This paper focuses on intervention effects obtained by embedding a topic constituent (either a displaced topic or a clitic left-dislocated topic) within the domain of wh-movement. We present the results of two acceptability judgment tests carried out in European Portuguese (EP), which indicate that only a subset of the constructions in which a topic intervenes in the path of wh-movement is judged acceptable by native speakers. The pattern that emerges can be described by the following generalization: (1)
Although the construction known as “Clitic Left Dislocation” (henceforth, CLLD) in Romance may appear in embedded contexts, such as (1), embedding a CLLDed topic under the domain of wh-movement has been reported to yield deviant results in European Portuguese (EP) (
(1)
a.
Não
not
sei
.know.
se,
if
à
to.the
Maria,
Maria
lhe
vou
go.
oferecer
give.
esse
that
livro.
book
‘I don’t know if I’ll give that book to Maria.’
b.
*Não
not
sei
know.
ainda
yet
que
which
prenda,
gift
à
to.the
Maria,
Maria
lhe
vamos
go.
oferecer
give.
no Natal
at.the Christmas
‘I don’t know yet which gift we will get for Maria this Christmas.’
c.
Não
not
sei
know.
ainda
yet
a
to
quem
whom
a
the
Maria
Maria
vai
go.
oferecer
offer.
essa
this
prenda.
book
‘I don’t know to whom Maria will offer this book.’
The following sentences show that the restriction against an intervening topic also applies in relative clauses (2) and in root questions introduced by a d-linked wh-constituent (3).
(2)
Relative clause
a.
Já
already
li
read.
os
the
artigos
articles
que
that
aquela
that
revista
journal
não
not
quis
want.
publicar
publish.
‘I have already read the articles that that journal didn’t want to publish.’
b.
*/?Esse
that
é
is
o
the
editor
editor
que,
that
os
the
teus
your
artigos,
articles
nunca
never
os
them
quis
wanted
publicar
publish.
nessa
in.that
revista.
journal.
‘That is the editor that never wanted to publish your articles in that journal.’
(3)
Root question with d-linked wh
a.
Que
which
exame
exam
a
the
Maria
Maria
está
be.
a
at
pensar
think.
mostrar
show.
a
to
esse
that
médico?
doctor
‘Which exam is Mary thinking of showing to that doctor?’
b.
*Que
which
exame,
exam
a
to
esse
that
médico,
doctor
estás
be.
a
at
pensar
think.
mostrar-lhe?
show.
‘Which exam are you thinking of showing to that doctor?’
There are two main lines of analysis of CLLD in the literature. One influential approach is that of Rizzi (
(4) | V [CP [C se] [TopP [ à Maria ] [Top’ Top … [ TP lhe vou dar esse livro ]]]] |
The other approach (
(5) | V [CP [C se] [TP [ à Maria ] [ TP lhe vou dar esse livro ]]] |
Under both accounts, one might attempt to rule out the deviant cases above by invoking minimality. On the assumption that the relative clause in (2) involves movement of a null operator to Spec-CP, in all of the deviant examples above the topic intervenes between the wh-operator/phrase and its base position. For concreteness, we quote Rizzi’s (
(6) | Y is in a Minimal Configuration (MC) with X iff there is no Z such that | |
1. | Z is of the same structural type as X, and | |
2. | Z intervenes between X and Y |
Assuming that both CLLDed topics and wh-phrases occupy A-bar positions, the presence of the topic in (1b), (2b) and (3b) would prevent chain formation to take place between the wh-phrase (or the null operator in [2b]) in Spec, CP and its trace.
(7) | [CP |
Rizzi (
(8) | Italian ( |
|
a. | un uomo a cui, il premio Nobel, lo daranno senz’altro | |
‘a man to whom, the Nobel Prize, they will give it undoubtedly.’ Rizzi ( |
||
b. | ?Mi domando a chi, il premio Nobel, lo potrebbero dare. | |
‘I wonder to whom, the Nobel Prize, they would give it.’ Rizzi ( |
Haegeman (
(9)
French (Haegaman 2012: 57–58)
a.
Voici
here.is
l’étudiant
the’student
à
to
qui,
whom
ton
your
livre,
book
je
I
le
it
donnerai.
give.
‘Here is the student to whom, this book, I will give.’
b.
J’aimerais
I’like.
savoir
know.
à
to
qui,
whom,
ton
your
texte,
text,
tu
you
comptes
count.
le
it
montrer
show.
d’abord.
first
‘I would like to know who, your text, you intend to show to first.’
Rizzi (
Concerning EP, there is one observation that renders a minimality account of the deviance of (1b, 2b, 3b) hard to maintain. Besides CLLD, EP has another means of expressing the Topic-Comment relation, whereby the topicalized element is simply associated with a gap in argument position and no clitic is present (
(10)
[Esse
that
manuscrito],
manuscript
enviei
send.
[-]
[-]
hoje
today
para
to
a
the
editora.
publishers
‘That manuscript, I sent today to the publishers.’
Now let us consider the topicalized counterparts to (1b), (2b) and (3b):
(11)
??Não
not
sei
know.1
ainda
yet
que
which
prenda,
gift
à
to.the
Maria,
Maria
vamos
go.1
oferecer
offer.
no
at.the
Natal.
Christmas
‘I don’t know which gift, to Maria, we will offer for Christmas.’
(12)
??Esse
that
é
is
o
the
editor
editor
que,
that
os
the
teus
your
artigos,
articles
nunca
never
quis
wanted.
publicar
publish.
nessa
in.that
revista.
journal.
‘That is the editor that, your articles, never wanted to publish in that journal.’
(13)
??Que
which
exame,
exam
a
to
esse
that
médico,
doctor
estás
be.
a
at
pensar
think.
mostrar?
show.
‘Which exam are you thinking of showing to that doctor?’
Even though examples such as (11–13) are not fully acceptable, they are clearly better than (1b), (2b) and (3b), with a clitic.
(14)
a.
CLLD: no strong cross-over effects
Esse
that
rapaz,
boy
disseram-me
tell.
que
that
sabe
know.
que
that
não
not
oi
procurámos.
search.
‘That boy, they told me that he knows that we didn’t search for him.’
b.
