Chantraine P. Grammaire homérique, t. 2: Syntaxe 1953 Paris
Chantraine P. Dictionnaire étymologique de la langue grecque. Histoire des mots, t. 1 1968 Paris
Ebeling H. Lexicon Homericum 1885 Leipzig Hildesheim
George C. " The Spatial Use of ἀνά and κατά with the Accusative in Homer " Glotta 82 2006 70 95
Haug D. Bubenik V. , Hewson J. , Rose S. " Does Homeric Greek Have Prepositions? Or Local Adverbs? (And What’s the Difference Anyway?) " Grammatical Change in Indo-European Languages 2009 Amsterdam/Philadelphia 103 122 http://dx.doi.org/10.1075/cilt.305.12hau
Horrocks G.C. Space and Time in Homer. Prepositional and Adverbial Particles in the Greek Epic 1981 New York
Langacker R.W. Foundations of Cognitive Grammar vol. 1: Theoretical Prerequisites 1987 Stanford
Luraghi S. Rijksbaron A. , et al " The Opposition Total/Partitive and the Use of Cases with Prepositions in Ancient Greek " In the Footsteps of Raphael Kühner 1988 Amsterdam 177 192
Luraghi S. " Cause and Instrument Expressions in Classical Greek " Mnemosyne 1989 Vol 43 294 308
Luraghi S. Studi su casi e preposizioni nel greco antico 1996 Milano
Luraghi S. Soares de Silva A. , Torres A. , Gonçalves M. " The Container Schema in Homeric Greek " Linguagem, cultura e cognição: estudios de linguística cognitiva 2004 Braga 25 41
Luraghi S. " The History of the Greek Preposition µετά " Glotta 2005 Vol 81 130 159
Luraghi S. Barðdal J. , Celliah S. " The Evolution of Local Cases and Their Grammatical Equivalent in Greek and Latin " The Role of Semantics and Pragmatics in the Development of Case 2009 Amsterdam/Philadelphia 283 305
Palmer L. The Greek Language 1962 London
Schwyzer E. Griechische Grammatik, Bd. 2: Syntax 1950 München
Taylor J.R. Zelinsky-Wibbelt C. " Prepositions: Patterns of Polysemisation and Strategies of Disambiguation " The Semantics of Prepositions 1991 Berlin/New York 151 175
FN1 1)I would like to thank Mnemosyne’s anonymous referee for helpful comments on an earlier version of this paper.
FN2 2)On the trajector/landmark alignment see Langacker 1987, 231-6.
FN3 3)See Taylor 1991. Note that, although the name ‘trajector’ might be taken to imply movement, this is not the case, and the trajector/landmark alignment also applies to stationary situations (this can also be the case with διά, though less frequently, see section 4.1.3).
FN4 4)Uniplex entities may occur with prepositions that require biplex trajectors. Biplex entities can also be indicated by singular count nouns when the two-sided nature of the entity is emphasized, as is the case with landmarks occurring with ἀµφί, cf. Luraghi 2003, 256.
FN5 5)Indeed, the fact that example (6) sounds more normal without the addition of the second clause reflects profiling properties of into: one is likely to use intowhen one needs to specify that the trajector does not remain outside the landmark. Since this is the common inference also when one uses to, when describing the normal situation in which someone goes to the station to take a train one usually does not add unnecessary information regarding the end segment of the trajectory.
FN6 6)See the recent discussion in Haug 2009.
FN7 7)As is well known, spatial meaning of Greek cases is connected with case syncretism; the three prepositions ἐν, εἰς and ἐκ were already much more common than plain cases in Homeric Greek, and can be seen as substitutes for cases in the encoding of the three basic spatial relations locative, allative and ablative, see Luraghi 2009.
FN8 8)See Luraghi 1988 and 2003. The fact that the plain accusative could not only indicate direction and function as an allative in Ancient Greek is described in reference grammars, which mention the so-called ‘accusative of extension’ indicating extension in space or duration in time, see Schwyzer 1950, 67-70, Chantraine 1953, 45-6.
