Post on 30-Apr-2023
Remarks on the fundamental left-right asymmetryof natural languages and verb clusters
Klaus AbelsUCL
k.abels@ucl.ac.uk
July 1, 2015Leizpig, IGRA colloquium
1 Introduction(1) Syntax should provide a theory of word order.(2) Cinque 2005 provides a descriptively adequate theory of neutral word order
in the noun phrase and is a step in the right direction.(3) Cinque’s theory is explanatory in that it needs to postulate only readily
learnable paramaters (see Abels 2015 for discussion), though the prove-nance of the universal hierarchy remains in the dark.
(4) Cinque’s theory easily generalizes since it relies on:a. notions of hierarcyb. notions of headednessc. constraints on movement
(5) We will look in some detail at how to generalize the theory and apply theresult to Germanic verb clusters.Clusters are interesting because they are a proper word-order challenge(Wurmbrand 2006, to appear) and we have a lot of data (Barbiers, Auw-era, et al. 2008; Dubenion-Smith 2010; Eroms, Röder, and Spannbauer-Pollmann 2006; Kaufmann 2007; Louden 2011; Patocka 1997; Seiler 2004;Wurmbrand 2006, to appear).
(6) Success would shed light on the derivation of the cluster, in particular theissue of a linear asymmetry (Wurmbrand 2006, to appear).It would shed light on the nature of the elements that cluster, lexical vs.
1
functional (see Cinque 2006; Öhlschläger 1989; Reis 2001; Wurmbrand 1998a.o.).
(7) Structure:a. Discussion and generalization of Cinque’s theoryb. Application to unrestricted three-element clustersc. Restriction to auxiliary, modal, ‘let’, and main-verb clustersd. Extension to particlese. Four-element clustersf. Conclusion
2 Elements of a universal theory of neutral wordorder
2.1 The fundamental left-right asymmetry(8) Universal 20 simplified Greenberg 1963
a. Dem Num A Nb. N A Num Demc. N Dem Num Ad. *A Num Dem N
(9) Rigidity on the left of the lexical head, flexibility on the right.(10) Cinque, 2009 suggests this is part of a pattern: Rigid ordering of satellites
(S1–S3 in (11)) before the lexical head (L4 in (11)) and variable orderingafter it.
(11) a. S1 S2 S3 L4b. L4 S3 S2 S1c. L4 S1 S2 S3d. *S3 S2 S1 L4
(12) Cinque, 2009 exemplifies with attributivea. adjectives of size, color,and nationalityb. directional and locative PPsc. tense, mood, and aspect markersd. circumstantial PPs of time, place, and mannere. adverbsf. restructuring verbs (calling them ’aux’ uniformly)
(13) Abels and Neeleman 2012b free Cinque’s account of Universal 20 of its LCA
2
strictures:(i) The underlying hierarchical arrangement of demonstrative, numeral,
descriptive adjective, and noun within the extended projection of thenoun is fixed in such a way that the demonstrative c-commands theremaining three elements, the numeral c-commands the adjective andthe noun, and the adjective c-commands the noun.
(ii) Phrase structure obeys the non-tangling condition.(iii) All movement involved in deriving unmarked word orders must move
a constituent containing the lexical head.(iv) All such movements land in a position within the extended projection
of the noun so that the moved element strictly c-commands (in thesense of sister containment) the launching site of movement.
(v) Movement is uniformly to the left.(14)
dem num a n n a num dem
dem num n a a n num dem
num a n dem dem n a num
dem a n num num n a dem
3
(15)
dem N num a
t
N dem num a
t
N num a
t
dem N a
t
num dem
a n dem num
t
n a dem num
t
(16) There are 14 derivable unmarked orders and they coincide perfectly withthe attested ones (Cinque 2005, though see Cinque, 2014; Dryer, 2009).
(17) Underivable Dem-A-Num-N without movement
dem a num ndem a num n
(18) Underivable Dem-A-Num-N with wrong movement
4
dem a num
t
N dem
t
a num N
num N dem
t[num N]
a
t[N]
2.2 It’s a class society(19) What exactly are the operative concepts in this account?(20) a. Head-dependent (which I will call head-satellite) distinction
b. Syntactic hierarchyc. Classes of satellites (Cinque speaks of functional heads and modifiers
vs.–presumably–arguments)(21) Why exclude arguments?(22) a. Das
theBuchbook
lesenread
hättehad
erhe
nichtnot
sollen.should
He shouldn’t have read the book.b. Bestätigen,
certifydassthat
FritzFritz
schwimmenswim
kann,can
musstmust
duyou
mindestensat.least
könnencanYou must at least be able to certify that Fritz can swim.
(23) a. … dassthat
erhe
dasthe
Buchbook
nichtnot
hättehad
lesenread
sollen.should
…that he shouldn’t have read the bookb. … dass
thatduyou
mindestensat.least
bestätigencertify
könnencan
musst,must
dassthat
FritzFritz
schwimmenswim
kanncan
5
…that you must at least be able to certify that Fritz can swim(24) Cinque 2009 mentions both auxiliary verbs and adverbs to illustrate his
generalization…(25) a. Schön
beautifullysingensing
hathas
erhe
früherformerly
könnencan
He formerly used to be able to sing beautifully.b. dass
theerhe
häthas
chönecan
schöönbeautifully
singesing
that he used to be able to sing beautifully
c. dassthat
erhe
früherformerly
schönbeautifully
hathas
singensing
könnencan
d. datthat
hijhe
vroegerformerly
prachtigbeautifully
heefthas
kunnencan
zingensing
(26) The problem we run into here is reminiscent of Bobaljik 1999; Nilsen 2003;Svenonius 2002.
