found in Insular Sc andinavian (see Lundquist 2013a and Lundquist 2013b for disc ussion on other types of non­loc al
reflexives):
(3)
Huni trodde
hun gjorde det som var best for segi selv i.
she think.PAST she did
it
REL
was best for
‘Shei thought shei did what was best for heri.
RFLX
(Ic e.)
SELF
2. Results
2.1 Nordic Syntactic Database (NSD)
In the Sc anDiaSyn survey, the following sentenc e was tested in Norway, Sweden and Finland:
(4)
folki
leser
vel
bare de brevent
som er
til segi selv i.
people read.PRES presumably only the letter.PL.DEF REL be.PRES to
‘Peoplei presumably only read the letters whic h are for themi’
RFLX
(#116) (Nor.)
SELF
In this sentenc e, the subjec t of the relative c lause is inanimate, and long distanc e binding ac ross this subjec t is expec ted to be possible. As seen in the
Map 1 below, this sentenc e is ac c epted in most of the measure points:
Lundquist
Long binding: Relative
NALS Journal
Map 1: Binding into relative clause. (#116: Folk leser vel bare de brevene som er til seg selv. 'People
presumably only read the letters which are for them.') (White = high score, grey = medium score, black = low score)
There are some plac es where the sentenc e only gets a medium sc ore, but there are no straightforward dialec tal patterns. A c loser look at the informants
judgments shows that over 80 per c ent of the informants gave the sentenc e a full sc ore.
In Denmark, a set of four sentenc es was tested, where the c o­referent element either has the form of a simple reflexive, a c omplex reflexive, a
personal pronoun or a c omplex personal pronoun (‘herself’). The subjec t of the embedded c lause is non­referential.
(5)
Huni mente
hun gjorde det som var bedst for sigi.
for RFLX
(#164) (Dan.)
hun gjorde det som var bedst for sigi selv i.
(#165) (Dan.)
she c laim.PAST she did
it
REL
was best
‘She i c laimed she i did what was best for heri.
(6)
Huni mente
she c laim.PAST she did
it
REL
was best
‘She i c laimed she i did what was best for heri.
(7)
Huni mente
Huni mente
RFLX SELF
hun gjorde det som var bedst for hendei selv i.
she c laim.PAST she did
it
REL
was best
‘She i c laimed she i did what was best for heri.
(8)
for
for
RFLX
(#166) (Dan.)
SELF
hun gjorde det som var bedst for hendei selv i.
for RFLX
SELF
(#167) (Dan.)
she c laim.PAST she did
it
REL
was best
‘She i c laimed she i did what was best for heri.
As the maps below shows, the c omplex reflexive form is the most ac c eptable form of the given alternatives (Map 3). The c omplex pronominal form is in
general ac c epted as well (Map 5), while both simple forms are not ac c epted, with a few exc eptions.
507
Lundquist
Long binding: Relative
Map 2: Simple reflexive in relative clause. (#164: Hun mente hun gjorde det som var bedst for sig. 'She claimed she
did what was best for her.') (White = high score, grey = medium score, black = low score)
Map 3: Complex reflexive in relative clause. (#165: Hun mente hun gjorde det som var bedst for sig selv. 'She claimed she did what was best for her.')
(White = high score, grey = medium score, black = low score)
508
NALS Journal
Lundquist
Long binding: Relative
NALS Journal
Map 4: Binding into an embedded object
(#166 Hun mente hun gjorde det som var bedst for hende selv. 'She
claimed she did what was best for her.') (White = high score, grey = medium score, black = low score)
Map 5: Complex co­referential pronoun in relative clause. (#167 Hun mente hun gjorde det som var bedst for hende selv. 'She claimed she did what was best for her.') (White = high score, grey = medium score, black = low score)
The maps show that there is dialec tal variation here, espec ially when it c omes to the c omplex pronominal form (map 5), whic h is not ac c epted on Fyn
and southern Sjælland. It is also interesting that simple pronominal form (map 4) is only ac c epted on one measure point, namely Ærø. The ac c eptanc e for
the simple pronoun is probably muc h higher in the rest of Sc andinavia.
3. Discussion
Below I will first look at some data from other sourc es that c over Ic elandic and Faroese. Thereafter I will disc uss different fac tors that fac ilitate binding
into relative c lauses.
3.1. Other data sources
As was shown in the introduc tion, long distanc e binding of the type tested above (example 4, 5 and 6) is not ac c epted in Ic elandic . In Faroese, however,
binding into a relative c lause is ac c epted to a higher degree. In the dialec t survey c arried out on the Faroese Islands (Thráinsson et al. 2008­2009, see
disc ussion in Strahan 2011), it was found that 67 per c ent of the informants found the following sentenc e ac c eptable:
(9)
Honi visti
ongan, sum var forelskaður
í
she know.PAST no­one REL was enamorate.PART in
’She i knew no­one who was in love with heri.’
særi.
(Far.)
RFLX. DAT
Note that a sentenc e with the same struc ture as (9) above is not ac c eptable in Ic elandic : only 22 per c ent of the Ic elandic informants in the Ic elandic
dialec t survey ac c epted the c orresponding Ic elandic version of (9) (Thráinsson et al. 2005­2007, see disc ussion in Strahan 2011). Note however that
binding into relative c lauses is possible in Ic elandic as long as the c lause hosting the relative c lause is embedded under a verb of saying or believing, that
triggers subjunc tive marking on the embedded verbs, see Thráinsson (1990) and (2008), and Holmberg and Platzac k (1995) for disc ussion on the so­c alled
‘Domino effec t’. The following example is from Holmberg and Platzac k (1995):
509
Lundquist
(10)
Long binding: Relative
Jóni segir
að
þetta sé
stúlkan sem elski
NALS Journal
sigi.
Jon say.PRES.3RD that this is.SUBJ gril
SEM love. SUBJ
’Johni says that this is the girl that loves himi.
(Ic e.)
RFLX. ACC
We c ould suspec t that the reflexive in the Danish test sentenc e (5)­(6) is also lic ensed by the superordinate verb of saying mene ‘c laim’, just like in the
Ic elandic example above. However, the Ic elandic long distanc e reflexive is always simplex. If Danish had a domino effec t, we would expec t (5) (simple
reflexive) rather than (6) (c omplex reflexive) to be good.
3.2. Restrictions the intervener and the antecedent
As disc ussed in Lødrup (2009) and Strahan (2011), binding into relative c lauses is lic it in Mainland Sc andinavian when the subjec t of the relative c lause is
inanimate or non­referential. Lødrup also shows that the intervening inanimate subjec t has to be a theme rather than a c ause for LDR to be lic it. The
Swedish referenc e grammar (Teleman et al. 1999) also gives examples where the relative c lause subjec t is an indefinite pronoun, like ingen ‘no­one’,
similar to the Faroese example (9). When the subjec t of the relative c lause is animate and spec ific , LDR is not lic it in any of the North Germanic
languages, as far as I am aware. Lødrup (2009) gives the following Norwegian ungrammatic al sentenc e that presumably is bad all over Sc andinavia:
(10)
*Deii kjenner
mannen som kritiserar
segi (sjølv i) /bror
sini.
they know.PRES man.DEF REL c ritisize.PRES RFLX SELF
love./BROTHER
int. ’They i know the man that c ritic izes themi/theiri. brother.’
(Nor.)
RFLX. POSS. SG
We c an tentatively c onc lude that LDR is lic it in relative c lauses only when an intervening subjec t is not likely to be a binder of an anaphor. For further
disc ussion on this topic , I refer the interested reader to Thráinsson (2008), Lødrup (2009) and Strahan (2011).
The Norwegian/Swedish test sentenc e (4) and the Danish test sentenc es (5­8) differ with respec t to the nature of the antec edent, i.e., the main
c lause subjec t. The antec edent in the Danish examples (5­8) is a spec ific third person pronoun, while the antec edent in the Norwegian/Swedish example
(4) is a generic ally referring noun. As is disc ussed in Lundquist 2012c , generic , non­bound reflexives are ac c epted by many speakers in Mainland
Sc andinavia. There is slight c hanc e that the informants that ac c ept (4), treat the reflexive as a non­bound generic reflexive rather than a long distanc e
reflexive. This possibility is brought up by Lødrup (2009), but ac c ording to him, at least some Norwegian speakers rejec t generic non­bound reflexives
while ac c epting sentenc es like (4). In Swedish, many speakers (inc luding myself) would find (4) equally marked (or unmarked) as (12) below, where the
reflexive lac ks a syntac tic antec edent, and rec eives a generic interpretation (or possibly a first person/speaker interpretation):
(11)
?Det är
alltid
spännande med brev
som är
till sigi själv i.
it
be.PRES always exc iting
with letter.PL REL are to
‘It is always exc iting with letters that are for one self.’
(Swe.)
RFLC SELF
However, there are Swedish speakers who allow binding into relative c lauses even when the reflexive has a non­generic referenc e. The Swedish referenc e
grammar (Teleman et al. 1999, 2:343) gives the following example, where the antec edent is spec ific (and the intervening relative c lause subjec t is
indefinite), whic h is said to be ac c eptable for at least some speakers:
(13) Hani har ingen som ser till sigi nuförtiden (Swe) He has no­one REL look.PRES to RFLX nowadays ‘Hei has no­one that looks after himi these days.’
(13)
Hani har
ingen
som ser
till
sig nuförtiden.
(Swe.)
he have.PRES no­one REL look.PRES after RFLX nowadays
‘Hei has no­one that looks after himi these days.’
Ac c ording to the Teleman et al. (1999), the main c lause verb influenc es the possibility of relative c lause LDRs. Ac c ording to them, binding into a relative
c lause is only possible when the main c lause verb denotes some kind of possession, i.e., when the subjec t is a possessor of the objec t that c ontains the
relative c lause with the anaphor. If this c laim is c orrec t, there are as many as four different fac tors that influenc e the ac c eptability of long distanc e
reflexives in relative c lauses:
1. The nature of the main c lause subjec t/antec edent: Generic vs. spec ific ..
2. The nature of the intervening relative c lause subjec t: definite/spec ific vs. indefinite/non­referring.
3. The main c lause predic ate: verb of possession vs. other types of verbs.
4. The embedded predic ate: agentive/c ausative verbs vs. stative verbs (i.e., long distanc e binding is not possible when the intervening subjec t is an agent/c auser)
More researc h needs to be done to establish whether all these fac tors are relevant in all of the North Germanic languages.
References
Faarlund, Jan Terje, Svein Lie and Kjell Ivar Vannebo. 1997. Norsk referansegrammatikk. Universtitetsforlaget, Oslo
Holmberg, Anders and Christer Platzac k. 1995. The Role of Inflec tion in Sc andinavian Syntax, Oxford University Press, Oxford.
Lundquist, Björn. 2013a. Long Distanc e Binding 1: Complement Clauses. Nordic Atlas of Linguistic Struc tures http://www.tekstlab.uio.no/nals#/c hapter/20.
Lundquist, Björn. 2013b. Mid­distanc e binding. Nordic Atlas of Linguistic Struc tures (NALS). http://www.tekstlab.uio.no/nals#/c hapter/24.
Lundquist, Björn. 2012c . Reflexives without a sentenc e­internal antec edents. Nordic Atlas of Linguistic Struc tures http://www.tekstlab.uio.no/nals#/c hapter/25.
Lødrup, Helge. 2009. ‘Animac y and long distanc e binding in Norwegian,’ Nordic Journal of Linguistic s 32.1, 111­136.
Strahan, Tania E. 2003. Long­Distanc e Reflexives in Norwegian: A Quantitative Study, Linc om Europe, Munic h.
Strahan, Tania E. 2011. ‘A typology of non­loc al reflexives in the Sc andinavian languages,’ Nordic Journal of Linguistic s 34.2, 157­178.
Teleman, Ulf, Staffan Hellberg and Erik Andersson. 1999. Svenska Akademiens Grammatik II: Ord. Svenska Akademien, Stoc kholm.
Thráinsson, Höskuldur. 2007. The Syntax of Ic elandic . Cambridge University Press, Cambridge.
510
(NALS).
(NALS).
Lundquist
Long binding: Relative
Web sites:
Nordic Atlas of Language Struc tures (NALS) Online: http://www.tekstlab.uio.no/nals
Nordic Dialec t Corpus: http://www.tekstlab.uio.no/nota/sc andiasyn/index.html
Nordic Syntax Database: http://www.tekstlab.uio.no/nota/sc andiasyn/index.html
511
NALS Journal