FN9 9)Μετά with the accusative could also occur with motion verbs, in which cases it had a directional meaning, see Luraghi 2005.
FN10 10)The description in this section is partly based on Luraghi 2003, 168-75.
FN11 11)On the etymology of διά see Chantraine 1968 s.v. This particle is somewhat exceptional with respect to most other Greek prepositions/preverbs, because it never occurs in Homer as a free standing adverb, see Chantraine 1953, 95.
FN12 12)Further semantic roles expressed by διά with the genitive are time, instrument and intermediary; see Schwyzer 1950, 450-4, and Luraghi 2003, 176-87.
FN13 13)On the meaning of κατά in Homer, see Chantraine 1953, 112-5, and Luraghi 2003, 197-204.
FN14 14)Contrary to διά with the genitive, διά with the accusative is not limited to space expressions in Homeric Greek, but can also indicate cause. Cause expressions are left out of consideration in the current discussion.
FN15 15)The same type of motion is described in Il.11.147, in which the headless whole body of a dead hero is described as rolling through the fighting soldiers with the verb κυλίνδεσθαι.
FN16 16)The PP διὰ προµάχων occurs thirteen times starting in the second foot, and twice starting in the third; the similarity of the verses where it occurs points toward the highly formulaic character of the PP.
FN17 17)The anonymous referee points out that “Leto’s (intended) trajectory does indeed terminate outside the landmark Panopeus”. However, while I agree on the referee’s remark that “[t]he fact that she does not cross Panopeus completely is not expressed by διά + gen. but rather by the present stem”, it still remains that the landmark is not completely crossed: so διά with the genitive is not limited to occurrences in which it indicates complete crossing of a landmark.
FN18 18)Even though it can be remarked that Peiraeus was leading Odysseus to the place purposefully.
FN19 19)The fact that the trajectory does not remain inside the landmark in such occurrences of διά with the accusative obviously creates problems for Ebeling’s interpretation of the meaning of landmarks in the accusative with this preposition, which he sees as necessarily connected with trajectories enclosed in the interior of landmarks conceived as closed areas. Thus he attempts various explanations, partly from ancient scholiasts: regarding (30) he mentions a gloss which interprets διὰ τάφρον as ἐπὶ τάφρον i.e. ‘up to the ditch’ (without crossing it), with no regard for the overall meaning of the text, which clearly implies crossing.
FN20 20)The latter is the case in the passage from which example (37) is taken. The following verses indicate that the spear continued and describe the further trajectory: ‘and straight on beside his flank the spear shore through his tunic; but he bent aside and escaped black fate’ ( Il.3.359-60).
FN21 21)Horrocks (1981, 255) leaves both possibilities open without going through all examples. Temporal meaning of διὰ νύκτα is also assumed in the recent book by Pietro Bortone, who considers this an example of “[i]ndiscriminate use of cases in PPs”, since, in his opinion, διά could indicate duration in time both with the accusative and with the genitive. As examples, he gives (out of context) the passage quoted here as (39) and a passage from Herodotus, in which διά with the genitive does indeed express time (Bortone 2010, 159). Apart from considerations regarding the real meaning of διὰ νύκτα, we should bear in mind that neither does διά with the genitive express time in Homeric Greek, nor is the PP διὰ νύκτα ever found in time expressions after Homer, a fact reported in all handbooks.
FN22 22)Verbs of perception are often considered as metaphorically corresponding to motion verbs in Greek, and there is clear evidence in later Greek theories of perception for such interpretation (see Luraghi 1989). In Homer, there is also evidence for the eyes to be conceived as containers, rather than channels, which could imply a stationary concept of perception (see Luraghi 2004). A discussion of this interesting issue is beyond the scope of this paper; in any case, in the light of the discussion below, it does not matter much in this context whether one takes the verb ‘see’ as implying metaphorical motion or not.