(27) Cinque 2009 mentions both auxiliary verbs and prepositional phrases toillustrate his generalization…
(28) a. Inin
jedemevery
Saalhall
singensing
hathas
erhe
früherformerly
können.can
He used to be able to sing in the concert hall.b. dass
thaterhe
häthas
chönecan
iin
jedemevery
Saalhall
singesing
that he used to be able to sing in every concert hallc. dass
thaterhe
früherformerly
inin
jedemevery
Saalhall
hathas
singensing
könnencan
d. datthat
hijhe
vroegerformerly
inin
dethe
zaalhall
heefthas
kunnencan
zingensing
that he formerly used to be able to to sing in the hall(29) A more systematic study would investigate further across-class orderings.
Such a study should also need to evaluate empirically how the variouswithin class orders interact with each other.This is perfect IGRA fodder!
(30) Once we restrict Cinque’s generalization to coherent classes, we can returnto arguments and find the situation better than hopeless. The unexpectedOSV order is virtually absent (Haspelmath et al. 2005) and double objectconstructions seem also to be well behaved (Pearson 2000).
(31) a. IO DO V
6
b. V IO DOc. V DO IOd. *DO IO V
2.3 A formulation(32) Theory: Let Ln be a lexical head and S1…n-1 dependents of Ln such that
‚ all si are members of the same morphosyntactic class‚ all si occur in the extended projection of Ln‚ for all pairs Sj, Sj+1, Sj is hierarchically more prominent (scope, con-
stituency, selection, government,…) than Sj+1thenpossible neutral orders of S1 …Sn-1 Ln are all those orders given by flex-
ibly linearizing structure [ S1 [ S2 …[ Sn-1 Ln ] …] ] without violating thenon-tangling condition and by moving Ln or constituents containingLn to strictly c-commanding positions and to the left.
(33) The number of possible orders thus generable is given by the followingformula:f(0) = 1
f(n) =n
ÿ
i=1
f(i ´ 1)f(n ´ i)
3 Three-element clusters(34) In clusters we generally know what the hierarchy is (though not what should
count as a satellite), so we can check for conformity with the theory.(35)
S1 S2 L3 L3 S2 S1
S1 L3 S2 S2 L3 S1
7
(36)
L3 S1 S2
t
(37) The remaining logically possible order is 2-1-3. It cannot be generated asan unmarked order without violating some constraint of the theory.
3.1 2-1-3 in three-verb clusters(38) 2-1-3 cluster order comes up in a number of works as an alternate order
(e.g., Heilmann 1999; Schmid and Vogel 2004; Schwalm 2013). They areirrelevant.
(39) 2-1-3 in the West Frisian ‘third construction.’ Some people doubt that‘third constructions’ are proper clusters because they lack IPP effect andhave ‘to.’
(40) … datthat
erhe
myme
datthe
boekbook
ferbeanfofbidden
hathas
teto
lêzenread
that he has prohibited my reading the book Haan 2010b, 204 ex. 22c(41) 2-1-3 in Zürich German (M. Salzmann, p.c., based on Lötscher 1978, 3 fn
2). The order is unmarked.Is this a ‘third construction’? Note the lack of IPP.
(42) a. wowhere
sthey
merme
sthe
gschierdishes
{*hälffenhelp
| ghulffe}helped
händhave
abwascheup.wash
when they helped me clean the dishesb. wo
wheresthey
merme
sthe
gschierdishes
händhave
{hälffenhelp
| ghulffe}helped
abwascheup.wash
(43) …there certainly is clause union and there is no ‘to.’(44) wo
wheresthey
merme
{*(s)it
ghulffehelped
händhave
(*s)it
| *(s)it
händhave
ghulffehelped
(*s)}it
abwascheup.wash
when they helped me clean it (M. Salzmann, p.c.)(45) 2-1-3 in the Vorarlberg.
2-1-3 is obligatory with phase verbs (Schallert 2014). Some varieties showIPP effect, at least when the main verb is intransitive (Shallert:2014a )No doubt this is a cluster.
8
(46) Sishe
feandfound
’sit
nüdnot
gut,good
asthat
arhe
aafangobegin.inf
hathas
rouchosmoke
She didn’t like it that he has started smoking Schallert 2014, 229 ex. 355a(47) Louden 2011 shows that in current Pennsylvania Dutch the 2-1-3 order is
possible and indeed obligatory in clusters with perception verbs, motionverbs, and benefactives as the second member, but not with modals, whichexhibit obligatory 3-1-2 order. With causatives as the second member,we find alternation between 3-2-1 and 2-1-3 with a subtle distinction inmeaning that Louden takes to indicate a lexical status of the causativewhen the order is 2-1-3 and a functional status when the order is 3-2-1.
(48) The examples fromWest Frisian, Zürich German, the Vorarlberg and Liecht-enstein, and Pennsylvania Dutch show that there are verb clusters where2-1-3 is the unmarked or, indeed, the only order. If, as is commonly as-sumed, verb clustering requires clause union in some sense, then these datadisprove Cinque’s (2009, p. 168) conjecture that the linear order of (all)“auxiliary and restructuring (or clause union) verbs (Cinque 2006)” is re-stricted by the generalization of universal 20 from the previous section.
(49) Is there only one type of restructuring?(50) a. CP
C1 TP
T2 VP
V3 CP
C4 TP
T5 VP
V6 …
b. CP
C1 TP
T2 VP
V3 TP
T4 VP
V5 …
c. CP
C1 TP
T2 passP
pass3 VP
V4 …
(51) 2-1-3 comes up only with verbs that have a lexical feel to them not withauxiliaries, modals (and ‘let’). These may be purely scopal. While clustersof the 2-1-3 type might restructure in some sense, auxiliaries, modals (and‘let’) may restructure in the stronger sense of all being satellites of the mainverb.
9
3.2 Three verb clusters with auxiliaries, modals, and ‘let’(52) For clusters made up of Aux1, Mod2, and V3, all five theoretically expected
orders are attested as neutral orders.2-1-3 is absent as a neutral order. (See also Wurmbrand, 2006, to appear.)a. Barbiers, 2005 gives 2-3-1 and 3-2-1 variants in substantial numbers
and 1-3-2 in small but locally coherent numbers for Dutch dialects ofthe Netherlands and Belgium.
b. Seiler, 2004 finds the orders 1-2-3 and 3-1-2 in Swiss German dialects.c. Patocka, 1997, p. 278 finds three orders in the Bavarian dialects of
Austria: 1-3-2, 3-1-2, and 1-2-3.d. Standard German also has 1-3-2 as an unmarked order for Aux>Mod>V
structures.e. None of these authors report the 2-1-3 pattern to be possible.
(53) DutchVerteltell
maarjust
nietnot
wiewho
zijshe
hadhad
kunnencan.inf
roepen.call.inf
Just don’t say who she could have called. Barbiers, 2005, 237 ex. 3(54) Swiss German
Sthe
Telefonphone
häthas
gradjust
glüütet,rung
won=iwhen=I
hanhave
wellewanted
gaago
The phone just started to ring when I wanted to leave. Seiler, 2004,372 ex. 6a
(55) dassthat
erhe
hathas
arbeitenwork
müssenmust
that he has had to work Patocka, 1997, p. 278(56) Indeed, all five of these orders show up as obligatory.
a. 1-2-3 is the only possible order in a large part of the area covered bythe SAND project (Barbiers, Auwera, et al. 2008, 20a).
b. 3-2-1 is obligatory for example in West Frisian (Haan 2010c; Barbiers,Auwera, et al. 2008, 20a).
c. 1-3-2 appears to be obligatory in a number of the Dutch dialects whereit occurs (Barbiers, Auwera, et al. 2008, 20a), it is the most unmarkedorder in standard German for such clusters (Bader and Schmid, 2009).
d. 2-3-1 is the standard order in Afrikaans (Robbers 1997, p. 57) and anumber of places West and South of Antwerp (Barbiers, Auwera, et al.2008, 20a).
e. 3-1-2 is obligatory in some Bavarian dialects (Eroms 2004; Eroms,Röder, and Spannbauer-Pollmann 2006, map 5), in Eastern Hessian
10
(Schwalm 2013, 63 map 8), and in Pennsylvania Dutch (Louden 2011).f. 2-1-3 is never obligatory with Aux1 Mod2 V3 clusters.
(57) Other cluster types using auxiliaries, modals, the causative, and main verbsshow less variability. Neither 2-3-1 nor 2-1-3 show up in Mod1 Aux2 V3,Mod1 Mod2 V3, or Aux1 Aux2 V3 clusters (Wurmbrand to appear, table 2for an overview).
(58) 2-1-3 is attested only as a (marked) alternative order (Heilmann 1999;Schmid and Vogel 2004 etc.).
(59) The only possible true exception that I am aware of are the following ex-amples from Höhle 2006, p. 74.
(60) a. duyou
hesds0have.2ndsg.sbjv
jåprt
li:wǝrather
los2let.inf
khün1caninf
gǝsai3be.inf
you should have preferred to abstain from it attributed to Luthardt1963, p. 370
b. içI
håusn�0have.him.it
los2let.inf
khün1can.inf
gǝmåx3do.inf
I could make him do it.(61) ich
Iwü:ǝrǝn0would.him
nednot
hå:1have.inf
laǝs3let.inf
kön2can.inf
gǝruf4call.inf
I wouldn’t have been able to have him called. attributed to Reichardt1914, p. 207
(62) Maybe ‘let’ was included in error? I haven’t seen Höhle’s sources, so can’tbe sure what is going on and whether these are unmarked.
(63) Overall then, when clusters are restricted to those consisting of auxiliaries,modals, causative ‘let’, and main verbs, we find strong support for thetheory of unmarked word order from section 2.
3.3 Digression: verbal particles(64) What class do particles belong to? Are they low verbal heads (den Dikken
1992; Neeleman and Weerman 1993; Ramchand and Svenonius 2002)?(65) Particles do not scramble, hence are unlike arguments and adverbs and like
verbs.(66) Particles cannot be stranded under partial VP-fronting, hence are unlike
arguments and adverbs and like verbs.(67) Note: adding particles as a low head leaves the conclusions so far untouched.
