LINGUIST List 9.32

Sun Jan 11 1998

Sum: Jakobson Quote

Editor for this issue: Martin Jacobsen <>


  1. Hartmut Haberland, Jakobson Quote

Message 1: Jakobson Quote

Date: Fri, 09 Jan 1998 12:09:40 +0100
From: Hartmut Haberland <>
Subject: Jakobson Quote

In Linguist 9.20, I had a query, "Roman
Jakobson is supposed to have said something like 'Languages differ not
so much in what they can say but in what they must say' (in the sense
that some languages force you to make some choices - number in nouns,
aspect in verbs - which others are indifferent to). I have quoted it
myself, but cannot find the reference at the moment. Any help out

Only a few hours later, answers came in, from
Alex Alsina
Remi Jolivet
Stuart Payton Robinson
Dietmar Zaefferer.

Dietmar Zaefferer pointed out to me that he had asked a similar
question a few years ago, the answers to which can be found in the
LINGUIST archives (LINGUIST List 6.411, Wed 22 Mar 1995).

The essence of the answer is that the quote is from Roman Jakobson
(1959) 'On linguistic aspects of translation', in Reuben A. Brower
(ed.), On translation, Cambridge, Mass.: Harvard University, Press.
232-239. Reprint New York: Galaxy Books 1966. (The quoted sentence is
on p. 236.) Also in Jakobson, Roman (1971) Selected Writings vol. II,
The Hague: Mouton. 260-266. (The quoted sentence is here on p. 264.)

And the quote is "Languages differ essentially in what they _must_
convey and not in what they _may_ convey."

(There is a lot more in the posting in LINGUIST 6.411
, including a related quote.)

In his 1995 discussion of the issue, Dietmar Zaefferer referred to
Huang's temperature metaphor in the description of anaphora. My
interest in the quote goes into a similar direction and originated in
my interest in the status of pragmatics in Linguistic Typology. If
pragmatics is universal, as one (a horrible simplification, of course)
could infer from reading Habermas and Grice, there is not much room
for pragmatics in linguistic typology. Still, serious students of
linguistic typology deal with pragmatic phenomena all the time, as
e.g. is clear from the results of the EUROTYP project of the European
Sciernce Foundation which are beginning to be published now: to
mention one of the relevant parameters, the coding of the
thetic/categorical distinction is something some languages 'must'
convey while others don't have to. When Kuno suggested that the
phenomena analyzed by Kuroda as expressing the thetic/categorical
distinction could be analyzed be reference to information structure
(given/new etc.), this can not only be seen as two different
explanatory approaches to the same phenomena, but also as a reflection
of different preferences built into the grammar of different
languages: some languages 'must convey' (have to code) what I would
call 'discourse-pragmatic' distinctions (Kuno), others 'must convey'
(have to code) what I would call 'situation-pragmatic' distinctions.

Hartmut Haberland
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue