Coulter
GEORGE The Spatial Use of kata
and ana
with the Accusative in Homer
The Greek prepositions have long been
neglected by scholars of Greek linguistics. While the
standard handbooks all give useful compilations of
examples, they do little to explore the
interrelationships between them. Silvia Luraghi's recent
monograph, On the Meaning of Prepositions and
Cases, goes a long way towards addressing this odd lack
of literature in schematizing the relationships between
the different semantic roles that prepositions can
express. But while she is right to take many of her
examples from the poems of Homer, in which the underlying
spatial meanings of the prepositions are most prominent,
she pays too little attention to the constraints imposed
on Homeric language by the rules of oral composition and
the dactylic hexameter. In particular, this paper will
focus on the spatial use of kata
and ana
with the accusative in Homer and will argue that meter
cannot be ignored in determining the factors that explain
which preposition is used in a given context.
The spatial significance of kata with the accusative is undeniably
complicated. Two of its uses are relevant here. First, it
can be used to mark an action that takes place at many
points within the area marked by the preposition; in such
passages, the preposition can usually be translated "in,
throughout", e.g. polloi gar kata astu mega Priamou
epikouroi (Il. 2.803). Second, kata is found in
expressions where a soldier is struck in a particular
part of the body, e.g. kata stēthos balein.
In reference to these uses, Luraghi contrasts kata with
ana: in
expressions of the first type, kata,
it is argued, denotes "throughout" in a less exhaustive
way than ana, while in
those of the second type, kata
is viewed as indicating a less precise location than
ana does.
Unfortunately, neither of these explanations seems to
work. First, kata can denote as
exhaustive a motion as ana, as can be seen in Od.
1.144&endash;5: hoi men epeita
/ hexeiēs hezonto kata klismous te thronous
te. The adverb
hexeiēs suggests a
more regular procession than does ephoita in Il. 5.528, which Luraghi cites to
illustrate the "exhaustive" use of ana: Atreïdēs d' an' homilon
ephoita polla keleuōn.
Second, the locations specified by kata are no less precise than those
marked by ana,
as can be seen in Il. 11.108: ton men huper mazoio kata stēthos
bale douri.
Far from being a vague description of where the spear
struck, the narrower specification huper mazoio, together with kata stēthos, provides a
rather precise portrayal of the wound. Such additional
details about the location of the wound are also found in
other similar passages; ana, on the other
hand, is not found in any comparable wounding scenes in
the Iliad. Indeed, its
most common objects denote places that are larger than a
body part: dōma,
straton, astu.
We might expect the difference between kata and ana to be brought
into sharper relief in the few passages where they occur
side by side, for example ban d' ienai kath' homilon
ana straton eurun Achaiōn (Il. 4.209) and
tis d' houtos kata nēas ana straton ercheai oios
(Il. 10.82). But
there is little that homilon and nēas have in common, as against
straton, that would
explain why the former occur with kata and the latter with ana. Instead,
where the two prepositions differ is in the metrical
contexts in which they are used: for instance, kata
straton, found twenty-one times in Homer, occurs without
fail just before the bucolic diaeresis, while ana
straton, which occurs
only ten times, is found in three different places in the
line. Given, then, that in such passages there is no
clear semantic distinction between kata and
ana and, further,
that there are metrical restrictions on which preposition
is used in which part of the line, we are left with the
conclusion that, while certainly not always
interchangeable, the two prepositions could be used as
equivalent to one another in order to serve the metrical
needs of the poet. We may compare up the road
and down the road
in English: in the context of a road that has a clear
change in elevation, only one of the two would be
admissible. But in describing motion along a level road,
either may be used.