11
(68) a. Aux-Mod-V, particle variable(i)
Aux1 Mod2 V3 prt4
(ii)
Aux1 Mod2 prt4 V3(iii)
Aux1 prt4 Mod2 V3
t
(iv)
prt4 Aux1 Mod2 V3
t
b. Aux-V-Mod, particle variable
Aux1 V3 prt4 Mod2 Aux1 prt4 V3 Mod2
prt4 Aux1 V3
t
Mod2
c. V-Mod-Aux, particle variable
V3 prt4 Mod2 Aux1 prt4 V3 Mod2 Aux1
d. Mod-V-Aux, particle variable
12
Mod2 V3 prt4 Aux1 Mod2 prt4 V3 Aux1
prt Mod2 V3
t
Aux1
e. V-Aux-Mod, particle variable
V3 prt4 Aux1 Mod2
t
prt4 V3 Aux1 Mod2
t
f. V-Aux-Mod cannot be derived(69) What are the facts?
13
Table 1: Summary of orders for separable prefixesexpected attested
Prt = 4(68)ai 1-2-3-4 yes yes English, Danish, Swedisch, Norwegian…(68)aii 1-2-4-3 yes yes Dutcha
(68)aiii 1-4-2-3 yes yes Dutchb
(68)aiv 4-1-2-3 yes yes Dutchc
1-3-2-4 no no(68)bi 1-3-4-2 yes !(68)bii 1-4-3-2 yes yes Standard German(68)biii 4-1-3-2 yes !
3-2-1-4 no no3-2-4-1 no no
(68)c i 3-4-2-1 yes !(68)c ii 4-3-2-1 yes yes Standard German, Frisian
2-3-1-4 no no(68)di 2-3-4-1 yes !(68)dii 2-4-3-1 yes yes West Flemish(68)diii 4-2-3-1 yes !d
3-1-2-4 no no3-1-4-2 no no
(68)e i 3-4-1-2 yes !(68)e ii 4-3-1-2 yes yes Bavarian
2-1-3-4 no no2-1-4-3 no no2-4-1-3 no no4-2-1-3 no no
a Obligatory in many Belgian varieties of Dutch (Barbiers and Bennis 2007, 31a).b Obligatory in one variety of Dutch in the Netherlands (Barbiers and Bennis2007, 31a).
c Obligatory in a number of varieties of Dutch in the Netherlands (Barbiers andBennis 2007, 31a).
d We can construct the following grammatical Afrikaans example based on Rob-bers 1997, 61 ex. 40a
(i) datthat
hyhe
homhim
wegaway
laatlet
gaango has
het
that he let him leave
but I exclude it here because it is a dispreferred alternative to the 2-4-3-1 order.
14
(70) Good: None of the 10 excluded orders are attested.(71) Bad: Why is V-partcile order so restricted?(72) Abels shows that Wagner 2004, 2005 theory of of nuclear stress predicts
that in 2-3-1 clusters the prosodic phrasing groups 3 and 1 into a singleprosodic constituent to the exclusion of 2. This, of course, goes againstthe syntactic structure, which groups 2 and 3 to the exclusion of 1. Abelssuggests that this mismatch makes the 2-3-1 order dispreferred and difficultto learn. All the theoretically expected but unattested orders (except for the4-1-3-2, more on which below) give rise to the mismatch between syntaxand prosody discussed by Abels. Their absence is therefore not entirelysurprising.
(73) The order 4-1-3-2 is expected but unattested as a neutral order. Koopmanand Szabolcsi 2000 rule this order out categorically, but it is possible forsome speakers as a marked alternative (J. van Craenenbroeck, A. Neeleman,L. Haegeman, L. Aelbrecht, p.c.) in Mod1 Aux2 V3 prt4 clusters. 4-1-3-2in the noun phrase is extremely rare.
(74) We have a good fit between theory and facts and tentatively incorporateparticles.
4 Word order in four-element clusters(75) There is not atlas data for four-element clusters, so the attestations are
more anecdotal.
15
Table 2: Summary of orders for four-element clustersexpected attested
Prt=4 V=4 cause required1-2-3-4 yes yes yesa no1-2-4-3 yes yes yesb no1-4-2-3 yes yes yesc no4-1-2-3 yes yes yesc no1-3-2-4 no no nod
1-3-4-2 yes ! yese no1-4-3-2 yes yes yesf no4-1-3-2 yes ! yesg no3-2-1-4 no no no3-2-4-1 no no no3-4-2-1 yes ! yesh yes4-3-2-1 yes yes yesi no2-3-1-4 no no no2-3-4-1 yes ! yesj
2-4-3-1 yes yes !k4-2-3-1 yes ! yesl yes3-1-2-4 no no no3-1-4-2 no no no3-4-1-2 yes ! yesm yes4-3-1-2 yes yes yesn no2-1-3-4 no no no2-1-4-3 no no !o no2-4-1-3 no no no4-2-1-3 no no nop
a Standard order in Enlish, Danish, Swedish, Norwegian, and Standard Dutch.b Standard order for Mod1 Mod2 Aux3 V4 in Belgian Dutch (Besten, 1981, p. 6)c Standard Dutch (Geerts et al., 1984, p. 600)d Though see example (61) and its discussion abovee Mod1 Aux2 Mod3 V4 clusters in West Flemish (Haegeman 1998b, p. 277; denDikken 1994, p. 83)
f Standard German in Aux1 Mod2 Aux3 V4 clusters (Bader and Schmid, 2009,p. 214), Stellingwerfs Mod1 Mod2 Aux3 V4 clusters (Zwart, 1995, p. 9)
g Afrikaans Mod1 Aux.perf2 Aux.pass3 V4 (Donaldson, 1993, 261 ex. 918)h Zürich German Aux.perf1 Aux.pass2 let3 V4 (M. Salzmann, p.c.). The orderalternates with 1-3-4-2 and 4-3-2-1 but in the latter case with the full participleof ’let.’
