[b-greek] Re: Prototype theory

From: Kimmo Huovila (kimmo.huovila@helsinki.fi)
Date: Tue Mar 06 2001 - 11:13:38 EST

"Rodney J. Decker" wrote:
> That's about as clear as mud--at least for non-linguists.

Probably true. Sorry. I just tried to save time.

> And most of us
> are not subscribed to b-hebrew.

You can visit the archives without subscribing. Try
and choose 'visit b-hebrew without joining', and then 'read messages',
then 'jump to date 1999/12/13' and scroll to almost bottom, pick Peter's
message titled 'Prototype Theory and Hebrew Tense/Aspect' and click
'View message'. Or better still, read the first chapter of my thesis
(Peter's message does not really go deeper into prototype theory than I
did in my thesis).

> Would you mind explaining it to those of us
> who are just ordinary exegetes? I would sincerely like to understand the
> theoretical problems that a few see with my proposal, but since I don't
> have a PhD in linguistics, I'm not even able to propose an intelligible
> defense.

I understand your concern. Reading the first chapter of my thesis gives
a foundation. It is available at
For further reading I heartily recommend John Taylor's Linguistic
categorization: Prototypes in Linguistic Theory and on tense-aspect
specifically, ึsten Dahl's Tense and Aspect systems.

Grammatical meaning, as well as lexical meaning, is (usually) learned by
being exposed to multiple examples of valid use. People are able to
abstract something (a prototype) that is in common with the examples and
use the expressions correctly, and yet often they are unable to explain
why they use grammar the way they do. With reference to tense and
aspect, people learn typical situations when aorist indicative for
example is used (past perfective). Yet they see that not all examples
fit the pattern. Another pattern emerges (perfective timeless or
future). This could be described as another, secondary prototype, which
is not unrelated to the first. Yet, because the overwhelming majority of
aorist indicatives are past, there is some pastness in the meaning of
the aorist, which may be cancelable. In a neutral context (if there is
no indication otherwise) it is considered past. Yet perfectivity is more
central, because the temporal meaning can be canceled, but not

With the imperfect the issue is even more clear. There are quite few
prototypical uses: past imperfective, and imperfective 'unreality'
(which has two subtypes: counter-factuality in conditions and polite use
with verbs of desire, a weakened form of 'unreality'). Perhaps some
others, I just took these off the top of my head. It basically means
that whenever there is an imperfective, the meaning must fit into one of
these categories (unless it is a novel extension - but this leads us
much deeper into prototype theory, namely category extension - a very
fascinating and rewarding topic to study, BTW). The fact that there is
another focus of the prototype category does not negate the fact that in
the majority of cases the imperfect really grammaticalizes past
imperfective (both temporal and aspectual meaning), and the reference
could not be any other time. Or conversely, if the reference is not past
time, the use must follow another focus of the category (it must refer
to counter-factuality or be a polite use of a verb of desire - or
whatever categories I may have missed).

This is quite parallel with English. 'I ran.' must refer to past time.
'If I ran...' does not. In the first example the imperfect
grammaticalizes past time, in the second 'unreality'. It would not make
sense, from the prototype theory point of view, to say that the second
example ('If I ran') proves that 'I ran' does not grammaticalize past

I am not sure how much this helps. Probably still muddy. To make this
really clear, I should lay the whole foundation for prototype theory,
but yet it is more feasible for me just to give the references. If this
is not any clearer after reading the first chapter of my thesis, do not
hesitate to ask for clarification. I am glad to help more thoroughly.

Kimmo Huovila

B-Greek home page: http://metalab.unc.edu/bgreek
You are currently subscribed to b-greek as: [jwrobie@mindspring.com]
To unsubscribe, forward this message to leave-b-greek-327Q@franklin.oit.unc.edu
To subscribe, send a message to subscribe-b-greek@franklin.oit.unc.edu

This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:36:52 EDT