
8
ABSTRACT
NASCIMENTO, Mauro José Rocha do. Repensando as vogais temáticas
nominais a partir da gramática das construções. Tese de Doutorado em Língua
Portuguesa. Rio de Janeiro: Universidade Federal do Rio de Janeiro, 2006.
This work takes the Cognitive Linguistics as its theoretical basis, more specifically the
Constructions Grammar model developed by Goldberg (1995), adapting it into
morphologic constructions. The first proposal presented here is that word categories
can be considered grammar constructions by itself. We focused nouns and verbs,
particularly the formation processes which have relation with these two categories. It
is argued that the process when a noun is formed from a verb using no affixes,
traditionally called “regressive derivation”, it is in fact a case of a morphological re-
frame, as it is proposed .to be called. The same process may occur in an opposite
way, i.e. from the noun to the verb, relating just the elements which characterize
these categories. Having this proposal as its basis, it is argued that a variation of this
process occurs inside the noun category, relating noun sub-categories. These sub-
categories are also sets of constructions. Its form is related to thematic vowels and its
meaning is related to gender. About form, it is proved that speakers relate
automatically X-o constructions to masculine gender and X-a constructions to
feminine gender. About meaning, basing this hypothesis in Lakoff (1980), who claims
that people conceptualize the world from basic notions of the body experience, it is
claimed that speakers conceptualize grammatical notion of gender in basis of sex
differences. Because of this, there is that relation between gender and sex in most of
Indo-European languages. Inside the set of gender constructions, there is a basic
set, where gender and sex are directly linked (menino ‘boy’ / menina ‘girl’, gato ‘male
cat’ / gata ‘female cat’). In this group of constructions, there is a re-frame from -o, -e
and -∅ theme masculine constructions to –a theme feminine constructions, that is,
feminine gender is formed from masculine gender. That is because masculine gender
is the prototypical one, while feminine gender moves away from this prototype. The
prototypicity of masculine gender is a cultural data, which interferes in language. As
its consequence, masculine gender is in language the basic one, the more general
one, while feminine gender is always used just in specific situations. The most
productive re-frame related to livened up beings is the one which links –o and –a
constructions, like menino ‘boy’ / menina ‘girl’. A second group of constructions is
derived from this most productive pair, related to non-livened up beings. In this set of
constructions, gender is not related to sex (they are the constructions like jarro ‘jar’ /
jarra ‘specific kind of jar’). These constructions inherit from the basic group masculine
gender being the prototypical one and feminine gender being the less prototypical
one. In the first group the relation of prototypicity is about a more concrete
characteristic, i.e., the sex of the referent. On the other hand, in the second set the
prototypicity occurs in a less objective way. In that group, masculine gender