In Ancient Greek, topics can be expressed as intra-clausal constituents but they can also precede or follow the main clause as extra-clausal constituents. Together, these various topic expressions constitute a coherent system of complementary pragmatic functions. For a comprehensive account of topic organization, therefore, a limited focus on the clause proper is insufficient. In this paper, I will argue that it is possible to distinguish five different structural positions in which topic constituents may appear in Ancient Greek. These are: (i) Theme, (ii) clause-initial, (iii) postverbal in Setting, (iv) postverbal in main clause and (v) Tail. Each of these positions in the sentence is associated with a specific pragmatic function: Resumed Topic, Contrastive/New Topic, Given Topic or clarification of Given Topic. In linguistic theory, topic and focus are often seen as independent aspects of information structure instead of complementary functions. It is, therefore, attractive to posit two separate sets of constructional templates: on the one hand, a topic set comprising the aforementioned topic constructions and, on the other hand, a focus set containing two (narrow and broad) focus-constructions. This results in a flexible system in which the word order of each sentence is determined by a combination of a focus construction plus one or more topic constructions.
RuijghC.J.EichnerH.RixH.La place des enclitiques dans l’ordre des mots chez Homère d’apres la loi de WackernagelSprachwissenschaft und Philologie. Jacob Wackernagel und die Indo-Germanistik heute1990Wiesbaden213233[ = Scripta Minora, II, 627-47]
E.g. Givón 1983 and2001, 2, 227, 254. Gómez-González 2004 refers to the D-Topic’s retrospective and prospective potential.
See Rijksbaron 2006, Napoli2009, 592. The anaphoric function also explains the tendency which I found in my data that names with focus or names used to (re)introduce discourse participants (Themes) tend to lack the article. There may also be diachronical, genre-specific and even author-specific tendencies. Thucydides, for instance, does provide Themes with articles. The use of the article with proper nouns certainly deserves more study. I will not pursue this issue here any further.
See S. Dik1997, 1, 323; 2, 216-7; also H. Dik 1995, 2.
See e.g. Chafe1980, Sanford & Garrod 1981, Ariel 1990, 2001, 33, Van Vliet 2008, 44-7.
See Chafe1976, 34, Lambrecht 1994, 291-5, Givón 2001, 262. My category of Contrastive/New Topic by and large comprises exclusive contrastive topics and frame-setting topics as proposed by Matić 2003. I see no reason to maintain a distinction between exclusive contrastive topics and frame-setting topics. As Matić himself admits, there is no intonational evidence for this distinction. Furthermore, I am not aware of any examples in which both types of topics occur side by side. Contrastive Topics subsume Helma Dik’s Subtopics, i.e. topics relating to parts or aspects of a hierarchically superordinated entity (Dik 1995, 27-8).
The term Given Topic I owe to S. Dik1997, 1, 314. Matić 2003 uses the term continuous topic.