Topicalization: strong cross-over effects
*Esse
that
rapazi,
boy
disseram-me
tell.
que
that
sabe
know.
que
that
não
not
procurámos [-]i.
search.
(Examples adapted from
Previous analyses of CLLD and topicalization in EP indeed assume that the latter involves some sort of A-bar movement whereas the former doesn’t.
Another relevant observation concerning this type of intervention effects is that there is a contrast between high dative Experiencers (15) and lower dative Goals (2b), regardless of the presence of the clitic. The following examples are taken from Barbosa (
(15)
Vi
saw.1
hoje
today
a
the
casa
house
que
that
à
to.the
Maria
Maria
mais
more
(lhe
(
convém
is.convenient
comprar
buy.
‘Today I saw the house that, for Maria, it is more convenient to buy.’
The following examples illustrate other contexts in which a preverbal dative Experiencer is allowed with or without a resumptive clitic (our own judgments):
(16)
Sabes
know.2
quando,
when
ao
to.the
Pedro,
Pedro,
mais
more
(lhe)
to.him
convém
is convenient
lá
there
ir?
go.
‘Do you know when, to Pedro, it is more convenient to go there?’
(17)
Que
which
discos,
records,
ao
to.the
João,
João,
mais
more
(lhe)
to.him
agradará
please.
receber?
receive.
‘Which records will please João the most?’
Curiously, in this case, the presence of the clitic doesn’t appear to contribute to a decrease in acceptability.
In their seminal paper on the topic, Belletti & Rizzi (
These observations, however, have never been checked against a sufficiently large pool of native speaker informants. In this paper, we report on the results of two acceptability judgment tasks designed to determine the extent to which such structures with multiple dependencies are accepted by native speakers of EP and the factors that contribute to improved acceptability.
Study 1 (Section 2) seeks to determine the effect of embedding a fronted dative Experiencer as opposed to a fronted dative Goal within the domain of wh-movement, while at the same time controlling for the effect of the presence of the clitic. Our results show that dative Goals yield lower ratings than dative Experiencers. We will examine the hypothesis that preverbal dative Experiencers are quirky subjects in light of this result. We will discuss evidence that the clitic that occurs with preverbal Experiencers is a marker of inherent case, but we will argue that the pattern of responses obtained for the dative Experiencers does not quite warrant the conclusion that preverbal dative experiencers are subjects. We will examine the problems raised by the quirky subject analysis of these preverbal dative Experiencers and we will conclude that there is not reason to think that they are not topics.
Dative Experiencer arguments differ from dative Goals in two ways: (i) they are assigned inherent case; (ii) in the base, they are located higher than the nominative argument (the Theme). In (15–17), the Theme is the infinitival clause, which contains the trace of wh-movement. Schematically, we have the structure in (18) for (15–17), in which the base position associated with the Experiencer is higher than the trace of the wh-phrase:
(18) | [ |
In this configuration, the full topic-(cl)-gap dependency is contained within the path of wh-movement. In the examples with a dative Goal topic (1b, 3b), by contrast, the trace of the wh-phrase/operator c-commands the argument position associated with the topic, as illustrated in (19):
(19) | [ |
Likewise, (2b) also involves a crossing dependency (the trace of the wh-subject c-commands the argument position associated with the object CLLDed topic). Thus, while (15–17) involve nested dependencies, (1b, 2b, 3b) involve crossing dependencies. Nested dependencies have been argued to be favored over crossing dependencies quite generally (
In order to evaluate whether a version of the no-crossing constraint is operative in these multiple dependency constructions, a second acceptability judgement task was created (Section 3). This time, the function of the topic was held constant and we manipulated the height of the base position of the relative operator, subject
(20)
Esse
that
é
is
o
the
editor
editor
[
que,
that
os
the
teus
your
artigos
articles
nunca
never
[-]i
(osk)
(
quis
want.
publicar [-]
publish.
nessa
in.that
revista].
journal
‘That is the editor that never wanted to publish your articles in that journal.’
(21)
Esse
that
é
is
o
the
revisor
referee
a
to
quemk,
whomk
os
the
teus
your
textosi,
textsi,
já
already
(os)
(cl.
tentei
try.
enviar [-]i [-]k
send.
várias
several
vezes.
times
‘That is the referee to whom, your texts, I have already tried to send several times.’
An ordinary logistics regression analysis revealed a significant difference between the two conditions:
(22) |
We discuss the challenges faced by a purely syntactic account of (22) and suggest that a more promising line of approach would be to derive (22) from processing constraints. Viewed from a parsing perspective, (22) is equivalent to the claim that maintaining an active filler in the course of processing a wh-filler gap dependency is costly for the human processor. Our hypothesis is the following: if, at the point of retrieval of a wh-filler, an active topic (i.e., a displaced topic looking to be integrated with its subcategorizer) intervenes between the wh-gap and the wh-filler, the amount of cognitive resources required to retrieve and integrate the wh-filler in the representation raises to a threshold that results in perception of unacceptability.
One result of the present study that would receive a natural explanation on these grounds is the effect of the presence of the clitic. Previous studies on the processing of CLLD (
The goal of this study is to determine the effect of embedding a fronted dative Experiencer as opposed to a fronted dative Goal within the domain of wh-movement, while at the same time controlling for the effect of the presence of the clitic.
Sixty-six undergraduate and postgraduate students from University of Minho participated in Experiment 1 on a voluntary basis (average age 21,6, SD = 6,8; 51 females, 15 male). All participants were native speakers of European Portuguese, naive regarding the purpose of the experiment and reported not to speak any other language at home.
Twelve experimental sentences and twenty-four fillers were created for this experiment. The experimental items consisted of relative clauses containing dative phrases in pre-verbal position (see Appendix). The sentences were manipulated to examine the role of two independent factors: (i) the theta-role of the dative phrase and (ii) presence
(23)
Experimental conditions (Study 1)
a.
Dative-Experiencer and clitic present (DatExp-Clit)
Já
already
li
read.
o
the
artigo
article
que,
that
ao
to.the
editor
editor
mais
more
to.him.
interessa
interest.
publicar
publish.
nesta revista.
in.this journal
‘I already read the article that the editor is most interested in publishing in this journal.’
b.
