LINGUIST List 8.749

Sun May 18 1997

Sum: Right Dislocation

Editor for this issue: Ann Dizdar <>


  1. Lawrence Y. L. Cheung, Sum: Right Dislocation

Message 1: Sum: Right Dislocation

Date: Mon, 19 May 1997 00:24:43 +0800
From: Lawrence Y. L. Cheung <>
Subject: Sum: Right Dislocation

- ----
Dear Linguist Listers,

Three weeks ago, I posted a query about the properties of Right
Dislocation in different languages. Thanks a lot to the following
respondents for their valuable contribution:

Joaquim Brand=E3o de Carvalho
Javier Perez Guerra
Rob Pensalfini
Asya Pereltsvaig
Patrick Sauzet
Gregory Ward
Colin Whiteley

The following is a summary of the responses to my three questions:

- -------------------------------

a. Do any of the languages (esp. in colloquial form) you know of
exhibit RD?

In addition to 6 languages (Cantonese, Mandarin Chinese, Tamil,
English, French, Catalan) mentioned in my query, contributors report
that RD is also found in Jingulu, Portuguese, Spanish, Russian, and
Occitan, making up a total of 11 languages.

- -------------------------------

b. What can be dislocated? Cantonese and Mandarin allow dislocation of
a wide variety of categories, e.g. Subj./Accusative Obj. NP, modals,
most adverbs, and some others. However, French, Catalan, and Tamil are
reported to be rather limited in this aspect, allowing NP and
PP. Please correct me if I am wrong.

The answer to the question is not clear. Many only indicate in their
examples that NPs, as expected, are possible in RD. I will summarize
the properties of RD in the languages.

1. _Jingulu_:

Rob Pensalfini:

Jingulu is fully nonconfigurational. Verbal words show agreement with
both subject and object (third person agreement is null), and nominals
generally have case marking (Ergative).

Words or phrases that occur at the edges of clauses (left or right)
can occur without the expected case marking. This is usually (though
not always) accompanied by an intonation break:

Darrangku-warndi warlaku maya-nu, wawa.
stick-INST dog hit-did child
'The child hit the dog with a stick.' OR
'He hit the dog with a stick, the child.'

The dislocated element can be co-referent with another nominal in the

Jama-rni maya-nu warlaku, wawa.
that(masc)-ERG hit-did dog child
'That child hit the dog.' OR
'That one hit the dog, the child.'

In Jingulu the default case is Absolutive (unmarked).

2. _English_:

Rob Pensalfini:

The disappearance of expected case marking on a clause-peripheral
nominal (left or right periphery) is diagnostic of dislocation. This
is also true in English. If you have a dislocated pronoun, that
pronoun shows up in the default case (which in English is Accusative),
irrespective of the grammatical role it is co-referent with:

That man hit me, him.
*That man hit me, he.

3. _Portuguese_:

Joaquim Brand=E3o de Carvalho:
Portuguese does not permit right dislocation. French-like structures
as the one you mentioned (il attend a la porte, le garcon) are
precisely a very typical feature of many French-born Portuguese people
when speaking Portuguese.

Javier Perez Guerra:
I have discussed your informant's observation with two colleagues
(Portuguese), and, in principle, RD seems to work in Portuguese
exactly as it does in English or Spanish.

4. _Spanish_:

Javier Perez Guerra:
Spanish, which is quite 'particular' as far as LD is concerned, is
absolutely 'conservative' as regards RD. Everything commented on in
the literature on English RD applies to Spanish as well.

5. _Russian_:

Asya Pereltsvaig:
There is RD in Russian. In many respects it is similar to RD in
English. A characteristic difference is that the RD NP in Russian
clearly exibits Case form (the same as the pronoun within the clause).

Compare the following:
I don't like them, the cops.
Ja ne ljublju ix, polizejskix.
I not like them:acc cops:acc
Ani lo ohev otam, et ha-shotrim.
I not like them:acc the-cops:acc ("et" is Acc Case marker for
definites only)

As you can see in both Russian and Hebrew the RD NP shows the case
that corresponds to there syntactic position within the clause. In
English, however, one cannot determine the case of the RD NP. "The
cops" can be either accusative or nominative. The only case exhibited
by full NPs in English is Genitive. It seems to me that RD NP
corresponding to Gen. NP will have Gen form too.

I've seen his brother, John's.
* I've seen his brother, John. (unless "John" is a vocative)

Another example:
Ona napisala emu, Aleksu.
she wrote he:dat Alex:dat

As opposed to:
Aleks, one emu napisala.
Alex:nom she he:dat wrote.=20

It is impossible for the RD NP to correspond to an empty pronoun.

Since Russian is a relatively free word order language, it is not an
easy question to determine whether some element (particularly
adverbials) is base-generated at the rightmost psoition, moved
rightwards (f.e. by RD) or the rest of the sentence moved leftward
(F.e. leaving modal behind) as in the following example:

Prochest' etu knigu on prochital...
to-read this book he read:past

Prochest' on etu knigu prochital...
to-read he this book read:past

It is possible to analyse this sentence as involving RD moving modal
(Tense)or as involving VP preposing moving the infinitival form. Of
course the verb has to appear twice since Tense cannot appear without
being attached to a verb. To my mind, the second analysis seems more

6. _Occitan_:

Patrick Sauzet:
As a contribution to your right dislocation survey, I would like to
mention that (not too surprisingly) it also appears in Occitan, a
Romance Language spoken in Southern France and closely related to

The right dislocated element does represent old information, as you
mention. In that respect it is similar to the basic use of left
dislocation in that language (but there is another type of left
dislocation where the dislocated element is a focus).

The right dislocated element is normally uttered with a flat and
rather low intonative pattern.

An interesting difference between right and left dislocation is that,
in case the dislocated element corresponds to a PP in the matrix
clause, you can left dislocate the DP embedded in the PP alone (and
you normally do so), but you must right dislocate the whole PP (see
for instance the paradigm in 13-16 below).

In the examples I give you below, the right dislocated element is
separated by a comma.

1 . A donat de pan al can, Peire.
 Has given DET bread to-the dog, Peter
"Peter has given bread to the dog."
(Possible answer to: "What has Peter done?")

[Occitan is a pro-drop language. The absence of clitic in case the
dislocated element is correferential with the subject is the expected

2 Peire n'a donat al can, de pan.
Peter CLIT:obj-indef. has given to-the dog, DET bread
"Peter has given some (scil. "bread") to the dog"
(Possible answer to "Why is there so little bread left?")

3 Peire li a donat de pan, al can.
Peter CLIT:dative has given DET bread, to-the dog.
"Peter has given some (scil. "bread") to the dog"
(Possible answer to "Has the dog been fed?")

(With [+human] NPs, it is possible to right dislocate the DP instead
of the PP:
 3' Li as tornat son libre, (a) Peire?
 CLIT:dative have(2sg) returned his book, (to) Peter
 "Have you given his book back to Peter."

4 Lo cot=E8l, l'i metrem, dins lo tirador.
The knife, CLIT:object CLIT:locative we-shall-put, in the drawer.
"We shall put the knife in the drawer."

5 L'i metrem, dins lo tirador, lo cot=E8l.
CLIT:object CLIT:locative we-shall-put, in the drawer, the knife.
"We shall put the knife in the drawer."

6 L'i metrem, lo cot=E8l, dins lo tirador.
CLIT:object CLIT:locative we-shall-put, the knife, in the drawer.
"We shall put the knife in the drawer."

7 Los auc=E8ls cantan, al/lo mes de mai.
The birds sing, at-the/the month of May.
"Birds sing in the month of May.=20

(A bare right dislocated DP is possible there because "lo mes de mai"
can function as a PP in any position.)

8 Los auc=E8ls cantan, quand fa b=E8l.
The birds sing, when makes nice.
"Bird sing when the weather is nice."

9 *Los auc=E8ls cantan, sovent.
The birds sing, often.

10 *Peire tornar=E0 a l'ostal, plan planet / d'aise / leu-leu.
Peter will-return at the-house, slowly / quietly / hurrily.

11 Peire beur=E0 un c=F2p de vin, puei.
Peter will-drink a strike of wine, then.
"Then Peter will have a glass of wine."

12 L'as pagat, lo capel?
 CLIT:object have(2sg) payed, the hat.
"Did you pay for that hat?"

(A ritual question asked to someone wearing a weird hat. I exemplifies
the very common occurence of right dislocation in questions. Here "the
hat" is a topic inferred from the situation.)

13 Lo cinema, i vau sovent.
the cinema, CLIT:loc. I-go often.
"I often go to the cinema."

14 I vau sovent, al cinema.
CLIT:loc. I-go often, to-the cinema,
"I often go to the cinema."

15 Al cinema, i vau sovent.
at-the cinema, CLIT:loc. I-go often.
"I often go to the cinema."

16 *I vau sovent, lo cinema.
CLIT:loc. I-go often, the cinema,
"I often go to the cinema."

- -------------------------------

c. Do you know any references to this kind of phenomenon in any

Here is the list of references suggested by the contributors.

Sauzet, Patrick. 1989. "Topicalisation et prolepse en occitan", in
A.Rouveret et P.Sauzet =E9ds. La structure de la proposition dans les
langues romanes, Revue des langues romanes, 93-2, 235-273. [RD is
touched upon briefly.]

Perez-Guerra, Javier. (forthcoming) "Integrating right-dislocated
constituents: a study on cleaving and extraposition in the recent
history of the English language" Ed. J. Fisiak _English Historical
Linguistics_ Berlin: de Gruyter.

Ward, Gregory and Betty J. Birner. 1996. "On the Discourse Function of
Rightward Movement in English,'' in _Conceptual Structure, Discourse
and Language_, edited by Adele Goldberg. Stanford: Center for the
Study of Language and Information Publications. p. 463-479.

Ziv, Yael. 1994. Left Dislocation and Right Dislocation: Discourse
Functions and Anaphora, _Journal of Pragmatics_, 22:5, 1-17

- -------------------------------

Lawrence Cheung
Department of English
Chinese University of Hong Kong
Shatin, N.T.
Hong Kong
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue