Biblical Hebrew Forum

[Kimmo Huovila <kimmo.huovila AT helsinki.fi>]

Dear Rolf,

Wow! This thread has so many interesting comments so fast. Next time I
read my e-mail there were tons of comments to your theory. I understand
if you run out of time responding to each claim made so far, but I
comment on a few things anyway. (Also, a brief response to Dave at the
end.)

At the time it took me to write this, some of the points have already
been addressed, but I send my version anyway rather that go through the
whole text again deleting part of the text.

Rolf Furuli wrote:

> What you say above is exactly my point. it does not make sense to speak of
> a durative verb being punctual. And my intention was therefore to argue for
> the existence of fundamental linguistic units that can be said to represent
> "semantic meaning", i.e. meaning that under no circumstances can be
> canceled. Randall does not seem to accept the existence of such units. So
> when we accept the elementary and fundamental principle, the question is
> whether such verbs (with uncancelable meaning) do exist in reality, in
> Hebrew. Grice's principle is that "semantic meanings may not be canceled
> without contradiction or reinforced without redundancy." This means that
> the features of the verbal system which in no context and under no
> circumstances can be changed or canceled, represent semantic meaning, while
> features that can be changed or canceled represent conversational pragmatic
> implicature, even though they may seem to represent a uniform meaning.

I am afraid that you might be using a definitional point trying to prove
something about (linguistic) reality. It does not work.

I think that the above dichotomy between semantics and pragmatics is
overly simplistic (though rather common). I am afraid that it just helps
cloud some issues. Grice has a nice theory of pragmatic implicature, and
as far as I know, it has not been seriously challenged. However, to
limit linguistic meaning to uncancelable semantic meaning and pragmatic
implicature (such as Grice's theory handles) is to rob language a whole
lot of meaning. The sentence
that many pragmatics text books use as an example ('Could you pass me
the salt?') can be explained as a request by Grice's maxims. But the
maxims cannot explain why 'Are you capable of passing me the salt?' does
not produce the same effect.

The phenomenon here is conventionalization. One sentence used an idiom
that has become a conventional way of expressing a request. The other
does not. Now, my question is what component of our language theory will
deal with matters like this. I cannot see that any theory of pragmatics
could do it without making explicit statements about conventional
expressions (Grician pragmatics surely is inadequate). And if you
exclude this kind of data from semantics, what are you left with? If you
say that pragmatics deals with these phenomena, then you need a theory
of pragmatics that can adequately do it (and it comes out resembling
semantics theories in many points more than pragmatics theories).

I know that many have tried to draw a line between semantics and
pragmatics as to what is cancelable and what is not. Then we must be
ready to give pragmatics a whole lot of idiosyncracies and meaning that
cannot necessarily be handled by pragmatics theories.

Even with the cancelability thing, it is not always clear as to what is
cancelable and what is not (even for natives). I view
semantics-pragmatics as a continuum rather than a strict dichotomy. Any
strict distinction between semantics and pragmatics is somewhat
arbitrary. Conventionalization of an expression is not a binary thing.
It is continuum like. At the one end there is totally conventional,
semantic meaning. At the other there is pragmatic meaning that can be
adequately handled by Grice's conventional maxims.

Why go to such lengths at defining the difference between semantics and
pragmatics. I think that it is crucial if one tries to build a theory of
a linguistic form that a priori assumes that its meaning is or is not
cancelable. I think that there are both cancelable and uncancelable
elements of meaning in language. The meaning of syntactic elements is no
exception.

It is definitely a worthwhile attempt to characterize the uncancelable
properties of an expression. However, one may end up with a highly
abstract property that cannot explain all the data, though  is in no
contradiction with any of it (this is how I took Randall's basic
argument). For example, what is the uncancellable property of the
English genitive (that would characterize all possible uses of the
genitive)? Can someone come up with anything better than an asymmetrical
relation? Surely an adequate description of the English genitive must go
beyond that (would you call it going into pragmatics?).