Dative-Experiencer and clitic absent (DatExp-NoClit)
Já
already
li
read.
o
the
artigo
article
que,
that
ao
to.the
editor
editor
mais
more
interessa
interest.3
(…)
(…)
c.
Dative-Goal and clitic present (DatGoal-Clit)
Já
already
li
read.
a
the
mensagem
message
que,
that,
ao
to.the
diretor,
director,
tanto
so
querem
want.
enviar
send.
ainda
still
hoje.
today
‘I already read the message that they want so much to send to the director today.’
d.
Dative-Goal and clitic absent (DatGoal-NoClit)
Já
already
li
read.
a
the
mensagem
message
que,
that,
ao
to.the
diretor,
director,
tanto
so
querem
want.
enviar
send.
(…)
(…)
The experimental items were created so as to maximize naturalness and balance between the two sets of conditions. Even though (23a, b) and (23c, d) are not structurally identical (in the latter case the wh-object and the dative are coarguments; in the former, they are not), they are identical from the point of view of information structure: in both cases, the dative is topical and the rest of the clause constitutes the comment. The reason why we have decided to embed the trace of the wh-operator in an infinitival complement in all conditions is that, otherwise, the two sets of examples would differ with respect to the function of the relativized argument. In the Experiencer conditions, the relativized argument would be a subject and, in the dative condition, it would be a direct object, as shown in (24):
(24)
a.
Já
already
li
read.
os
the
livros
books
que,
that
ao
to.the
editor,
editor,
(lhe)
(
interessam
interest.
mais.
most
‘I already read the books that interest the editor the most.’
b.
Já
already
li
read.
os
the
livros
books
que,
that
ao
to.the
editor,
editor,
(lhe)
(
oferecemos
offer.
no Natal.
in.the Christmas
‘I already read the books that we offered to the editor for Christmas.’
Since the different function of the relativized argument would introduce an unwanted variable in the paradigms tested, our option was to embed it in an infinitival clause and keep the examples as identical as possible. So as to maximize the parallelism between the two sets of conditions, in the DatGoal-Clit condition, the clitic associated with the lower verb appears attached to the superordinate verb (all of the superordinate verbs chosen in this condition allow for clitic-climbing). In order to avoid a possible adjacency effect, we inserted an adverbial or quantifier between the fronted dative and the clitic.
To minimize the identification of the experimental items due to repetition of the syntactic structure, the twenty-four fillers also had repeated syntactic structures. Three sets of eight sentences were created (see 25 a–c): adverbial phrases introduced by “in general”; subordinate clauses prior to the main clause, all preceded by “while”; and syntactically varied sentences.
(25) | Fillers examples | |
a. | In general—NP—V(to be)—adjective-adjunct/complement | |
Em geral, as igrejas são bonitas na cidade de Florença. | ||
‘In general, churches are beautiful in the city of Florence.’ | ||
b. | [While—NP—Intransitive V]SUB [NP—VP]MAIN CLAUSE | |
Enquanto o rapaz estudava, a professora corrigia os testes de matemática. | ||
While the boy was studying, the teacher corrected the math tests.’ | ||
c. | [While—NP—Intransitive V]SUB [NP—VP]MAIN CLAUSE | |
A casa da Bárbara foi totalmente decorada pelos amigos do seu irmão. | ||
‘Barbara’s house was totally decorated by her brother’s friends.’ |
Participants received an access link from a teacher or one of the researchers and the task was performed in the classroom through the
Although we are aware of the inherent subjectivity of responses obtained through scales such as the one we used, we believe that this measure suits the purposes of the present study. Despite the controversies surrounding methodological and data analysis issues, the use of
The data were analyzed in the R statistics software, version 3.5.1 (
Study 1 – Generalized linear mixed model fitted by using Markov chain Monte Carlo techniques (
Coefficient | Lower bound (CI 95%) | Upper bound (CI 95%) | pMCMC | |
---|---|---|---|---|
Function_Exp | 1.075 | 0.626 | 1.444 | <0.001*** |
Clitic_NoClit | 0.125 | –0.059 | 0.354 | 0.240 |
Function_Exp:Clitic_NoClit | –0.218 | –0.471 | 0.013 | 0.090. |
Response ~ Function*Clitic, random = ~Participant + Item, data = Est1, verbose = FALSE, family = “ordinal”. Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1.
The data from Study 1 revealed a significant effect for the “Function” factor (
Mean and Median acceptability ratings on a scale from 1 to 7 for Study 1.
The contrast between the “dative-Goal” and “dative-Experiencer” structures, shown above, can clearly be seen from Figure
Distribution of acceptability ratings on a 7-point Likert scale in the Study 1 (1 = ‘not acceptable’; ‘7 = totally acceptable’).
In addition to the statistical approach described above for ordinal data, we also fitted a linear mixed model with Gaussian family by using MCMCglmm package. We also used the lmer function with the lme4 package (
Study 1 – Generalized linear mixed model fitted by using Markov chain Monte Carlo techniques (
Coefficient | Lower bound (CI 95%) | Upper bound (CI 95%) | ||
---|---|---|---|---|
Function_Exp | 1.450 | 0.907 | 1.925 | <0.001*** |
Clitic_NoClit | 0.204 | –0.070 | 0.474 | 0.172 |
Function_Exp:Clitic_NoClit | –0.334 | –0.742 | 0.042 | 0.100. |
Response ~ Function*Clitic, random = ~Participant + Item, data = Est1, verbose = FALSE, family = “gaussian”. Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1.
Study 1 – Generalized linear mixed model fitted by using lmer function in the lme4 package (with Gaussian family).
Coefficient | Std.Error | |||
---|---|---|---|---|
Function_Exp | 1.447 | 0.250 | 5.791 | <0.001*** |
Clitic_NoClit | 0.210 | 0.141 | 1.487 | 0.137 |
Function_Exp:Clitic_NoClit | –0.343 | 0.200 | –1.715 | 0.087 |
Response ~ Function*Clitic, random = ~Participant + Item, data = Est1, verbose = FALSE, family = “gaussian”. Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1.
