A similar approach to phrasal syntax and morphology, advocated by Autolexical Syntax (Sadock 1991), Role and
Reference Grammar (Van Valin and LaPolla 1997), and recent Lexical-Functional Grammar (Bresnan 2001), treats
them as somewhat independent systems that accomplish partially overlapping functions. Section 5.6 referred to these
systems as the tiers of phrasal syntax and morphosyntax, and pointed out some of the differences in the sorts of
combinatoriality they offer. These systems, like those involved in depth perception, interweave with each other,
sometimes redundantly, sometimes not, but both helping make explicit the semantic relations among components in
an utterance. For instance, phrasal syntax may signal thematic roles (who did what to whom) through the order of
phrases in relation to the verb. Inflection may do the same thing by means of verb agreement with the subject (and in
some languages, with the object as well). Inflection may alternatively express semantic roles through a system of case
marking, as in German, Russian, and Latin. In particular, many of the so-called“semantic cases”in these languages
have much the same character as the phrasal markersforandwithdiscussed in the previous section. And parallel to
verbs in English that require“governed prepositions”(section 5.8), case-marking languages often have verbs that
govern so-called “quirky case” on an argument. Languages tend to mix and match these strategies in different
proportion; languages with rich inflectional systems often allow more freedom in word order for different purposes,
usuallyfor focus–topicinformation. On theotherhand, inflection can be used (freelyor redundantlywithwordorder)
to indicate focus or topic as well, for example the Japanese suffix-wa, which typically marks topic redundantly with
initial position in the clause.
Thus we might think of phrasal syntax and morphosyntax as independently evolved systems, each built on top of the
syste mof protolanguage, each refining communication through its own expressive techniques. In a similar vein, Casey
and Kluender (1995) suggest that agreement inflection evolved as an extra syste mto provide redundant (and hence
more reliable) information about semantic relations of arguments. I see no immediate reason to assert the temporal
priority of one of these systems over the other in the course of evolution.
Notice that parts of these systems, especially inflection, depend strongly on the noun–verb distinction. Verbs are
marked for agreement with nouns; nouns are marked by verbs for case.
An important innovationthat remains is the system of grammatical functions: subject, object,and (insome languages)
indirect object. Section 5.10 suggested that this syste m may well for ma separate tier of syntax,“GF-structure,”
interfacing with both phrasal syntax and morphosyntax and relating them to semantic functions. The present
evolutionary context invites the intriguing