The (non)veridicality theory of polarity and negator selection

In this chapter the theory of nonveridicality is presented, along with the relevant evidence from Standard Modern Greek regarding negator choice. It is argued that negator choice in Modern Greek must be understood as a polarity phenomenon: NEG2 μη‎( ν‎) /mi(n)/ is a polarity item licensed in nonveridical contexts. For this reason NEG2 cooccurs with subjunctives, questions and other nonveridical elements. Furthermore, crosslinguistic extensions are identified of the idea that negator choice depends on nonveridicality regarding a number of indoeuropean (Vedic, Hittite, Armenian, Albanian), as well as typologically and genetically unrelated languages, such as Zulu and Algonquian.

Late Medieval and beyond: the renewal of NEG1

This chapter discusses negator distribution in Late Medieval Greek with texts from the twelfth to the fourteenth centuries AD. Spoken Greek of this period depicts a stage of variation in the forms of NEG1 and NEG2. Both elements have developed counterparts with which they appear in free variation, as the former NEG1-thing, οὐδέν‎ /udhén/, and NEG2-thing, μηδέν‎ /midhén/, indefinites have bleached into plain sentential negators. Two instances of parameter resetting in Late Medieval Greek are identified: (i) the specifier-to-head shift in the syntactic status of the negators and (ii) the loss of NEG2 from the conditional protasis. The NEG2 was lost from the conditional antecedent for reasons that relate to its syntactic status shift and exact location on the expanded CP.

Negation, mood, and (non)veridicality in Classical Greek (fifth to fourth centuries BC)

This chapter provides a mapping of the functions of the Attic Greek NEG1 οὐ‎(κ‎) /u:(k)/ and NEG2 μή‎ /mε‎:/particles and a unified explanation of their distribution regulated by the notion of nonveridicality. Core properties of the Attic dialect are presented, regarding verbal morphology (grammatical information on verbs, infinitives and participles) and word order. The phrasal status of the Attic negators is discussed and their inconsistent relation to mood is exposed. The non-negative functions of NEG2 μή‎ /mε‎:/ are also identified: as a particle introducing yes/no questions, a complementizer and an attitudinal marker. The polarity behavior of the Attic Greek negators is extended to the morphologically negative paradigms of Attic indefinites: NEG1-words (e.g. οὐδείς‎/ u:de:s/ ‘no one’, οὐδέν‎ /u:den/ ‘nothing’ οὐδέποτε‎ /u:depote/ ‘never’) and NEG2-words (e.g. μηδείς‎ /mε‎:de:s/ ‘no one’, μηδέν‎ /mε‎:den/ ‘nothing’ μηδέποτε‎ /mε‎:depote/ ‘never’). NEG2-words are exclusively licensed in nonveridical environments.

Renewal and stability

This chapter discusses the Greek negator transformations in relation to Jespersen’s Cycle. The developments of NEG1 and NEG2 in Greek do not properly qualify as instances of Jespersen’s Cycle in the traditional understanding of the phenomenon, as it did not manifest a doubling stage. A new approach for Jespersen’s Cycle is proposed, which accommodates not only Greek, but also various other atypical languages that deviate in one way or another from the traditional morphosyntactic description of the phenomenon. It is proposed that Jespersen’s Cycle is a diachronic phenomenon whose regularities are to be found in the semantics. An overview is also provided of the diachronically stable functions of NEG2, which are the COMP-related functions of NEG2 μη‎. It is argued that NEG2 μη‎ did not eventually renew, because of the inertial pressures of its several nonnegative functions, which, being nonnegative, were not affected by Jespersen’s Cycle phenomena.

Developments in Hellenistic-Roman times and the Nonveridicality projection (third century BC to fourth century AD)

This chapter examines sentential negation during the Hellenistic Koine stage of Greek based on non-atticizing texts mainly from the first century BC to the second century AD. Structural developments of the language are presented that support a treatment for nonveridicality as encoded in a syntactic projection, independent from morphological mood and independent from complementizer position. A treatment of the licensing of polarity items is proposed—among which is the Greek NEG2—in terms of syntactic agreement. Nonveridical operators are taken to introduce the Nonveridicality Phrase (NONVERP) in syntax, encoding the observation that nonveridical environments tend to be morphologically marked in ways that can be distinct from mood marking. Furthermore, in the Koine Greek stage, NEG2 gets more specialized in its lexical negation function at the expense of NEG1, while Negative Concord structures get significantly reduced, a change that was linked to Greek word-order particulars.

Conclusion

We have tracked the history of the Greek negator system through a timespan of over 2.5 thousand years, through quantitative evidence from three major stages of spoken Greek: Classical Greek, Koine Greek, and Late Medieval Greek and qualitative evidence from Homeric Greek up to Standard Modern. The presence of a second negative element, NEG2, that is particular to nonveridical environments in the sense of Giannakidou (1998 ...

Introduction

The introductory chapter provides the basic facts on the diachrony of negation in Greek, the renewal of NEG1 and persistence of NEG2 in its core functions. A periodization of the Greek language is given along with details on the methodological approach and text selection. A brief historical overview on the matter of negator selection in Greek is provided starting from the Classical scholars of the 18th to early 20th century until more recent studies within generative grammar. The general reasoning of the argumentation and analysis is condensed regarding the positing of a Nonveridicality syntactic projection, discussed in relation to the notion of markedness, for which pragmatic grounding is posited: the Gricean maxim of quality.