http://www.jhsonline.org/Articles/article_155.pdf
Here's an excerpt from the author's opening words:
The goal of this paper is to provide a new definition of a Biblical
Hebrew (BH) verbal construction, usually referred to as wayyiqtol
. . .
There are two types of commonly employed functional and semantic
definitions of the formation. First, when referring to taxis3-
aspect-tense-mood (TATM) properties, the gram has most frequently
been equaled with a past (definite past or preterite) or perfective
past (other proposals, on the contrary, identify the gram
with a present tense and imperfective aspect^ for a general review
of descriptions posited by temporal, aspectual, historicalcomparative
and psychological schools, as well as those offered by
the first generation of grammaticalization framework, consult footnote
4 below). Other theories (especially, those developed by the
syntactical approach) have adjoined a value of sequentiality to the
TATM load of the construction. Second, when emphasizing its
discourse pragmatic characteristics—and disregarding TATM values—the
expression has been classified as a principal form (foreground)
of the narrative backbone. However, the two descriptions
are reductionist and greatly simplify the nature and substance of the
wayyiqtol.
The former ignores or minimizes the fact that (as will be indicated
in section 2) the formation provides not one but a broad
range of uses and hence cannot be reduced to a single value such as
past or perfective past. Nor is it appropriate to understand such
frequently proposed labels (i.e. past or perfective past) as überfunctions
from which other meanings are derivable, i.e. hardly can
the use of the formation with a future or stative present force be
explained as a realization of its past or past perfective value (cf.
section on 3.1 and the discussion on a dynamic vision of synchronic
grammatical phenomena). As for the discourse-pragmatic classification
(principal narrative construction), it is reductionistic in the
sense that it ignores the evident semantic content of the gram as
well as the fact that the formation expresses determined temporal
and aspectual meanings, entailing a failure to denote others. This
means that to account for the entire nature and behavior of the
wayyiqtol is neither easy nor straightforward and that, in particular, it
cannot be swept under simplifying reductionistic definitions, such
as a past, a perfective past or a narrative form. There furthermore
exists a third group of descriptions which, although highly valuable,
are limited to a mere taxonomy (cf. e.g., Waltke & O’Connor 1990
in footnote 3). They introduce a detailed—not reductionistic—
account of the semantic content of the construction without, however,
providing an explanation for it: they fail to account for the
relation between uses of the gram and its internal consistency.4