[B-Greek] Re: Predicate as topic

Elizabeth Kline kline-dekoning at earthlink.net
Sun Jun 26 17:30:46 EDT 2005


On 6/26/05 12:41 PM, "R Yochanan Bitan Buth" <ybitan at mscc.huji.ac.il> wrote:

> Back to pragmatic marking on verbs:
> In general, there is an ambiguity in a verb-initial syntactical system (like
> both Greek and Hebrew, despite there greatly differing frequencies of
> functions) and Œfronted¹ verbs can be overlooked.
> With biblical Hebrew, there are some examples with fronted participles both as
> Œtopical¹ and as Œfocal¹(see Buth, ³Verbless Clause...² in  Miller, ed., The
> verbless clause in BH, Eisenbraun¹s 1999. I would call the volitional
> stuructures in Hebrew where volitional verbs are joined to vav (Œand¹) without
> using the seuqential system as another instance of pragmatic marking on a
> verb, though more than likely ŒFocal¹ rather than Œtopical¹.

Randall,

Here is a rather long quote from Helma Dik. Would you or Ivar Larsen could
interact directly with Helma Dik's statements :

4.2 Topic assignment and the predicate


In standard Functional Grammar, Topic assignment has always been restricted
to terms.95 Undeniably, terms, especially those of the concrete and animate
type, are the most common elements to be chosen as the starting point for an
utterance. In section 4.1, we have seen some instances of this. In this
section I would like to show that the theoretical restriction in FG of Topic
assignment to terms may be a reflection of prototypical Topic assignment,
but not a universal borne out by empirical data.


When first embarking on the study of Greek word order with FG as the
theoretical paradigm, I quite soon arrived at the conclusion that much of
the evidence can be described very nicely in terms of Topic and Focus
assignment to elements in the initial part of the sentence. The snag,
however, was that the initial part of the sentence was defmed as that
preceding the predicate.96 This meant that there was a group of sentences in
the corpus that my theory could not even begin to deal with, as there was
nothing preceding the predicate in those sentences.


Given a clausepattern Topic-Focus-Predicate-Remaining elements, and the
possibility of Focus, but not Topic, assignment to predicates, there appear
to be two possible information structures for predicate-initial clauses. The
first one being that there is no Topic, and the predicate has Focus function
(as in (5) and (6) in 4.1); and alternatively, that there is no pragmatic
function assignment at all.97 In other words, these clauses are about
nothing and do not make a point. Neither description is satisfactory for a
considerably large number of instances.


At this point, there are two options in describing the data. One consists of
taking for granted that Topic function cannot be assigned to the predi-
cate, and, therefore, of looking for pragmatically marked constituents in
the 'Postfield' (the elements following the predicate). This approach leads
to a stipulation of two different ordering patterns. In one pattern, marked
elements are placed in the initial part of the sentence, preceding the
predicate, and in the other, marked elements follow the predicate. In such a
description, the factors determining the choice between the two patterns
would have to be established as well. It is not clear to me how this could
be done without reference to the special pragmatic status of the predicate
in the predicate-initial clauses.


The second option involves setting aside theoretical prejudice and having
another look at the clause-initial predicate and its textual function. Can
it be compared in a fruitful way to other clause-initial constituents, thus
providing a unified description with, admittedly, some cases more
prototypical than others? It goes without saying that my answer to this
question is affirmative. The second option will be the one I adopt in this
study, and in this section I will discuss instances of tpatsioj.at for which
it can be argued that the predicate has Topic function. However, in order to
avoid any misunderstanding, I should point out here that I use 'predicate'
in the broad pre-theoretical sense, which in most cases is equivalent to the
finite-verb constituent'. In the course of the discussion, the question will
be raised as to exactly what element or combination of elements within a
constituent is assigned Topic function when that constituent as a whole is
assigned Topic function.98 More specifically, I will argue that in cases
where I hold that the predicate has Topic function it is not the implicit
subject of the finite verb that has been assigned Topic function, as some
have claimed. Such a position is invalidated by the fact that the subject
constituent can be expressed separately as a Noun Phrase or can even be the
element with Focus function when the predicate has Topic function in that
clause.


A final point before turning to the actual instances may be of importance.
Although I will cite quite a few instances of predicates-as-Topics, I should
point out that their actual frequency is not very high compared to the
'normal cases of concrete and/or animate entities with Topic function; they
can be associated with contexts in which the predicate presents"

---Quoted from pages 64-65

Helma Dik, Word Order in Ancient Greek: A Pragmatic Account of Word Order
Variation in Herodotus. Amsterdam Studies in Classical Philology 5.
Amsterdam:  J. C. Gieben, 1995.  Pp. xii + 294. f. 143.10.  ISBN
90-5063-457-5. 


Elizabeth Kline 





More information about the B-Greek mailing list