i Standard German Mod1 Aux.perf2 Aux.pass3 V4 clusters; West Frisian all clus-ters (Haan, 2010c)
j West Flemish Aux1 Mod2 Mod3 V4 clusters (L. Haegeman, p.c.).k As an alternate in West Flemish Aux.perf1 Mod2 Aux.pass3 V4 clusters (L.Haegeman, p.c.)
l Standard German Mod1 Aux2 let3 V4 “Skandalkonstruktion” (Vogel, 2009)m Afrikaans Mod1 Aux.pass2 let3 V4 (Robbers, 1997, 64 ex. 46)n Standard German(Bader and Schmid, 2009, p. 214), Vorarlberg (Schallert, 2014)o Lindhorst Low German (Bölsing, 2011) discussed below.p Spontaneously, at low frequency, possibly as alternate, in Wurmbrand’s 2004study.
16
(76) Datapoints for geeks – not in the literature:(77) 2-3-4-1 in West Flemish (L. Haegeman, p.c.)1
datthat
HansHans
vufor
zenhis
examenexam
100100
woordenwords
perper
minuutminute
moetenmust
keunencan
typentype
eethas
that Hans must have been able to type 100 words per minute for the exam(78) 3-4-2-1 in Zürich German (M. Salzmann, p.c. – needs completion)2
lalet
liggelie
wordenbecome
ischis
that the bag has been left lying(79) 4-2-3-1 Standard German “Skandalkonstruktion” Vogel, 2009
dassthat
sieshe
esit
sichself
entferntremoved
habenhave
lassenlet
sollshould
that she is supposed to have had it removed(80) 2-4-3-1 West Flemish (L. Haegeman, p.c.)3
a. dajthat-you
diethose
boekenbooks
nienot
keunencan
mee-doenwith-do
eethave
that you could not take those books along with youb. dat
thatdienenthis
boekbook
moetenmust
gelezenread
wordenbecome
isis
that this book has had to be read(81) The unexpected 10 orders are, for the most part, unproblematically unat-
tested, though see the 1-3-4-2 order from example (61).
4.1 2-1-4-3 in Lindhorster Platt(82) hei
hewerd0will
kont2could.sup
hemn1have.inf
mor’ntomorrow
emaat4mowed
hemn3have.inf
he will have been able to have mowed tomorrow Bolsing20111See Haegeman 1998a,b, 2001 for details on the form and placement of the auxiliary in West
Flemish.2The order alternates with 4-3-2-1 and 1-3-4-2 but then with the full participle of ’let.’3Aux.perf1 Mod2 Aux.pass3 V4 clusters in West Flemish alternatee with 1-2-4-3, 1-4-3-2, and
4-1-2-3.
17
(83) In two- and three-element clusters auxiliaries follow the participles andsupines they govern and modals precede the infinitives they govern in thisdialect.
(84) Three element clusters are well-behaved in the sense that they do not sur-face with 2-1-3 orders.
(85) While it might seem tempting to try to wriggle out of example (82) byclaiming that it really instantiates the expected 2-3-4-1 order on the basisof the repetition of the auxiliary ‘hemn,’ such an attempt would do violenceto the very strong generalizations just cited. The attempt also fails forexamples (94) below.
(86) These generalization produce the otherwise rare 2-3-1 order in the futureperfect of modals in main clauses:
(87) ekI
nieme an,assume
heihe
werd0will
kont2can
gautgood
schläapen3sleep
hemn1have
I assume that he will have been able to sleep well Bölsing 2011, p. 215(88) The expected form for example (82) is therefore the following:(89) hei
hewerd0will
kont2could.sup
mor’ntomorrow
emaat4mowed
hemn3have.inf
hemn1have.inf
(90) This is bad because of haplology (see also Bölsing 2011, 217 fn 36) andbecause of prosody.
(91) The dialect might have a non-movement way of resolving haplology, though.There is no licensor either for the supine or for the participle.Maybe ‘hemn’ is syncopated?Regular forms with one of the two in the infinitives exist and show regularmeaning.
(92) heihe
werdwill
kontcould.sup
mor’ntomorrow
emaatmowed
hemnhave.inf
he will have been able to have mowed tomorrow Bolsing2011(93) The haplology reasoning does not apply to the expected 2-(5-)4-3-1 order in
the following examples, though, while the prosodic reasoning does. Bölsingreports all of them with 2-1-(5-)4-3 order:
(94) a. datthe
Päardhorse
werd0will
kont2can.sup
hemn1have.inf
vandäagetoday
nichshoee
beschläan4become.inf
wiern3
The horse will have been able to be shoed today Bölsing 2011, p. 216b. dat
thePäardhorse
werd0will
scholt2must.sup
hemn1have.inf
vandäagetoday
beschläan5shoed
18
wuern4become.sup
sein3be.inf
The horse will not have had to have been shoed today. Bölsing 2011,p. 216
c. däi
theWäagencar
werd0will
most2must.sup
hemn1have.inf
alalready
längstlongest
estrieken4painted
sein3be.infThe car will have had to be painted a long time ago. Bölsing 2011,p. 217
d. däi
theWäagencar
werd0will
most2must.sup
hemn1have.inf
alalready
längstlongest
estrieken5painted
sein44
be.infhemn3have.inf
The car will have had to have been painted a long time ago. Bölsing2011, p. 217
(95) I do not know if these orders alternate, but Bölsing does not suggest thatthey do.