> The difference between stative and fientive verbs in Hebrew is very
> important. But if we apply the cancelability test to stative verbs with the
> question: «Is stativity a semantic property?», we find that the answer is
> no. As in the case with the verb used in (5) and (6), verbs which generally
> are stative may in some contexts be interpreted as fientive. This does not
> mean that we should discard "stativity" as a meaningful concept, or that
> the importance of it is reduced. But it means that stativity is the most
> common interpretation of particular verbs, not the only interpretation.

The basic lexical aspect of many verbs can vary. In other words, many
lexemes are aspectually polysemous. This does not necessarily exclude
these aspects from being part of semantics any more than lexical
polysemy (cf. Wittgensteins example of 'game'). (I view stativity etc.
(Vendlerian categories) as aspectual. There is no clear, semantic
distinction with what has been called Aktionsart. Both phenomena are
best handled by the same theory.)

> In relation to Aktionsart (or the Vendlerian categories) there are three
> fundamental semantic units in Hebrew and English: dynamicity, durativity,
> and telicity, but punctuality (such as "hickup", "find" etc) are not
> semantic because any "punctual" verb can also have a durative
> interpretation.

I do not think this is quite true. Many punctual verbs can (I view this
as aspectual polysemy), but not all. As a matter of fact, there are
linguistic studies whose objective is to categorize punctual verbs into
those that allow for a durative interpretation and those that do not.

> The reverse is not true; verbs with durative Aktionsart can
> not have a punctual interpretation. To show that some semantic units are
> uncancelable I challenged Randall to find an event where "run" or "sing"
> could have a punctual interpretation.

But this is because of the way we define durativity. It does not
describe linguistic reality per se. I do not think that this proves
anything about the uncancelability of semantic units. You asked Randall
something that is definitionally impossible. On the matter of
uncancelable semantic units, I think they exist, but some semantic units
are cancelable. Our general theory must be equipped to deal with both.
We cannot force our data to fit a theory if it does not. If our theory
excludes either kind of semantic units, then we are in danger of this
very thing. (Here I do not mean that we need to call those aspects of
meaning as 'semantic'. Call them pragmatic, if you will. But they have
to be included in the analysis.)

(Some authors, e.g. Fanning, Verbal Aspect in NT Greek, p. 150, do give
positive tests of punctuality: "I bought this book at five o'clock", but
"*I read this book at five o'clock." Yet this hardly disproves my point.
The issue with this test of Fanning's is not aspect per se, but our
knowledge of the world. "I drank a glass of water at five o'clock."
would be OK, just because it takes less time, not because the aspect is
different.)

The situation would be opposite if we were to define durativity as a
semantic feature that specifies that the activity is predicated to have
taken time, and punctuality as the converse, then any durative verb
could also be punctual (eg. 'I sang' - it is not predicated as taking up
time any more that the usual example of punctuality 'the cable snapped';
we are just relying on our knowledge of the world, and perhaps partly on
the fact that 'snapped' does not allow for a durative interpretation
(thus being a counter-example to your claim that any punctual verb can
be durative).).
(What a complex sentence I just wrote. Perhaps I should study how to
write readable texts :-)
If we were to define our terms the way I described above, linguistic
reality would remain unchanged. This is why I feel uncomfortable with
the conclusions I think you are making from this point.

> Regarding tense, Broman Olsen's scheme of semantic meaning is as follows:
> Past tense: Reference time comes before the deictic point.
> Present tense: RT coincides with C.
> Future tense: RT comes after C.
> This relationship is uncancellable, and it can be applied to Hebrew. If we
> find one form with a particular morphology, to be a tense, its occurrences
> should have a uniform pattern as to the relationship between RT and C. No
> such form exists, and therefore Hebrew does not have grammaticalized tenses.

I think here we are in danger of making the same mistake Porter made (in
my opinion) with his analysis of Greek. Yes, we find that there is no
time relation with any Greek 'tense' that could not be canceled (contra
Olsen, as argued by Decker). But that does not justify the idea that
Greek has no tense. Take English, for example. There are excpetions to
the past meaning of the imperfect (If I was to come...). So English does
not have tense either (you could find exceptions to any English tense).
Or Finnish, even more temporal language (though without an obligatory
future tense). The Finnish imperfect does code past tense almost always
(including conditionals). But there is a future or present referring,
rare use, based on presupposition. If I assume that you know something
already, I could, in some contexts, refer to it with the imperfect even
if the situation is in the future (the logic being that this WAS how the
situation would stand - as you know it from the past information, even
though the situation is in the future). If
the assumption is not shared, the expression can result in
misunderstandings. So the Finnish imperfect is no past tense either? But
in almost any context (with some exceptions, like the one I described
above) the native intuition is that the form codes exactly past tense,
and does not allow for other interpretations.

However, I think that we should strive for understanding the core
meaning of Hebrew verbs. Also relevant is the semantic and syntactic
constraints on canceling (assuming that the canceling is not just by
implication and that it can be done with any sentence; perhaps this is
how you see it) like those on the Finnish future referring imperfects.
If it can be argued that apart from certain, well-defined situations,
Hebrew verbs do all behave according to a temporal theory, then we have,
in my opinion, proved that Hebrew has tense (to a degree). Of course, it
is possible that we come up with
another theory with less exceptions that is not based on tense per se,
but that has temporal implicaitions in the semantics of the language.
This sort of phenomenon would be like tense-mood in some languages. The
verbal
system can be analysed from a temporal or modal viewpoint, and results
can be very close. In such cases one may argue that the language codes
mood and another that it codes tense. Does it have to be an either-or
necessarily (though some analysis may capture more data). Prototype
theory of linguistic categorization does not force us to make it an
either-or situation (but allows us to do it if data directs us in that
direction).

The interesting question is if there are contexts where the Hebrew verb
allows only for a certain temporal reference (grammatically speaking,
not discourse coherence). If there are, then at least in those contexts
the form grammaticalizes tense (though this does NOT necessarily mean
that tense is the best analysis for the Hebrew verb).

I am not arguing that the Hebrew verbal system is purely temporal. Nor
am I arguing it to be purely aspectual. Or purely modal. Just too many
exceptions. What is it then? Perhaps a combination of these. I do not
know (but I want to).

Dave Washburn wrote:

> Could you give a constructed
> > example of a sentence that would demonstrate a punctual use of 'run' or
> > 'sing'?
> 
> "He ran to the door."  "She sang one staccato note."

Punctuality does not mean short duration. Our knowledge of real world
events does not necessarily correspond to linguistic punctuality. If
'the cable snapped' is punctual, we still could analyse the time that it
took for the cable to emit the sound. However, if we try something like
'the cable snapped for two milliseconds', we have made an iterative
expression or we mean that the results of the snapping remained for two
milliseconds. Punctuality means that the event cannot be predicated over
time. Durativity is its converse. "She sang one staccato note." is no
more punctual than "she sang five arias." Both can take an adverbial of
minimum duration ("she sang five arias for twenty minutes."; "She sang
one staccato note for X time" (I do not know how long it would take)).
(The
aspectual behavior of such sentences is not adequately handled by just
analyzing punctuality here. Something happens also to perfectivity when
we add the adverbial of minimum duration. Here the concept of nesting
helps.)

Well, Dave, if you want to define punctuality so that your senteces are
punctual, feel free to do so. However, I assume that Rolf would not
consider these sentences (in English or Hebrew) to be counter-examples
to his theory. The definition I gave above is roughly the one I have
seen most used. I do not claim it to be sacrosanct.

P.H. Matthews defines punctual as "marking an action, etc. taking place
at an undivided moment of time" (Concise Dictionary of Linguistics).

Comrie defines punctual as follows:
"The opposite of durativity is punctuality, which thus means the quality
of a situation that does not last in time (is not conceived of as
lasting in time), one that takes place momentarily. It should be noted
that the crucial point here is that punctual situations do not have any
duration, not even duration of a very short period. Thus a punctual
situation, by definition, has no internal structure, and in a language
with separate imperfective forms to indicate reference to the internal
structure of a situation, then clearly punctuality and imperfectivity
will be incompatible." Aspect, p. 42.

Frawley's definition: "If an event is momentary and has no temporal
duration, it is punctual; if it is necessarily distributed over time, it
is durative." Linguistic Semantics, p. 306.

> 
> Next question.

Would you explain the whole universe and give two good examples? :-)

Kimmo