The syntax of Scandinavian verb particles has been described by a number of researchers,
and seminal work include Johannisson (1939), Sandøy (1976), Svenonius (1996, 2003 and 2005),
Taraldsen (1983) and Åfarli (1985). The main patterns for particle placement are illustrated below.
In Norwegian, Icelandic and Faroese, a verb-particle can either precede or follow an NP direct object,
as shown for Norwegian and Icelandic in (1) and (2) below:[1]
(1)
a)
Vi kastet hunden ut.
(Nor.)
b)
Við
hentum
hundinum
út.
(Ice.)
we
threw
dog.def
out
'We threw the dog out.'
(2)
a)
Vi kastet ut hunden.
(Nor.)
b)
Við
hentum
út
hundinum.
(Ice.)
we
threw
out
dog.def
'We threw out the dog.'
In th discussion section below (section 3) below we will look at some Norwegian dialects where particle placement with respect to NP
objects is less free, and at least partly determined by semantic factors. Particles can however not precede
pronominal direct objects in Icelandic and Norwegian (with the exception of some particle constructions that
are discussed in section 3):
(3)
a)
Vi kastet den ut.
(Nor.)
b)
Við
hentum
honum
út.
(Ice.)
we
threw
it/him
out
'We threw him/it out.'
(4)
a)
*Vi kastet ut den.
(Nor.)
b)
*Við
hentum
út
honum.
(Ice.)
*we
threw
out
it/him
*'We threw it/him out.'
In Danish, particles obligatorily follow both NP and pronominal direct objects:
(5)
a)
Boris
skruede
musikken/den
ned.
(Dan.)
Boris
turned
music.def/it
down
'Boris turned the music/it down.'
b)
*Boris
skruede
ned
musikken/den.
(Dan.)
*Boris
turned
down
music.def/it
*'Boris turned down the music/it.'
In Swedish, on the other hand, particles precede both NP and pronominal objects, as shown below:
(6)
a)
Vi
kastade
ut
hunden/den.
(Swe.)
we
threw
out
dog.def/it
'We threw out the dog/it.'
b)
*Vi
kastade
hunden/den
ut.
(Swe.)
*we
threw
dog.def/it
out
*'We threw out the dog/it.'
2. Results
2.1 Nordic Syntactic Database (NSD)
In the ScanDiaSyn survey, particle placement in active sentences was only tested in Swedish-speaking areas
(i.e., in Sweden and parts of Finland), except for the souternmost parts of Sweden. As far as we are aware, the particle pattern in the southernmost parts is the same as the standard Swedish pattern. The aim was to investigate how robust the Swedish pattern given in (6) above
really is. Particle placement was tested with both pronominal and NP direct objects. The following sentences were
used to test particle placement with respect a full NP object:
(7)
Jag
satte
på
radion.
(#1421) (Swe.)
I
turned
on
radio.def
'I turned on the radio.'
(8)
Jag
satte
radion
på.
(#1420) (Swe.)
I
turned
radio.def
on
'I turned the radio on.'
The results are shown in Map 1 and Map 2 below:
Map 1: Particle preceding NP object
(#1421:Jag satte på radion. 'I turned on the radio.')
(White = high score)
Map 2: Particle following NP object
(#1420: Jag satte radion på. 'I turned the radio on.')
(White = high score, grey = medium score, black = low score)
As can be seen clearly, particles preceding the direct object is the standard word order in the Swedish speaking areas.
The opposite order is rejected by most informants. The interesting exceptions are three adjacent locations in southern Finland (Dragsfjärd, Snappertuna and Kyrkslätt), Larsmo in Österbotten, Finland, and
Överkalix in Northern Sweden, where several of the informants gave test sentence (8) a high or intermediate score.
The following two sentence were used to test particle placement with respect to a pronominal object:
(9)
Jag
satte
på
den.
(#1422) (Swe.)
I
turned
on
it
'I turned it on.'
(10)
Jag
satte
den
på.
(#1423) (Swe.)
I
turned
it
on
'I turned it on.'
The results are shown in Map 3 and Map 4 below:
Map 3: Particle preceding pronominal object
(#1422: Jag satte på den. 'I turned it on.')
(White = high score, grey = medium score, black = low score)
Map 4: Particle following pronominal object
(#1423: Jag satte den på. 'I turned it on.')
(White = high score, grey = medium score, black = low score)
As can be seen from Map 4, a pronoun preceding a particle is not accepted by any informants in Sweden, and by only very informants in Finland. It is interesting to note that pre-particle NP-objects are accepted to a higher degree than pre-particle pronominal objects in the Swedish speaking parts of Scandinavia.
Also, remember that particles cannot precede pronominal objects in the other Scandinavian varieties (with exceptions,
to be returned to). There is in other words a very clear isogloss here, which follows the border between Sweden and Norway
and Sweden and Denmark. Some informants gave low scores for sentence (9), and at the moment we do not know why this is,
but probably it is not due to the particle placement, given that sentence (10) was judged equally bad or worse by the same
informants.
3. Discussion
Vinka (1999) reports on some Northern Swedish dialects where certain particles optionally follow pronouns.
He calls these particles predicative particles, and they behave differently from non-predicative particles with respect
to a number of tests. The clearest test is whether a bare particle can follow a copula, as in the following example:
(11)
a)
Tv'n
är
på.
(Predicative part.)
TV.def
is
on
'The TV is on.' (i.e., 'The TV is turned on.')
b)
*Vinet
är
upp.
(Non-predicative part.)
*wine.def
is
up
*'The wine is up.' (int. 'The wine is/has been drunk up.')
Vinka thus predicts that the sentence (#1423), repeated below, should be grammatical:
(12)
Jag
satte
den
på.
(#1423) (Swe.)
I
turned
it
on
'I turned it on.'
As shown in Map 4 above, this sentence has not been judged as grammatical at any locations in Northern Sweden.
However, it should be noted that post-NP particles got higher scores in the northern parts of Sweden (and Finland)
than in other parts. More detailed fieldwork in the relevant areas is needed to establish exactly where, and to which extent, sentences like (12) are acceptable in northern Sweden. For now we can only conclude that the order between particles and pronouns seems to be rigid
throughout Sweden. We also do not have any evidence that the distinction predicative - non-predicative has any
effect on particle placement.
Sandøy (1976) reports that there is a distinction between directional and non-directional particles, at least in
the Romsdal dialect in Western Norway, which is the focus of Sandøy's study. Clearly non-directional particles have
to precede the direct object in the Romsdal dialect (13), while directional particles can either follow or precede the
direct object (14).[2] The following two examples illustrate the difference (curly brackets indicate possible or impossible
positions for the particle):
(13)
Han
he
rekna
{ut}
prisan
{*ut}
(Rom.)
he
has
calculated
out
prices.def
out
'He has calculated the prices.'
(14)
Han
bar
{?ut}
fangst'n
sin
{ut}
åt
dei
fattige.
(Rom.)
he
carried
out
catch.def
his
out
to
the
poor
'He carried his catch out to the poor.'
Whether the semantic distinction between the particles in (13) and (14) should be described as directional vs.
non-directional is debated. Aa (2010) argues that the correct distinction is rather dynamic versus resultative,
and that dynamic particles tend to precede the object, while resultative particles follow the object. It should however
be pointed out that the semantic function of the particle plays a fairly small role in determining the position of the
particle in other parts of Scandinavia. Both in Swedish and Danish, particle placement is fixed. In many Norwegain
dialects (and in Insular Scandinavian), information structure plays a more important role: objects that are new in the
discourse (and therefore carry strong stress) are more likely to follow particles than given objects (most clearly seen
in the case of pronouns).
Aa (2010) reports from fieldwork in Trøndelag that some dialects show more or less a Swedish pattern, especially
dialects spoken in areas close to the Swedish border. In Nordli, both pronouns and full NP objects obligatorily follow
a directional particle. In Fosen, closer to the coast, NP objects preferably follow the particle, while pronouns precede
the particle.
(15)
a)
Han
kasta
{ut}
hunden
{*ut}.
(Nordli)
he
threw
out
dog.def
out
'He threw the dog out.'
b)
Han
kasta
{ut}
han
{??/*ut}.
(Nordli)
he
threw
out
him
out
'He threw it out.'
(16)
a)
Han
kasta
{ut}
hunden
{??ut}.
(Fosen)
he
threw
out
dog.def
out
'He threw the dog out.'
b)
Han
kasta
{??ut}
han
{ut}.
(Fosen)
he
threw
out
it
out
'He threw it out.'
Note that the particle ut in the examples above at least potentially can be interpreted as either a directional
or a resultative particle.
In the Nordic Dialect Corpus, utterances with a particle preceding a non-stressed pronoun are fairly easy to find.
