Syntactic conditions on null arguments in the Indo-European Bible translations

Abstract

In this paper I examine the use of null arguments in coordinated structures where both conjuncts have the same object. The data come from the Greek New Testament and its translations into old Indo-European languages. I conclude that in Greek, a null argument in the second conjunct is practically obligatory, both in cases of syntactic coordination and in cases of discourse coordination of syntactically subordinate participle clauses. The translation languages basically behave in the same way in licensing null arguments in coordinations, but tend to avoid discourse coordination of syntactic subordinates.

Keywords::

• anaphora
• syntax
• discourse
• null arguments

Notes

¹ Arguments can also be left out in cases where the referent is not easily recoverable, but rather relatively unimportant. For example, the equivalent of English one and German man can be expressed through a third person plural verb without a subject; and objects can be left out from otherwise transitive verbs when they are used generically. Such cases are not treated here. Glosses follow the Leipzig abbreviations, except for AOR = aorist.

² Unless we take autoû to have this function rather than being a possessive going with kheîra. But observe that autós in non-subject function rarely precedes its verb. This only happens in around 10% of the cases (249 vs 2253 cases according to the PROIEL corpus).

³ See Haug et al. (2009a, 2009b) for a description of the corpus and its annotation. The corpus can be browsed at http://foni.uio.no:3000 where it is also possible to download the source files. All Greek and Latin examples will be cited from the PROIEL corpus, which is based on the editions of Tischendorff (1869–1872) and Weber (1969).

⁴ Blass and Debrunner (1961,155) observe that pleonastic autos in relative clauses is suggested by Semitic usage, although it is also found in classical and later Greek.

⁵ ‘Backward’ is to be taken in the structural sense, so ‘upward’ would probably be more appropriate.

⁶ We simplify the situation here. It is likely that some translations were done not only on the basis of the Greek text, e.g. the Armenian version shows influence from Syriac and the Gothic translator Wulfila probably consulted a Latin version as well.

⁷ There is of course a rich literature on the early translations of the Bible, to which we cannot even begin to do justice here. But see Metzger (1977) for an overview, and de la Villa (2002) on the specific issue of the translation of participles in the Vulgate.

⁸ It is tempting to see the cause for the extreme literalness of these translations in the fact that Gothic and Slavic were in fact created as languages through these translations. The translators had no previous literary tradition to draw on. The situation was obviously very different for the Latin translators. Regarding the rendering of participles, it is of course also relevant that Slavic and Armenian, unlike Gothic and Latin, have an active past participle.

Additional information

Notes on contributors

Dag Trygve Truslew Haug

Dag Haug got his PhD in 2001 from the University of Oslo with a thesis on the Homeric language. He is now an associate professor of Greek and Latin at the University of Oslo. He works on the syntax and semantics of Greek and Latin, and is also interested in comparative Indo-European linguistics.