(96) The examples are excentric for further reasons.a. In verb clusters scope bearing elements take scope over (not neces-
sarily all) material to their right. The Lindhorster forms flout thisgeneralization, (94b).
b. The supines of the modal verbs seem to drift towards the Wackernagelposition, preceding all other material in the middle field:
(97) AlseWhen
ekI
ümmeat
halwighalf
sessesix
nochstill
niksnothing
harrehad
kläipernbang
hüer’n,heard,
hawwehad.1stsg
kontcan.sup
ekI
mei
megleikimmediately
alprt
watsomething
denken.think
When I still didn’t hear any banging at half past five, I was immediatelyable to draw some conclusions.
(98) Lindhorster Platt might be a real counterexample in which case we have toeither retreat to the position of Abels and Neeleman 2012a: Leftward move-ment is not categorically ruled out but only disfavored or assume that hereelements other than the verbal head move (to the Wackernagel position).
4The infinitive of the passive auxiliary is clearly the wrong choice of form here, since thesupine ‘esein’ is required. The error seems to have been caused by cutting and pasting. This,together with the fact that Bölsing marks this form as very rare, casts some doubt on the realityof this particular data point.
19
5 Discussion(99) I have highlighted the concepts of head, satellite, and classes of satellites
underpinning Cinque’s theory of neutral word order.(100) I have shown that the theory applies nicely to verb clusters if we assume
that modals are functional but perception verbs, phase verbs, benefactives,etc. are not.
(101) We have not considered the question here of how the ordering amongsatellites in different classes interacts. We know that they can be inter-spersed without forming a rigid single hierarchy (Bobaljik 1999; Nilsen2003; Svenonius 2002). We also know that they can stack up on differentsides of a head in classical nesting patterns and that two sets of satellitescan stack up independently on the left in the cross-serial pattern. Thecross-serial pattern might be a special (trivial) case of Bobaljik’s ‘shuf-fling together.’ What interactions there are to the right of the lexical headis less clear. the order amongst satellites of the same class is more flexibleafter the head but interactions between classes may be more restricted.This conjecture is supported by the compactness of verb clusters, a prop-erty which means that verb clusters may be interrupted by non-verbalmaterial to the left of the head but not to its right (Bobaljik 2004 amongothers). Further, Abels, 2007 claims that the mirror image of the SwissGerman and Dutch cross-serial pattern is never found, which again sug-gests an asymmetry in the interactions between orders. A systematicstudy of these questions and a general theoretical analysis do not exist.
(102) It seems unlikely that a movement approach to cross-class interactionswill give satisfactory results. We saw that, on the assumption that move-ment of a limited kind is implicated in generating within-class orders,paradoxes arises for across-class orders. Moreover, the reordering that wefind, for example in dialects with verb projection raising, do not have theusual properties of movement (see Salzmann 2011). At the same time, itis not obvious how to generalize existing non-movement accounts, whichtypically involve the upward inheritance of theta-grids, to optional mod-ifiers. A deeper rethinking of structure-building operations will likely benecessary.
ReferencesAbels, Klaus (2007). “Towards a Restrictive Theory of (Remnant) Movement”. In:
Linguistic Variation Yearbook 7, pp. 53–120.
20
Abels, Klaus (2015). “Syntax”. In: Exploring Language and Linguistics. Ed. byNatalie Braber, Louise Cummings, and Liz Morrish. Cambridge: CambridgeUniversity Press. Chap. 6, pp. 137–167.
Abels, Klaus and Ad Neeleman (2012a). e. Talk presented at the Workshop onempty categories, GLOW 35.
— (2012b). “Linear asymmetries and the LCA”. In: Syntax 15.1, pp. 25–74.Bader, Markus and Tanja Schmid (2009). “Verb clusters in colloquial German”.
In: Journal of Comparative Germanic Linguistics 12.3, pp. 175–228.Barbiers, Sjef (2005). “Word order variation in three-verb clusters and the division
of labour between generative linguistics and sociolinguistics”. In: Syntax andvariation : reconciling the biological and the social. Ed. by Leonie Cornips andKaren P Corrigan. Philadelphia, PA: John Benjamins, pp. 233–264.
Barbiers, Sjef, Johan van der Auwera, et al. (2008). SAND: Syntactische Atlas vande Nederlandes Dialecten Deel II / Syntactic Atlas of the Dutch Dialects PartII. Amsterdam: Amsterdam University Press.
Barbiers, Sjef and Hans Bennis (2007). “The Syntactic Atlas of the Dutch Dialects”.In: Nordlyd 34, pp. 53–72.
Besten, Hans den (1981). “Some accompanying remarks on den Besten and Ed-mondson ‘The verbal comples in continental West Germanic’”. In: GroningerArbeiten zur Germanistischen Linguistik 19, pp. 2–10.
Bobaljik, Jonathan D (1999). “Adverbs: the hierarchy paradox”. In: GLOT Inter-national 4.9, pp. 27–28.
— (2004). “Clustering Theories”. In: Verb Clusters: A Study of Hungarian andDutch. Ed. by Katalin É. Kiss and Henk van Riemsdijk. Philadelphia: JohnBenjamins, pp. 121–145.
Bölsing, Friedrich (2011). Niederdeutsche Sprachlehre: Plattdeutsch im KirchspielLindhorst, Schaumburg-Lippe. Hildesheim: Olms.