They seem to be absent in Northern Norwegian, and very uncommon in Southern Norwegian. In Trøndersk and Eastern
Norwegian they are fairly frequent, though particles following pronouns are still more common. The following examples
is from Inderøy, in Nord-Trøndelag, where first a DP-object follows a particle, and then a non-stressed pronoun follows
a particle:
(17)
der
tar
de
inn
ungene
i
storkorpset
og
lærer
opp
dem
(inderoey_03gm)
there
take
they
in
kids.def
in
big.band
and
teach
up
them
'There they admit the kids to the big march band, and teach them'
It should further be noted that there are cases in Norwegian (and the other Scandinavian languages as well) where
something that looks like a particle has to precede a pronominal complement. Two examples are given below: the
preposition/particle til 'to' that 'creates' punctual events (see Tungseth 2006 for detailed discussion), and the
preposition/particle av 'off', that expresses source:
(18)
a)
Hun
slo
{til}
ham
{*til}.
(Nor.)
she
hit.past
to
him
to
'She hit him.'
b)
De
slet
{av}
ham
{*av}
genseren.
(Nor.)
they
tear.past
off
him
off
sweater.def
'They tore his sweater off.' (lit. 'They tore off him the sweater.')
Til is analyzed as a verbal preposition by Tungseth (2006), and might thus not constitute a counterexample to the
general particle-pattern. Also av in (18b) can be analyzed as a preposition, with ham as the complement. However,
preposition phrases can in general not precede the direct object, which makes av look more like a particle. It should
be noted though that ham in (18b) does not behave like a typical direct object either, as it cannot easily be promoted
to subject in a passive or topicalized in a regular active sentence.
From a historical perspective, it interesting to note that already Aasen (1864) noted that in Norwegian the
particle-object order is the unmarked order when the object carries heavy stress, while pronouns in the unmarked
case precede the particle. Aasen gives Trøndersk and the dialects spoken in the areas closest to Sweden in Eastern
Norway as exceptions.
Particle placement is also discussed in the NALS-chapters on particle placement in passives (Lundquist 2014a) and particle placement
and reflexive verbs (Lundquist 2014b).
References
Aa, Leiv Inge. 2010. 'Dynamic and Resultative Verb-Particle Constructions in Spoken Norwegian.' Ms, NTNU.
Åfarli, Tor A. 1985. 'Norwegian verb particle constructions as causative constructions.' Nordic Journal of Linguistics 8 1: 75–98.
Aasen, Ivar. 1864. Norsk Grammatik, Christiania, Oslo (1965).
Johannisson, Ture. 1939. Verbal och postverbal partikelkomposition i de germanska språken. Diss. Lindstedt, Lund.
Lundquist, Björn. 2014b. ‘Verb-particles: particle placement in reflexive verbs,’ Nordic Atlas of Linguistic Structures (NALS). http://www.tekstlab.uio.no/nals#/chapter/6.
Sandøy, Helge. 1976. Laust samansette verb i vestnordisk: ein samanliknande leddstillingsanalyse for islandsk, færøysk og romsdalsmål. Universitetet i Oslo, Oslo.
Svenonius, Peter. 1996. 'The optionality of particle shift.' Working Papers in Scandinavian Syntax 57: 47–75.
Svenonius, Peter. 2003. 'Swedish particles and directional prepositions.' In Grammar in Focus: Festschrift for Christer Platzack 18 November 2003, edited by Lars-Olof Delsing, Cecilia Falk, Gunlög Josefsson, and Halldór Á. Sigurðsson, vol. II: 343–351. Department of Scandinavian Languages, Lund University, Lund.
Taraldsen, Knut Tarald. 1983. 'Some phrase structure dependent differences between Swedish and Norwegian.' Working Papers in Scandinavian Syntax 9: 1–45.
Tungseth, Mai. 2006. Verbal Prepositions in Norwegian: Paths, Places and Possession. Diss. University of Tromsø.
Vikner, Sten. 1987. 'Case assignment differences between Danish and Swedish.' In Proceedings of the Seventh Biennial Conference of Teachers of Scandinavian Studies in Great Britain and Northern Ireland, edited by Robin M.A. Allan and Michael P. Barnes: 262–281. University College London, London.
Vinka, Mikael. 1999. 'Predicative and non-predicative verb particle constructions.' In WCCFL 18 Proceedings, edited by S. Bird, A. Carnie, J. Haugen, and P. Norquest: 570–585. Cascadilla Press, Somerville, Ma.
[1]
Pre-argumental particles in Faroese are quite rare, compared to Norwegian and Icelandic.
[2] When a directional particle modifies a PP as in (14), the particle preferably
follows the direct object. A non-directional particle can either precede or follow the object in this context.
