LINGUIST List 13.379

Mon Feb 11 2002

Sum: Focus and Event Relations in Clauses

Editor for this issue: Marie Klopfenstein <>


  1. sharbani, Re:13.261 Focus

Message 1: Re:13.261 Focus

Date: Sun, 10 Feb 2002 22:28:22 +0530
From: sharbani <>
Subject: Re:13.261 Focus

Sometime back I had asked the following questions relating to Focus:

1) Can anybody explain to me WHY and HOW Focus is related to the EVENT
of the clause?
2) I would also like to know which other language(s) clearly show such a
 I received the following responses:
AI)From: Werner Abraham <abrahamsocrates.Berkeley.EDU>

He sent me the following paper:

'The German Clause structure under discourse functional weight: Focus 
and Antifocus'----Werner Abraham & L=E1szl=F3 Molnarfi 
 He also sent me the Call for contributions in:

Volume of Folia Linguistica in 2003 guest edited by Werner Abraham & 
L=E1szl=F3 Molnarfi. Topic: Optionality in Syntax--Aspects of Word Order 
Variation in (West) Germanic

Werner Abraham, Visiting Professor 2000/2002
Dept. of German, 5333 Dwinelle, UCB
Berkeley, CA 94720-3243
Tel. +1-510-642-2004, fax +1-510-642-3243
as of mid-May 2002:
Maitschern 128, A-8942 Woerschach, tel=3Dfax: +43-3682-23175

B) Daniel Wedgwood <>
sent the following VERY informative reply:

 The clearest proposal that I know of regarding events and focus is
Elena Herburger's attempt to capture focus meanings using
neo-Davidsonian semantic representations, as expounded in her book
`What Counts' (2000, MIT Press). She proposes that, in a `tripartite'
schema of existential quantification over an event variable, the focus
is in the `nuclear scope' of the quantifier, while (only)
thematic/non-focus material is mapped in to the restrictor. So `Fred
[ate the BEANS]' (VP focus) is something like (1a), while `FRED ate
the beans' looks more like (1b) [and here the initial `E' should be
the existential quantifier]:

1a) Ee [AGENT(e,fred')] EAT(e) & PAST(e) & THEME(e,the-beans') &

`there is an event with Fred as its Agent such that Fred ate the beans'

b) Ee [EAT(e) & PAST(e) & THEME(e,the-beans')] AGENT(e,fred') & EAT(e) &
PAST(e) & THEME(e,the-beans')

`there is an eating event in the past with the beans as its Theme such
that Fred was the Agent of that event'

[this is a very rough version of her representations!]

In other words, Herburger proposes that the restrictor of the event
quantification is what the sentence is `about', leaving the nuclear
scope as focus. The paraphrases perhaps give an intuitive insight into
why focus and event semantics might be linked.

A review of this book appeared on the Linguist List last year and this
might give you a better description than I can manage of the ideas in
The review can be read at:

Herburger also has an article called `Focus and Weak Noun Phrases' in
the journal `Natural Language Semantics' (1997), which contains a
brief description of this general approach .

Herburger's ideas have a certain intuitive appeal, but, interestingly,
exactly the opposite proposal has also been made before: i.e. that
focused material is mapped into the restrictor of event
quantification, while thematic material related to the nuclear
scope. This idea was put forward by Larson & Lefebvre in the following

Larson, Richard & Claire Lefebvre. 1991. "Predicate Clefting in Haitian
Creole". In T. Sherer et al.(rds) Proceedings of NELS 1990:247-263.

In some ways, both of these approaches could be seen as
re-interpretations, in Davidsonian terms, of the `structured meanings'
approach to focus, in which focused items are lambda-abstracted out of
ordinary predicate-argument structures.

As for other languages showing evidence of event-focus relationships,
I don't know too many specific, detailed examples, but I know from
reading around the Hungarian literature that it seems to be common for
identifiable `focus positions' to be syntactically adjacent to the

As far as I remember just now, this is reckoned to be true of at least
Hungarian, Basque, Turkish, Aghem and Armenian (although I'm not
claiming to be absolutely sure about any apart from Hungarian!). In
the languages mentioned here (as well as in Korean), focus also seems
to occupy a position syntactically similar to that occupied by items
which can form `complex predicates' with the verb - and this process
could presumably be represented in terms of event semantics also. In
Hungarian, at least, a number of aspectual effects also involve the
presence or absence of certain elements in an immediately pre-verbal
position, the same position that focus occupies (at the
surface). Presumably, these could also be analysed by making reference
to event variables. So there seems to be scope for involving events in
some explanation that would, at some level, unite focus and these
other phenomena.

For more description and more references relating to languages with
focus positions, and how these might relate to the verb, you might
find the following the collection of papers useful (in case you don't
know it already) - especially the contribution by Mi-Jeung Jo:

`Discourse Configurational Languages', ed. Katalin E. Kiss (1995)
(Oxford University Press)

One more (possibly irrelevant) thing, since you are dealing with
Object Shift: I once read an article by David Adger (in the Canadian
Journal of Linguistics, 1997) which deals with shifted object pronouns
in Irish and Scots Gaelic. These pronouns - unusually - shift to the
right. This phenomenon had previously been dealt with in purely
syntactic terms, but Adger suggests that the correct way to deal with
it is to define the position of these pronouns phonologically: they
seem to encliticise to the constituent which bears the main stress in
the sentence. So this could be seen as a connection between focus and
(a kind of) object shift. Maybe that's relevant to you; maybe not.

Daniel Wedgwood
postal address:
Department of Theoretical and Applied Linguistics
University of Edinburgh
Adam Ferguson Building (Dept.) tel. 0131 650 3961
George Square (Dept.) fax. 0131 650 3962
Edinburgh EH8 9LL

C)With Dieter Wunderlich <>

The following question-answer session ensued:

Ans): My understanding is that focus can be related to everything in a
clause, to one of the core arguments, to one of the adverbials, to the
whole VP, but also to the verb itself (which means, to the event
encoded by the verb) and even to the assertation status of the clause
(if there is some relevant part of the utterance that can be
highlighted by intonation).

Q) Must every clause have a focussed phrase? Ans) No. The clause can
be out of the blue. If focus is grammaticalized (by movement to a
specific position, by using particles, or by cleft sentence) and not
just attributed by intonation, I expect that some sentences are
blocked from this device.

Q)The Bangla data shows that when the verb is a 'N-V' type, with the
DO incorporated, there cannot be any Focussed phrase in the clause.

Ans)Could one consider the incorporated noun to be in a focus? If not,
then there is something in the construcion that blocks focusing.

Q) In case of Dative subject constructions, there cannot be a Focus on
any phrase in the clause.

Ans) This is interesting. Again, either such a subject itself is focus 
or it blocks the focus construction, maybe because the subject undergoes 
some shift in that construction.

Q) When the aspect of the verb is completive, eg., 'to give away',
there cannot be a Focussed phrase in the clause etc.,

Ans)How is aspect marked? By some particle or preverb?

Q)In Bangla, the verb MUST move to the head of the Focussed phrase.

Ans) Does this mean that the verb combines with the focussed phrase?
Maybe some verbs cannot move in this way: because they have
incorporated a noun or have to license a dative subject, or have some
other material with them.

Q) My intuition is that it is the EVENT which DICTATES the presence or
absence of Focus in a clause.

Ans) Clarify what this intuition is about. I doubt that the event
itself dictates focus. Focus has to do with something that contrasts
with the preceding context or with a set of alternatives possible for
one of the elements in a clause. This is independent of the semantic
properties of eventualities. My suggestion is that all the blocking
phenomena you have described have to do with some morphosyntactic
complexity of the focus construction. If the verb has to be moved,
then some material strongly accociated with the verb may block
movement. However, since I have no idea of how the focused and
unfocused sentences actually look like, I cannot make any more
specific proposals.

Q)It is standardly assumed that in Wh-Qs the WhP has the Foc feature.
But in Bangla, that is not obligatory. There can be cases where the
WhP doesnot have the Foc feature, but some other phrase has the Focus

Ans) In German this is possible too. One can ask: Und wen hat PETER
besucht? ('and who has been visited by PETER?')

Prof. Dr. Dieter Wunderlich
Seminar fuer Allgemeine Sprachwissenschaft
Heinrich-Heine-Universitaet Duesseldorf
D-40225 Duesseldorf
Fax: +211-81-11325

D)With Alex Monaghan
 too, I had a similar question-answer session:
Ans) Focus is related to whatever the speaker/writer thinks is 
important. so, if he/she thinks the event is important, then focus may
well attach to the event in a high proportion of your data. also, if 
focusing mechanisms (topicalisation, clefting, etc.) include ways of 
making the focus coincide with the event, then there will be a 
correlation between the two.

however, there are plenty of examples where focus does not correspond
to the event in a clause, so this is not a necessary condition. in
certain genres (e.g. story telling), the event will be frequently
focused, and certain languages may tend to focus the event (just as
french tends to have focus at the end of the clause, and english tends
to mark *narrow* focus by moving constituents away from the end).

Q) What I am not able to understand is, Why there simply CANNOT be a
Focus feature in a clause under certain conditions---and that is what
has led me to conclude that it is the EVENT/ASPECT which DECIDES
whether there should be a Focus present in the clause at all.

 I have found quite a few cases in my work, with corresponding
consequences, for example, the WCO effect gets affected as a
result.Some of the cases are:

Dative subject constructions
N-V predicates(incorporated)
completive aspects(returned)--when the Focus can only be on the 
verb(contrasting with other verbs--not the aspect as such) etc
 Things would become clearer if one can identify cases where there 
CANNOT be any Focus feature in the clause.
Ans) Now i'd really need to know your definition of focus:
 - prosodic prominence?
 - new information?
 - syntactic/morphological marker?
 - other?
 I can suggest possible explanations of no-focus clauses and give some 
examples from English which I hope will be helpful.

The reasons one might expect focus NOT to be marked are as follows:
 - The focus is completely predictable (only one possible location), so 
marking is redundant
 - There is no important new information in the clause
 - All the information in the clause is equally new/important

 An example of the first type in english would be the much-discussed
word "even", as in "Even a linguist should be able to understand
calculus" or "Even a child can do that". in these cases, the semantic
constituent which immediately follows "even" is necessarily the focus.

 The latter two types are treated much the same in the literature on
prosodic focus in Germanic languages (Bolinger, Ladd, Schmerling,
Gussenhoven, Hirschberg, and my own work). they are both cases of
"broad" focus, i.e. the whole clause is equally focused, and the
neutral default rules apply (major prominence on the last prominent
item, etc.). thus, a conversation might go like this:

A: What's the matter?
B: John is an alien. (all new, broad focus)
A: So, John is an alien. (all old. broad focus)
B: and Sam is an alien. (Sam is new, the rest is old)

crucially, B's first utterance and A's response would have exactly the
same focus structure (and very similar prosody), whereas B's second
utterance would have very different prosody in english.

Following is Alex Monaghan's publication page:
 Dr Alex Monaghan
 Aculab plc, MK1 1PT, UK

 I am extremely grateful to all those who responded. Though I have
received very elaborate responses, more responses are still welcome.
This summary perhaps has clarified the questions more.
 I'll post the summary ofcourse, if I receive more replies.

Thanks a lot

Sharbani Banerji
(C/o Centre for Applied Linguistics & Translation Studies,
University Of Hyderabad
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue