摘要：In recent years the semantic map methodology has enjoyed increasing popularity in cross-linguistic studies. Although there are various ways to create semantic maps, they are all attempts to visually represent cross-linguistic regularity in semantic structure. It has become increasingly clear that these attempts to map out linguistic categorization provide an empirically testable tool for the study of semantic variation across languages. Furthermore, the semantic map approach has shown convergence with grammaticalization theory as well as with research using (implicational) hierarchies, as found in functional typology and optimality theory. Early approaches to drawing semantic maps can be found in Hjelmslev (1963: 53, discussed by Haspelmath 2003: 237), Lazard (2001[1981]: 54), and Anderson (1982). In recent years, semantic maps have been proposed for diverse aspects of linguistic structure, including tense/aspect (e.g., Anderson 1982; Janda. 2007; Croft & Poole 2008), modality (Anderson 1986; Van der Auwera & Plungian 1998), voice (Kemmer 1993; Croft 2001), pronouns (Haspelmath 1997a; Cysouw 2007), case-marking (Haspelmath 2003; Narrog & Ito 2007; Rice & Kabata 2007), clause linkage (Kortmann 1997; Malchukov 2004), spatial and temporal domain (Haspelmath 1997b; Levinson & Meira 2003), as well as for a number of syntactic domains, such as intransitive predication (Stassen 1997) and secondary predication (Van der Auwera & Malchukov 2005). Recently, the issue of applying semantic maps to lexical typology¡ªas anticipated already in the early studies by Hjelmslev and Lazard¡ªhas also been taken up by Majid et. al. (2008) and Fran.ois (2008)