These results confirm that, irrespective of the presence of the clitic, dative Experiencers differ from dative Goals. The latter yield lower ratings. Even though the clitic factor was not statistically significant, an interesting pattern emerges from the data. While the presence of the clitic contributes to lower acceptability rates with dative Goals, dative Experiencers display the opposite pattern.
These observations are in line with the findings by Belletti & Rizzi (
It is a well-known fact that negative QPs cannot be CLLDed (cf. [26b]). When fronted, they must leave a gap in argument position:
(26)
a.
Não
not
contes
tell.
a
to
ninguém
no one
o
the
que
what
me
me
disseste.
tell.
‘Don’t tell anyone what you have told me.’
b.
*A
to
ninguém
no one
lhe
to.him.
contes
tell.
o
the
que
what
me
me
disseste.
tell.
c.
A
to
ninguémi
no one
contes [-]i
tell.
o
the
que
what
me
me
disseste!
tell.
‘Don’t tell anyone what you have told me.’
Contrasts such as these fall naturally under an approach to CLLD that assumes that CLLDed topics are base-generated directly in place. The dislocated DP introduces an entity and the comment contains an ‘open’ position (a pronominal category) satisfied by the entity referred to by the dislocated DP. Since negative QPs are incapable of picking up a referent, they cannot be CLLDed. Thus, fronting of a nonreferring expression is only possible through movement, in which case it leaves a gap (cf. [26c]).
Interestingly, as originally noted by Belletti & Rizzi (
(27)
a.
É
it.is
que
that
agora
now
a
to
ninguém
no one
lhe
apetece
feels.like
escrever.
write.
‘The thing is that now no one feels like writing.’
b.
…
…
e
and
a
to
ninguém
no one
lhe
interessa
interests
despender
spend.
verba
money
em
in
actividades
activities
que
that
resultem
result.
em
in
desconforto
discomfort
‘… and no one is interested in spending money in activities that may result in discomfort.’
A search in Google for the sequence “
Belletti & Rizzi (
These observations support the thesis that preverbal dative Experiencers do not behave exactly like standard dislocated objects. However, as pointed out by Gutiérrez-Bravo (
The term “quirky subject” was originally applied to oblique arguments in Icelandic that behave like surface subjects in every relevant aspect. It was then extended to Italian and Spanish dative Experiencers (
The first property concerns person agreement. Icelandic quirky subjects block person agreement between the verb and the nominative argument (
(28)
a.
*Ég
I
veit
know.1sg
ađ
that
honum
him.
líkum
like.1
viđ
we.
b.
*Ég
I
veit
know.1sg
ađ
that
honum
him.
líkid
like.2
þiđ
you.
c.
Ég
I
veit
know.1sg
ađ
that
honum
him.
líka
like.3
þeir
they.
‘I know that he likes them’
EP behaves like Spanish (and German) and unlike Icelandic (we refer the reader to Gutierrez-Bravo (2006) for the Spanish data):
(29)
Ao
to.the
Pedro
Pedro
agrado
please.
eu/
I
agradas
please.
tu
you
/agradamos
please.
nós
us
‘I/you/we please Pedro.’
The second property concerns Conjunction Reduction, as in (30), where a nominative subject is coreferential with a deleted dative in the second conjunct.
(30)
Ég
I.
hafđi
had
mikiđ
much
ađ
to
gera
do.
og
and
___
var
be.
samt
still
ekki
not
hjálpađ
help.
‘I had much to do and was nonetheless not helped.’
Conjunction reduction with a dative Experiencer verb is impossible in EP (just like in Spanish, as originally pointed out by
(31)
*Ela
she
adora
love.
cães
dogs
e
and
___
agradam
please.
cavalos
horses
The third defining property of quirky subjecthood is that quirks can be controllees in infinitival clauses.
(32)
Hún
she
vonast
hope.
til
that
ađ
for
PRO-
leiđast
bore.
ekki
not
bókin
the.book.
‘She hopes not to find the book boring.’
As originally noted by Masullo (
(33)
*Ela
she
espera
hope.
agradar
please.
a
the
matemática.
mathematics
This paradigm shows that preverbal dative Experiencers do not have the behavior of quirky subjects in Icelandic.
Coming back to our experiment, the assumption that preverbal datives are subjects in EP would predict that the status of our test sentences with an intervening dative Experiencer should be no different from that of (2a), repeated here as (34), with a preverbal subject.
(34)
Já
already
li
read.
os
the
artigos
articles
que
that
aquela
that
revista
journal
não
not
quis
want.
publicar.
publish.
‘I have already read the articles that that journal didn’t want to publish.’
However, this prediction is not quite confirmed given that speakers’ ratings of the sentences with pre-verbal dative Experiencers are not exactly optimal, with ratings ranging from 1 to 7 and no detectable difference with respect to preverbal Goals in the middle points of the scale (cf. Figure
One other fact that constitutes an argument against a quirky subject analysis of these dative Experiencers is the observation that, within the Romance languages, only the null subject languages have been argued to have quirky dative subjects. As noted by Landau (
(35)
a.
Cette
this
musique
music
plaît
please.
à
to
Marie.
Marie
‘This music pleases Marie.’
b.
*À Marie plait cette musique.
This fact is surprising on a quirky subject analysis. If the preverbal dative Experiencers that are found in EP, Italian and Spanish were quirky subjects, it is not at all clear why they should be absent from French, a closely related language. Belletti & Rizzi (
(36)
Je
I
me
me
demande
ask.
quand
when
est
is
venue
come
Marie.
Marie
‘I wonder when Marie came.’
If, on the other hand, preverbal dative Experiencers are topics in all of these languages, the issue doesn’t arise. Since, in a null-subject language, Spec,TP doesn’t need to be overtly filled and the grammatical subject may be post-verbal (cf. [37]), a sentence with a topic Experiencer and a postverbal subject will always have the appearance of a sentence with a quirky subject (cf. [38]). This is not so in French, where the EPP forces overt subject movement to SPec-TP.
(37)
a.
Agrada-me
please.
isso.
that
‘That pleases me.’
b.
Apetece-me
crave.pres.3
um
an
gelado.
ice-cream
‘I am craving an ice-cream’
(38)
a.
Ao
to.the
Pedro
Pedro
agrada(-lhe)
please.
isso
that
‘That pleases Pedro.’
b.
Ao
to.the
Pedro
Pedro
apetece
crave.
(-lhe)
(
um
an
gelado.
ice-cream
‘Pedro is craving an ice-cream’
Landau (
Dative Experiencer arguments differ from dative Goals or direct objects in two ways: they are assigned inherent case and, in the base, they are projected higher than the nominative argument (the Theme). Thus, on second inspection, there is actually no reason to expect that dative Experiencer topics should pattern exactly like lower object topics. In fact, the rejection of a topic analysis of preverbal dative experiencers is grounded on the assumption that all left-dislocated topics are expected to pattern alike with regard to extraction. However, this assumption is factually wrong. Barbosa & De Cat (
(39)
J’habite
I live
la
the
rue
street
où
where
Nicolas
Nicolas
refuse
refuses
de
to
se
parquer
park
‘I live in the street Nicolas refuses to park his car in.’
(40)
Voici
here are
les
the
médailles
medals
que,
that
les
the
athlètes,
athletes
ils
they
sont
are
fiers
proud
d’avoir
to’have
remportées.
won
‘These are the medals the athletes are proud to have won.’
(41)
Voici
here are
les
the
athlètes
athletes
qui,
that
les
the
médailles
médals
d’or,
of’gold
les
them
ont
have
remportées.
won
‘These are the medals that the athletes are proud to have won.’
(42)
Elle
she
a
has
un
a
chien
dog
qui,
that
le
the
soir,
evening
se
transforme
transforms
en
into
chat.
cat
‘She has a dog that turns into a cat in the evenings.’
Stimuli were presented orally, and each test item was preceded by a short context. An ordinal regression analysis revealed a consistent picture, of which we highlight the following conclusions:
(43) | |
The strongest predictor of Intervenor effect is the position of the intervenor: Dislocated objects are significantly more disruptive of wh-chains than dislocated subjects, adjuncts or subjects. Subjects and adjuncts are less disruptive of wh-chains than dislocated subjects. |
These results show that embedding a dislocated subject within a wh-movement domain yields better results than embedding a dislocated object, regardless of the type of structure (i. e., relative clause or indirect question). Unsurprisingly, there is a partition between dislocated subjects, on the one hand, and adjuncts or subjects, on the other. However, the significant split between subjects and objects within the group of structures involving CLLD is something that, to our knowledge, has never been noticed before. This subject/object asymmetry indicates that height of the base position associated with the topic matters in determining whether it may or may not appear within the domain of wh-movement.
These data also argue against an analysis that attempts to explain this subject/object asymmetry by claiming that subject clitics are agreement markers, in which case the fronted DP is a subject rather than a CLLDed Topic (
This subject object asymmetry could be attributed to the particular status of subjects as natural topics. Since subjects are natural topics, subject CLLD is less marked that object CLLD.
A close look at the configurations involved in (40) versus (41) reveals that they differ from each other in the way the scopes of the wh-item and the topic interact. Schematically, (40) corresponds to the structure shown in (44a) and (41), to the structure shown in (44b).
(44)
a.
b.
While (44a) involves nested dependencies, (44b) involves intersecting dependencies.
With this much as background let us now reconsider the configurations involved in our EP study. Examples (23a,b) and (23c,d), reproduced here as (45a) and (45b), respectively, also differ from each other in the way the scopes of the wh-item and the topic interact. While (45a), with a dative Experiencers topic, displays nested dependencies (45b), (46a) exhibits intersecting dependencies (46b):
(45)
a.
Vi
saw.1
a
the
casa
house
que,
that
à
to.the
Maria
Mary
mais
more
(lhe
(
convém [-]
is.convenient
comprar [-]
buy.
‘I saw the house that is more convenient for Maria to buy.’
b.
(46)
a.
Já
already
li
read.
a
the
mensagem
message
que,
that,
ao
to.the
diretor,
director,
tanto
so
querem
want.
enviar
send.
ainda
still
hoje.
today
‘I already read the message that they want so much to send to the director today.’
b.
We thus observe that the EP data are consistent with the French data. In both cases, we find a similar pattern: the structures that involve nesting dependencies are rated significantly higher than those that display intersecting dependencies.
In fact, nested dependencies have been argued to be favoured over crossing dependencies quite generally (
(47) | [Which violin]k is [this sonata]i easy to play ti on tk ? |
(48) | *[Which sonata]i is [this violin]k easy to play ti on tk |
Baker (
(49) | a. | As a sentence is processed from left to right, a prospective tenant [=filler] |
b. | A prospective filler is assigned to the first unoccupied address [=gap] for which it is the most current of the eligible prospective tenants. |
Fodor (
(50) | Nested Dependency Constraint ( |
If there are two or more filler-gap dependencies in the same sentence, their scopes may not intersect if either disjoint or nested dependencies are compatible with the well-formedness conditions of the language. |
Finally, Pesetsky (
(51) | Path Containment Condition (PCC) ( |
If two paths overlap, one must contain the other. |
Pesetsky shows that the generalization in (51) has broad empirical scope.
The experimental items created for this experiment consisted again in relative clauses containing topicalization/left-dislocation constructions. This time, however, the function of the topic was held constant and we manipulated the height of the base position of the relative operator. In (52) and (53) the topic is a direct object. These sentences differ from each other with respect to the position occupied by the trace of the wh-operator: it may be a subject (52) or a dative Goal (53). As shown in (52b), (53b), the wh-subject condition displays crossing dependencies while the wh-dative condition exhibits nested dependencies.
(52)
Wh-subject
a.
Esse
that
é
is
o
the
editor
editor
[
que,
that
os
the
teus
your
artigos
articles
nunca [-]i
never
(osk)
(
quis
want.
publicar [-]
publish.
nessa
in.that
revista].
journal
‘That is the editor that never wanted to publish your articles in that journal.’
b.
(53)
Wh-dative
a.
Esse
that
é
is
o
the
revisor
referee
a
to
quemk,
whomk
os
the
teus
your
textosi,
textsi,
já
already
(os)
(
tentei
try.
enviar
send.
[-]i [-]k
várias
several
vezes.
times
‘That is the referee to whom, your texts, I have already tried to send several times.’
b.
a quemi os teus textosk, já (osk) tentei enviar [-]k [-]i
In the examples above, the clitic associated with the lower verb appears attached to the superordinate verbs
Sixty undergraduate and postgraduate students from University of Minho participated in Experiment 2 on a voluntary basis (average age 23,6, SD = 8,25; 49 females, 11 male). All participants were native speakers of European Portuguese, naive regarding the purpose of the experiment and reported not speaking any other language at home.
This experiment crossed the factors
(54)
Experimental conditions (Study 2)
a.
Wh-subject and clitic present (WhSub-Clit)
Esse
that
é
is
o
the
editor
editor
que,
that
os
the
teus
your
artigos,
articles
nunca
never
os
quis
want.
publicar
publish.
nessa
in.that
revista.
journal
‘That is the editor that never wanted to publish your articles in that journal.’
b.
Wh-subject and clitic present (WhSub-NoClit)
Esse
that
é
is
o
the
editor
editor
que,
that
os
the
teus
your
artigos,
articles
nunca
never
quis
want.
publicar
publish.
nessa
in.that
revista.
journa
‘That is the editor that never wanted to publish your articles in that journal.’
c.
Wh-dative and clitic present (WhDat-Clit)
Esse
that
é
is
o
the
revisor
referee
a
to
quem,
whom
os
the
teus
your
textos,
texts
já
already
tentei
try.
enviar.
send.inf
várias
several
vezes
times
‘That is the referee who I have already tried to send your texts to several times.’
d.
Wh-dative and clitic absent (WhDat-NoClit)
Esse
that
é
is
o
the
revisor
referee
a
to
quem,
whom
os
the
teus
your
textos,
texts
já
already
tentei
try.
enviar
send.
várias
several
vezes.
times
‘That is the referee who I have already tried to send your texts to several times.’
The same procedures described for the Study 1 were used in this experiment, except for the platform where it was implemented; this time, the Google Forms was used.
The same statistical approach described for Study 1 was used to analyze the results of this experiment. The data revealed a significant effect for the “Wh-height” factor (
Study 2 – Generalized linear mixed model fitted by using Markov chain Monte Carlo techniques (
Coefficient | Lower bound (CI 95%) | Upper bound (CI 95%) | pMCMC | |
---|---|---|---|---|
Wh-height_Dat | 0.633 | 0.084 | 1.100 | 0.006** |
Clitic_NoClit | –0.302 | –0.471 | –0.146 | <0.001*** |
Wh-height_Dat:Clitic_NoClit | 0.044 | –0.209 | –0.242 | 0.510 |
Response ~ Wh-heigt*Clitic, random = ~Participant + Item, data = Est2, verbose = FALSE, family = “ordinal”. Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1.
The pattern of responses represented by the means and the medians can be seen in Figure
Mean and Median acceptability ratings on a scale from 1 to 7 for Study 2.
The overall acceptability rates of the experimental stimuli were low. A higher percentage of responses at the lowest points of the scale (1–3: Wh-dative, 58.6%; Wh-subject, 70%) than in the highest points (5–7: Wh-dative, 26.7%; Wh-subject, 15%) was observed. At the midpoint (4) the response rate was 14.7% for the dative conditions and 15% for the Wh-subject conditions.
Nonetheless, similarly to the previous study, the distribution of the answers at each of the seven points of the scale reveals that at the extreme points, the pattern of responses for both types of sentences is reversed (Figure
Distribution of acceptability ratings on a 7-point Likert scale in the Study 1 (1 = ‘not acceptable’; ‘7 = totally acceptable’).
As in Study 1, we also used a linear mixed model with Gaussian family by using MCMC method (Table
Study 2 – Generalized linear mixed model fitted by using Markov chain Monte Carlo techniques (
Coefficient | Lower bound (CI 95%) | Upper bound (CI 95%) | pMCMC | |
---|---|---|---|---|
Wh-height_Dat | 0,657 | 0,096 | 1,201 | 0,020* |
Clitic_NoClit | –0,331 | –0,580 | –0,088 | 0,006** |
Wh-height_Dat:Clitic_NoClit | –0,016 | –0,354 | 0,322 | 0,936 |
Response ~ Wh-heigt*Clitic, random = ~Participant + Item, data = Est2, verbose = FALSE, family = “gaussian”. Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1.
Study 2 – Generalized linear mixed model fitted by using lmer function in the lme4 package (with Gaussian family).
Coefficient | Std.Error | |||
---|---|---|---|---|
Wh-height_Dat | 0,633 | 0,266 | 2,385 | 0,033* |
Clitic_NoClit | –0,356 | 0,129 | 2,762 | 0,006** |
Wh-height_Dat:Clitic_NoClit | –0,278 | 0,182 | –0,153 | 0,879 |
Response ~ Wh-heigt*Clitic, random = ~Participant + Item, data = Est2, verbose = FALSE, family = “gaussian”. Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1.
The results of this study indicate that there is a statistically significant preference for dative relative clauses over subject relative clauses containing an object topic. We consider these findings particularly striking in light of the well-known observation that, in head initial languages quite generally, subject relative clauses are preferred over object relative clauses on a number of different measures (
Compared to Study 1, the results of this study show a more marked concentration of the ratings at the lowest points of the scale, suggesting greater consistency among participants in their judgments than that seen in Study 1. Yet another result that is different from the previous study is that the clitic factor is statistically significant, with the presence of the clitic contributing to lower ratings in both conditions.
We start by noting that this preference for embedding a topic within the domain of a wh-dative chain over a wh-subject chain is consistent with the French judgments reported by Rizzi (
(55)
French Indirect questions
a.
Wh-dative
?Je
I
ne
not
sais pas
know.1
à
to
quii,
whom
ton
your
livre,
book
jek
I
pourrais
could. 1
lek
it
donner [-]k [-]i
give.
‘I don’t know to whom I could give your book.’
b.
Wh-subject
*?Je
I
ne sais
not know.1
pas
NEG
qui
who
ton
your
livrek, [-]i
book
pourrait
could.3
l’k
it
acheter [-]k
buy.
‘I don’t know who, your book, t could buy it.’
(56)
French relative clauses
a.
Wh-dative
?Un
a
homme
man
à
to
qui
whom
ton
.your
livrek,
book
je
I
pourrais
could.1
le
it
donner [-]k [-]i
give
‘A man to whom I could give your book’
b.
Wh-subject
*/?Un
a
homm
man
quii,
who
ton
your
livrek,
book
[-]
pourrait
could.3
l’
it
acheter [-]k
buy.
‘A man who could buy your book.’
Recall that, for Rizzi (
(57) | Je ne sais pas [qui C [ton livre Top … [ |
According to Rizzi, even if C is turned into a governor via agreement, it is too far away to license the subject trace due to the intervening Top head, a standard case of relativized minimality. Therefore, the structure is ruled out as an ECP violation. (55a), by contrast, is a mere subjacency violation.
This ECP-based account predicts that structures comparable to (55b) and (56b) should be fine in a null-subject language such as EP. As is well-known, many null subject languages lack
(58)
Quem
Who
achas
think.2
que
that
vai
go.3
faltar
miss.
à
to.the
aula?
class
‘Who do you think will miss class?’
The grammaticality of examples such as (58) has been attributed to the fact that the null subject languages may avail themselves of a strategy that is unavailable in English or French, namely extraction from post-verbal position, which is head governed by V raised to T (
Additional evidence against an ECP-based account of the contrasts in (55), (56) comes from the observation of French examples with wh-adjuncts and embedded CLLD, such as the following, which are taken from Barbosa & De Cat (
(59)
a.
Je
I
me
demande
wonder
quand,
when
ton
you
patron,
boss
il
he
va
will
nous
us
inviter
invite
‘I wonder when your boss will invite you.’
b.
???Tu
you
sais
know
quand,
when
le
the
voleur,
robber
on
one
l’
him
a
have
surpris [-] ?
surprised
‘Do you know when the thief got caught?’
In the acceptability judgement elicitation task carried out by Barbosa & De Cat (
On the other hand, a close look at (59a,b) shows that these examples differ from each other in the way the two dependencies interact. If we assume that the trace of the temporal adjunct occupies a position that is lower than the subject, but higher than the object (
(60)
a.
b.
A similar account carries over to the contrasts in (55) and (56). (55a, 56a) display nesting dependencies while (55b, 56b) exhibit crossing dependencies. Thus, in all of the multiple dependency constructions examined the illicit examples involve crossing dependencies while the others involve nesting dependencies. We therefore conclude that a restriction against crossing dependencies is operative in these constructions, with one important qualification. The restriction against crossing only applies to configurations in which the topic appears to the right of the wh-phrase, i.e., not in cases like (61), in which the topic precedes the wh-constituent (this holds independently of the presence of the clitic).
(61)
Ao
to.the
Pedroi
Pedro
[o que]k
the what
(lhei)
(
interessa [-]i
interests [-]i
mais
more
comprar [-]k ?
to.buy.
‘What is it that is in Peter’s best interest to buy?’
(61) involves crossing dependencies and yet it is acceptable. Moreover, the constraint also doesn’t apply to dependencies established between different topics. In general, there are no ordering restrictions on multiple topics, regardless of whether they are CLLDed (62) or simply topicalized (63).
(62)
CLLD
a.
Esse
that
livroi,
book
ao
to.the
Joãok,
João
não
not
lho
posso
can.1
dar [-]i [-]
give.
‘That book, to John, I can’t give it to him.’
b.
Ao
to.the
Joãok
João
esse
that
livroi
book
não
not
lho
posso
can.1
dar [-]i [-]
give.
‘To John, that book, I can’t give it to him.’
(63)
Topicalization
a.
Esse
that
livroi,
book
ao
to.the
Joãok,
João
não
not
posso
can.1
dar [-]i [-]
give.
‘That book, to John, I can’t give.’
b.
Ao
to.the
Joãok,
João
esse
that
livroi,
book
não
not.
posso
can.1
dar [-]i [-]
give.
‘To John, that book, I can’t give.’
This suggests that the constraint in question should be restricted to apply to chains created by wh-movement. The following descriptive generalization adequately captures the patterns observed:
(64) | A wh filler-gap dependency may tolerate a topic in its scope iff the full topic-(cl)-gap dependency is contained within the domain of the wh-filler gap dependency. |
As argued in Barbosa & De Cat (
However, if topics are base-generated in place, one cannot appeal to a constraint on movement, such as Pesetsky’s Path Containment Condition or any of its current instantiations, such as the Minimal Link Condition, to rule out the cases that do not fall under (64). Conversely, if we do assume that CLLD involves movement and that (64) is derived from a ban on intersecting movement operations, then (61), (62a) and (63a) are predicted to be ruled out, contrary to fact. In other words, we have no explanation for why the ban on intersecting dependencies applies whenever the wh-constituent precedes the Topic and not when the reverse order obtains. For these reasons, we will explore an account of (64) that doesn’t rely on principles of narrow syntax.
A purely syntactic account of the relevant contrasts faces yet one additional challenge. While full DPs are disruptive of wh-chains in crossing configurations, pronouns are not. In other words, CLLDed pronouns appear to be exempt from (64). In the study by Barbosa & De Cat (
(65)
Voilà
here
le
the
bateau
boat
qui,
that,
moi,
me,
m’a
toujours
always
fait
made
rêver.
dream
‘Here is the boat that has always made me dream.’
Structurally, (65) is similar to (56b). In our study on EP, we didn’t include pronouns, but a cursory search in Google retrieves a number of examples of subject relative clauses containing a CLLDed pronoun (66a–d):
(66)
a.
Castigada
punished
será
be.
a
the
Turkey
pela
for.the
decision
do
of.the
líder
leader
que
that
a
to
ele
him
lhe
sucede
succeeds
‘Turkey will be punished for the decision of the leader who will succeed him.’
b.
…
,mantendo-se,
keeping-
no entanto,
however
as
the
dívidas
debts
that
to
him
sobrevivem.
survive
‘…, keeping, however, the debts that have survived him.’
c.
As
the
palavras
words
foram
were
uma
a
surpresa
surprise
para
to
mim,
me,
são
are
palavras
words
bonitas
nice
que
that
a
to
mim
me
não
not
me
fazem
make
diferente
different
‘The words were a surprise for me, they are nice words that don’t make me different (…)’
These examples have precisely the same structure of (54a), which was rejected by our subjects.
One might speculate that the peculiar behavior of pronouns is due to Clitic Doubling. In effect, in French as well as EP, direct and indirect object strong pronouns are necessarily clitic doubled. However, it is not possible to deny that the pronoun is left-dislocated in these examples, given that it appears in the left-periphery. Configurationally, there are no differences between the structures containing CLLDed pronouns and structures containing non-pronominal CLLDed constituents.
Another possibility would be to attempt to derive the exceptional behavior of pronouns from the fact that they posssess the feature Person, which is not present in full fledged DPs. This approach, however, would only work under a minimality account of the deviance of (54a, b). A minimality account, however, would assign the same status to (54a, b) and (54c, d), contrary to fact. This is why we believe that (64) is not amenable to explanation on the basis of syntax alone.
On the other hand, there is some indication that (64) could be fruitfully investigated as deriving from processing constraints. In fact, two aspects of our data that are problematic for a purely syntactic account — the fact that the restriction against intersecting paths applies to wh-movement chains and not to topic-chains; the fact that CLLDed pronouns are exempt — are better understood when (64) is considered from a processing perspective.
Intuitive judgements of acceptability are known to be affected by processing difficulties (
(67) | a. | The reporter [who the senator attacked] disliked the editor |
( |
||
b. | # The reporter [who the senator [who John met] attacked] disliked the editor. | |
( |
The contrast above shows that processing overload may affect intuitive judgments of acceptability. Processing overload has been attributed to limitations of the computational resources of the language processor. It is now a well-established fact that dependencies of the kind investigated in this paper consume a certain amount of processing resources (
In addition, several electrophysiological studies have provided evidence of the memory cost of keeping an open dependency. In a study of Event-Related Potentials (ERPs) registered during the processing of subject and object questions, Kluender & Kutas (
Building on these results, Felser, Clahsen & Münte (
These findings are relevant for our purposes, because they constitute evidence that there is a difference in integration cost between topic fillers and wh-fillers. If (64) is related to processing constraints and if the integration costs of topic dependencies are lower than those incurred by displaced wh-phrases, it is no longer surprising that (64) should apply to wh-dependencies and not to topic-(cl)-gap dependencies (note that topics introduce a referent in discourse; wh-operators introduce operator variable dependencies).
The second fact that we considered problematic for a syntactic account of (64) concerns the difference between pronominal and phrasal intervenors. This effect, however, is unsurprising from the point of view of processing. Indeed, a similar asymmetry between pronouns and full DPs has been observed in a variety of phenomena in the literature on language processing. In the case of center-embedded sentences, for example, their intelligibility is increased if the subject of the most embedded relative clause is a pronoun, as in (68a) rather than a full DP, as in (68b) (cf.
(68) | a. | # The reporter [who the senator [who John met] attacked] disliked the editor. |
( |
||
b. | The reporter [who the senator [who I met] attacked] disliked the editor. | |
( |
Warren & Gibson (
For these reasons, we believe that deriving (64) from the processing integration cost of wh-fillers is a promising line of approach. Viewed from a processing perspective, generalization (64) amounts to the claim that maintaining an incomplete (topic) dependency while processing a wh filler-gap dependency is costly for the human processor. By contrast, if the topic filler-gap dependency is complete by the time the wh-gap is encountered, performing the integration of the wh-gap is easier.
One result of the present study that would receive a natural explanation on these grounds is the variable effect of the presence of the clitic in the different studies. In the case of Study 1, our results indicate that the factor “clitic” was not statistically significant. However, a close look at Figure
In a series of experiments on the online processing of CLLD constructions in Spanish, Pablos (
We conclude by observing that we do not claim to have reached an explanatory account of the patterns obtained. We established an empirical generalization and we formulated a hypothesis thus laying out the ground for further research on the online processing of multiple dependency constructions in the left-periphery.
In his paper, we have focused on intervention effects created by embedding a complement topic within the domain of wh-movement. We have presented the results of two experimental studies in Portuguese designed to determine the effect of two independent factors: height of the base position associated with the topic relative to the base position of the wh-gap and presence
(69) | A wh-movement dependency may tolerate a topicalized or CLLDed constituent in its scope iff the full topic-(cl)-gap dependency is contained within the scope of the wh-filler gap dependency. |
This generalization suggests that a version of the no crossing constraint (
As seems clear, this hypothesis needs to be tested against online measures, so the next step is to study the processing of these constructions in real time.
The additional file for this article can be found as follows:
Stimuli used in Studies 1 and 2. DOI:
In effect, once minimality is incorporated under the definition of
Duarte (
For Duarte (
Here, by ‘Topic Function’ we mean the theta-role of the DP in preverbal position.
Retrieved from
Retrieved from
Such an analysis has also been shown to be untenable for a number of reasons (De Cat 2005). In particular, non-referential quantified phrases are incompatible with a coindexed subject clitic:
(i)
a.
*Quelqu’un il vient.
someone he comes
b.
Quelqu’un vient.
someone comes
‘Someone is coming.’
In the French
The only exception to this pattern are multiple wh-questions in Bulgarian, where there is a preference for crossing over nesting (
Raposo (
Retrieved from
Retrieved from
The first, second and third authors were partially supported by the Centre for the Humanities of the University of Minho, financed by national funds through FCT-Fundação para a Ciência e a Tecnologia, I.P., under the project UIDB/00305/2020.. The second author received support from the “Fundação para a Ciência e a Tecnologia” (FCT-CEECIND/04331/2017) for part of this work. The fourth and fifth authors were partially supported by the Research Centre of Mathematics of the University of Minho with the Portuguese Funds from the “Fundação para a Ciência e a Tecnologia” (Project UID/MAT/00013/2013).
The authors have no competing interests to declare.