LINGUIST List 29.3455

Sat Sep 08 2018

Sum: Indexical Shift Data

Editor for this issue: Everett Green <>

***************** LINGUIST List Support *****************

Fund Drive 2018
28 years of LINGUIST List!
Please support the LL editors and operation with a donation at:

Date: 04-Sep-2018
From: Kadri Kuram <>
Subject: Indexical Shift Data
E-mail this message to a friend

Dear fellow linguists,

I need your help for my research on the effect of L1 in the L2 acquisition of the indexical shift parameter. For this, I need the shifting parameter in various languages. As far as I know, there isn’t a definitive list of shifting languages, apart from the ones famous in the literature – that is Turkish, Uyghur, Kazaki etc. That is why I need your help. Since testing hasn’t started yet and I will have little time afterwards, I will welcome the parameter from any language you may provide. There are three values of this parameter: Non-shifting, obligatory shifting, optional shifting.

Although various items can shift, the easiest one to find and the one I am interested in right now is first and second persons. Here is how to find out if your languages shifts:

First and second persons in the finite embedded clauses may refer to the speaker and the addressee of the discourse respectively (non-shift) or to the subject and the addressee of the verbs of reporting in the main clause (shift). This only works with verbs of reporting and verbs of presumption such as think and assume. Here is an example.

(1) [speaker]i [addressee]j Jack said to Jane that Ii want to take youj out for dinner.

Here, I and you can only refer to the speaker and the addressee. Jack and Jane are not going out for dinner. It is simpler with a single indexical.

(2) [speaker]i Jack thinks Ii am smart

Jack thinks it is the speaker who is smart, not himself. English is a non-shifting language (there are arguments that indexicals sometimes shift in English, but I am putting them aside for the time being).

(3) [speaker]i Ahmetj [ei/j çok akıllıy-ım] sanıyor
Ahmet very smart-1SG thinks
‘Ahmet thinks he is very smart’
‘Ahmet thinks I am very smart’

The null subject in (3) may refer to the speaker of the discourse or to the subject of the main clause. Turkish is an optional-shifting language (still I am not 100% sure it is optional, I am still testing the native speakers). Finally, if in your language the embedded first person subject can only refer to the main subject, it is an obligatory shifting language.

A few warnings: 1. Make sure the embedded clause is fully finite. Finiteness is not a dichotomy, there are degrees to it. It should be exactly like a main clause with tense and agreement. Nominalization, which is a common strategy in languages like Turkish, does not work – supposedly because the shifting operator resides the CP of the embedded clause. 2. Try this with verbs of reporting and verbs of presumption such as say, tell, think, assume, believe etc. although there may be some interpretational difference between the two. It will be bonus if you can report such differences. 3. Shifting is not an easy reading to get even for native speakers. If you are reporting a language you are not a native speaker of, please make sure to consult a native for confirmation.

I will accept data through my email ( or in any way linguistlist allows.


Linguistic Field(s): Syntax

Page Updated: 08-Sep-2018