TAM-inflected nominals

[Matthew S Dryer]

In response to Rachel Nordlinger's query "Does anyone know of any
languages in which clausal information such as tense, aspect, mood or
polarity is encoded on nominals or free pronouns (either in place of, or
in addition to marking on the verb)?":

I do think that it is necessary to distinguish a number of distinct
candidates for phenomena of this sort, which it may be a mistake to lump
together.  One is instances in which clause level information can show up
inflectionally on nominal elements in the clause regardless of their
grammatical role in the clause.  The Supyire pronouns are an instance of
this.  Nouns in Apurina, an Arawakan language, can inflect for tense,
though it is possible that this should be analysed as an instance of the
next type.

A second type is one in which clause level information can show up as
clitics on nominal elements in the clause, where the fact that it shows up
on the nominal element is in some sense accidental and independent of the
fact that the host is a nominal element: second position clitics are a
case of this, and the phenomena in Kayardild might be viewed as a special
case of this.

A third type is one in which clause-level information shows up on nominal
elements only when they are in some sense serving a predicative function
within the clause.  Lango is an instance of this.  A possibly distinct
subtype of this occurs in some Wakashan languages, like Nootka, in which
one can find such marking on nouns apparently occurring in different
grammatical roles but where the nouns can be analysed as predicative under
an analysis in which all occurrences of nouns in the language are
predicative.

A fourth type involves words associated with verbs that can be analysed as
part of a verb complex which vary for pronominal features of the subject
(or sometimes the object) and for other clause-level information.  While
such words are sometimes described in grammars as pronouns, they typically
occur in a fixed position before the verb that is distinct from the
position in which independent noun phrases occur and are thus are not
really pronouns inflecting for tense or aspect any more than they are
tense or aspect words inflecting for pronominal features of the subject of
the clause, and in that sense are not really nominal elements.  These
elements are ones for which the GB notion of "INFL" seems appropriate.  I
believe that Hausa phenomenon is an instance of this type.

Matthew Dryer