Cinque, Guglielmo (2005). “Deriving Greenberg’s Universal 20 and its Exceptions”.In: Linguistic Inquiry 36.3, pp. 315–332.
— (2006). ““Restructuring” and the Order of Aspectual and Root Modal Heads”.In: Restructuring and Functional Heads. Ed. by Guglielmo Cinque. Vol. 4. TheCartography of Syntactic Structures. New York, New York: Oxford UniversityPress, pp. 81–98.
— (2009). “The fundamental left-right asymmetry of natural languages”. In: Uni-versals of Language Today. Ed. by Sergio Scalise, Elisabetta Magni, and An-tonietta Bisetto. Dordrecht: Springer, pp. 165–184.
— (2014). On the movement account of Greenberg’s Universal 20: Refinementsand Replies. Ms. Università Ca’Foscari.
den Dikken, Marcel (1992). “Particles”. Ph.D. Thesis.
21
den Dikken, Marcel (1994). “Minimalist Verb (Projection) Raising”. In: GroningerArbeiten zu germanistischen Linguistik 37, pp. 71–88.
Donaldson, Bruce C (1993). A grammar of Afrikaans. Berlin & New York: Moutonde Gruyter.
Dryer, Matthew S. (2009). “On the order of demonstrative, numeral, adjective,and noun: an alternative to Cinque”. In: Conference on Theoretical Approachesto Disharmonic Word Orders. Newcastle.
Dubenion-Smith, Shannon Andrew (2010). “Verbal Complex Phenomena in WestCentral German : Empirical Domain and Multi-Causal Account”. In: Journalof Germanic Linguistics 22.2, pp. 99–191.
Eroms, Hans-Werner (2004). “Wortstellung bei Modalverbkonstruktionen im Mit-telbairischen”. In: Kreuther Kräuterburschen: Beiträge zur 9. Bayerisch-österreichischenDialektogentagung in Wildbab Kreuth. Ed. by Ulrich Kanz and Alfred Wild-feuer. Regensburger Dialektforum. Regensburg: Edition Vulpes, pp. 151–170.
Eroms, Hans-Werner, Birgit Röder, and Rosemarie Spannbauer-Pollmann, eds.(2006). Bayerischer Sprachatlas / Regionalteil V: Sprachatlas von Niederbayern(SNiB) / Band 1: Einführung mit Syntaxauswertung. Vol. V. 1. Heidelberg:Universitätsverlag Winter.
Geerts, G et al., eds. (1984). Algemene Nederlandse Spraakkunst. Groningen, Leu-ven: Wolters.
Greenberg, Joseph (1963). “Some universals of grammar with particular referenceto the order of meaningful elements”. In: Universals of Language. Ed. by JosephGreenberg. Cambridge, MA: MIT Press, pp. 73–113.
Haan, Germen de (2010a). Studies in West Frisian Grammar. Vol. 161. LinguistikAktuell. Amsterdam & Philadelphia: John Benjamins Publishing Company.
— (2010b). “The third construction in Frisian”. In: Studies in West Frisian Gram-mar. Vol. 161. Linguistik Aktuell. Amsterdam & Philadelphia: John BenjaminsPublishing Company. Chap. 9, pp. 199–214.
— (2010c). “The verbal complex”. In: Studies in West Frisian Grammar. Vol. 161.Linguistik Aktuell. Amsterdam & Philadelphia: John Benjamins PublishingCompany. Chap. 8, pp. 165–198.
Haegeman, Liliane (1998a). “Verb movement in embedded clauses in West Flem-ish”. In: Linguistic Inquiry 29.4, pp. 631–656.
— (1998b). “V-Positions and the Middle Field in West Flemish”. In: Syntax 1.3,pp. 259–299.
— (2001). “Antisymmetry and verb-final order in West Flemish”. In: The Journalof Comparative Germanic Linguistics 3.3, pp. 207–232.
Haspelmath, Martin et al., eds. (2005). The World Atlas of Language Structures.Oxford: Oxford University Press.
22
Heilmann, Axel (1999). “Die VP im Schwäbischen”. PhD thesis. Stuttgart: Uni-versität Stuttgart.
Höhle, Tilman N. (2006). “Observing non-finite verbs : Some 3V phenomena inGerman-Dutch”. In: Form, Structure, and Grammar : A Festschrift Presentedto Günther Grewendorf on Occasion of his 60th Birthday. Ed. by Patrick Brandtand Eric Fuß. Vol. 63. Studia Grammatica. Berlin: Akademie Verlag, pp. 55–77.
Kaufmann, Göz (2007). “The verb cluster in Mennonite Low German : A newapproach to an old topic”. In: Linguistische Berichte 210, pp. 147–207.
Koopman, Hilda and Anna Szabolcsi (2000). Verbal Complexes. Cambridge, MA:MIT Press.
Lötscher, Andres (1978). “Zur Verbstellung im Zürichdeutschen und in anderenVarianten des Deutschen”. In: Zeitschrift für Dialektologie und Linguistics 45,pp. 1–29.
Louden, Mark L (2011). “Synchrony and diachrony of verb clusters in PennsylvaniaDutch”. In: Studies on German-language Islands. Ed. by Michael T Putnam.Vol. 123. Studies in Language Companion Series. Amsterdam & Philadelphia:John Benjamins Publishing Company, pp. 165–185.
Luthardt, E (1963). “Mundart und Volkstümliches aus Steinach, Thüringerwaldund dialektgeographische Untersuchungen im Landkreis Sonneberg, im Amts-gerichtsbezirk Eisfeld, Landkreis Hildburgshausen und in Scheibe im Amts-gerichtsbezirk Oberweißbach, Landkreis Rudolstadt”. PhD thesis. Hamburg.
Neeleman, Ad and Fred Weerman (1993). “The Balance between Syntax and Mor-phology: Dutch Particles and Resultatives”. In: Natural Language and Linguis-tic Theory 11 (3), pp. 433–476.
Nilsen, Øystein (2003). Eliminating Positions: Syntax and semantics of sentencemodification. Utrecht, NL: LOT.
Öhlschläger, Günther (1989). Zur Syntax und Semantik der Modalverben des Deutschen.Vol. 144. Linguistische Arbeiten. Tübingen: Niemeyer.
Patocka, Franz (1997). Satzgliedstellung in den bairischen Dialekten Österreichs.Frankfurt a. M: Peter Lang.
Pearson, Matthew (2000). “Two Types of VO Languages”. In: The Derivationof VO and OV. Ed. by Peter Svenonius. Linguistik Aktuell. Amsterdam andPhiladelphia: John Benjamins, pp. 327–364.
Ramchand, Gillian and Peter Svenonius (2002). “The Lexical Syntax and LexicalSemantics of the Verb-Particle Construction”. In: Proceedings of WCCFL 21.Ed. by L. Mikkelsen and Chris Potts. Summerville: Cascadilla Press, pp. 101–114.
Reichardt, Edinhard (1914). Die Wasunger Mundart, 2. Teil. Vol. 71. Schriften desVereins für Sachsen-Meinigische Geschichte und Landeskunde. Hildburgshausen.
23
Reis, Marga (2001). “Bilden Modalverben im Deutschen eine syntaktische Klasse?”In: Modalität und Modalverben im Deutschen. Ed. by Marga Müller Reimar undReis. Vol. 9. Linguistische Berichte Sonderheft. Helmut Buske Verlag, pp. 287–318.
Robbers, Karin (1997). “Non-finite verbal complements in Afrikaans”. PhD thesis.Amsterdam: University of Amsterdam.
Salzmann, Martin (2011). “Resolving the movmeent paradox in Verb ProjectionRaising. In favor of base-generation and covert predicate raising”. In: EmpiricalIssues in Syntax and Semantics. Ed. by Olivier Bonami and P. Cabredo Hofherr.Vol. 8. CNRS, pp. 453–485.
Schallert, Oliver (2014). Zur Syntax der Ersatzinfintivkonstruktion: Typologie undVariation. Vol. 87. Studien zur deutschen Grammatik. Tübingen: Stauffenburg.
Schmid, Tanja and Ralf Vogel (2004). “Dialectal Variation in German 3-Verb Clus-ters: A Surface-Oriented Optimality Theoretic Account”. In: Journal of Com-parative Germanic Linguistics 7, pp. 235–274.
Schwalm, Johanna (2013). “Verbalsyntax in den Mundarten Hessens: eine Un-tersuchung anhand der SyHD-Materialien”. MA thesis. Marburg a. d. Lahn:Philipps-Universität Marburg.
Seiler, Guido (2004). “On three types of dialect variation, and their implicationsfor linguistic theory. Evidence from verb clusters in Swiss German dialects”. In:Dialectology meets typology : dialect grammar from a cross-linguistic perspective.Ed. by Bernd Kortmann. Vol. 153. Trends in Linguistics. Berlin: Mouton deGruyter, pp. 367–399.
Svenonius, Peter (2002). “Subject Positions and the placement of adverbials”. In:Subjects, expletives, and the EPP. Ed. by Peter Svenonius. Oxford and NewYork: Oxford University Press, pp. 201–242.
Vogel, Ralf (2009). “Skandal im Verbkomplex: Betrachtungen zur scheinbar inkor-rekten Morphologie in infiniten Verbkomplexen des Deutschen”. In: Zeitschriftfür Sprachwissenschaft 28.2, pp. 307–346.
Wagner, Michael (2004). “Prosody as a Diagonalization of Syntax. Evidence fromComplex Predicates”. In: Proceedings of the North East Linguistic Society 34.Ed. by Keir Molton and Matthew Wolf. Vol. 2. University of Massachusetts atAmherst: Graduate Linguistic Student Association, pp. 587–602.
— (2005). “Prosody and Recursion”. PhD thesis. MIT.Wurmbrand, Susanne (1998). “Infinitives”. PhD thesis. Cambridge: Massachusetts
Institute of Technology.— (2004). “West Germanic verb clusters: The empirical domain”. In: Verb Clus-
ters: A Study of Hungarian, German and Dutch. Ed. by Katalin É. Kiss. JohnBenjamins, pp. 43–85.
24
Wurmbrand, Susanne (2006). “Verb Clusters, Verb Raising, and Restructuring”.In: The Blackwell Companion to Syntax. Ed. by Martin Everaert and Henk vanRiemsdijk. Vol. 5. Blackwell. Chap. 75, pp. 229–343.
— (to appear). “Verb Clusters, Verb Raising, and Restructuring”. In: The Black-well Companion to Syntax. Ed. by Martin Everaert and Henk van Riemsdijk.2nd revised edition. Blackwell Publishing.
Zwart, C. Jan Wouter (1995). A note on verb clusters in the Stellingwerfs dialect.
25