Semi-Lexical Categories
20th May 2016
Seminar organized by the research community CoCoLaC
Lecture hall 12, Metsätalo (Unioninkatu 40)
University of Helsinki
Abstracts
page
Lene Schøsler (Københavns Universitet): A diachronic perspective on support verb constructions
in French
2
Antonio Fábregas (Universitetet i Tromsø): On light verbs, light nouns and why there are no
light adjectives
2
Nicole Delbecque (KU Leuven): Constructions at the intersection of lexicalization and
grammaticalization. Some convergences and divergences between Spanish and French
3
Mona Forsskåhl (Tampereen yliopisto): Swedish motion verbs as parts of idiomatic phrases.
A frame semantic approach
3
Sari Mäittälä-Kauppila (Helsingin yliopisto): On translating the English negative prefix of
non- and its pejorative overtone into Finnish
4
Ángeles Carrasco Gutiérrez (Universidad de Castilla la Mancha): Progressive states?
5
Jukka Havu (Tampereen yliopisto): Estar + participle in (Ibero-)Romance languages
5
Alexandre Nikolaev, Marja Nenonen, Juha Mulli and Esa Penttilä (Itä-Suomen yliopisto):
How light can a light verb be? The constructional nature of verb ‘take’ in four languages
6
Ciro Imperato (Helsingin yliopisto): A cognitive linguistic-based contrastive analysis of
the Italian verb prendere and the Finnish verb ottaa
6
Lene Schøsler (Københavns Universitet)
A diachronic perspective on support verb constructions in French
Latin (almost) only has simple verbs. Why do all Romance languages have many complex verbal structures,
especially many so-called support verb constructions (svcs), like vivre ça me fout la trouille. Mourir plus
encore, mais vivre, ça me fait vraiment peur, (‘Life scares me. Death even more, but I am really afraid of life’)?
Two additional questions arise from my first one: 1. How can we define svcs and distinguish support verbs
from “full verbs” and from “auxiliaries”? This question is studied in part 1 of my paper. What are the reasons
for using svcs in cases where you have both simple verbs and svcs? - is there something that svcs do that
simple verbs cannot do? this question is studied in part 2 of my paper.
My hypotheses on svcs are: 1, that they are used where no verb exists; 2, that they are preferred to a simple
verb for the following reasons: a, they express aspects that a simple verb cannot express; b, they have a
syntactic structure that permits a focus on the verbal situation or action; and 3, that they are results of lexical
innovation processes based on existing elements, some of which - I claim - can be analysed as
“grammaticalised”. Svcs can be considered as cases of “lexical recycling”.
In part 3 of my paper I will address my initial question why Romance languages have svcs when Latin did not
have such analytical constructions. Afterwards, in part 4, I will study svcs in a diachronic perspective,
diachrony being one of the five parameters of variation, which is distinguished in variationnal linguistics.
Other parameters of variation will be taken up in part 5.
In part 6, I will defend my hypothesis that svcs can be considered as grammaticalised models of lexical
innovation.
Antonio Fábregas (Universitetet i Tromsø)
On light verbs, light nouns and why there are no light adjectives
The goal of this talk is to discuss the consequences that the existence of light verbs (and nouns) have for the
analysis of lexical categories; we will use them to argue that lexical categories are always derived in syntax
through a sequence of heads; this predicts not only that light verbs exist, but also that other classes, such as
auxiliaries, should exist. On the base of the less controversial properties of light verbs, we will identify a class
of light nouns. This will allow us to show that, while the properties of the light nouns are rigidly ordered in a
functional sequence, there is no evidence for the same rigid order in adjectives that combine with them.
Combined with a second observation, that light adjectives do not exist, we will argue that adjectives are not
primitive categories, or, in other words, that they do not exist as lexical classes.
2
Nicole Delbecque (KU Leuven)
Constructions at the intersection of lexicalization and grammaticalization. Some convergences and
divergences between Spanish and French
The tendency to polysemy and the exploitation of a wide range of possibilities between grammaticalization
and lexicalization, considered as the two poles of a continuum, can be assumed to be higher in Spanish than
in French. Empirical evidence for this claim will be explored for three kinds of phenomena.
(i) Reflexive constructions receiving a middle voice, a middle-passive or a 'reflexive' passive interpretation
represent a relatively open grammatical option in Spanish, while in French they are more a matter of verbal
lexicon. The point can further be illustrated by the larger paradigm of Spanish ‘pseudo-copular’ constructions,
with certain relatively lexicalized associations, as well as by the existence of other specific construction types
(e.g. se VP with dative clitic).
(ii) The Spanish non-periphrastic gerund construction encompasses the manner component expressed by the
French gerund (en V-ant) and the explanatory comment expressed by the French present participle (V-ant).
It extends over a continuum going from quasi periphrastic constructions to adjoined ones, which zoom in on
a co-event from a privileged insider’s perspective.
(iii) The Spanish verb dar ‘give’ admits a great diversity of constructions and lexical associations. Its syntactic
and semantic potential illustrates how metonymic and metaphoric extensions yield semilexicalizad uses as
light verb as well as more auxiliary ones. Although the French counterpart donner ‘give’ shares certain
combinatorial capacities and conceptual associations, its coverage is somewhat narrower.
Mona Forsskåhl (Tampereen yliopisto)
Swedish motion verbs as parts of idiomatic phrases. A frame semantic approach
Motions verbs have long been of special interest for both intra-linguistic, structural persepectives and for
cross-linguistic ones. From a typological point of view, Leonard Talmy (1972, 1985) model for categorizing
languages based on how they express “a motion situation” is one of the most widely influential. He suggested
each motion situation involves the semantic elements FIGURE (the object or person moving), GROUND, PATH
and MANNER, and these are expressed differently in different languages. Talmy’s notion of semantic
elements encoded in different forms is well aligned with a frame semantical and constructional perspective
(see Fillmore 1985; Fillmore and Baker 2001, 2010, Boas 2008), where word meaning is conceptualized as
semantical frames, consisting of specific sets of frame elements, and words are seen as constructions on the
same basis as other linguistic structures, e.g. as pairings of form and meaning. (e.g. Fillmore 1988, Fried &
Östman 2004; Croft et al. 2010). A main point in this paradigm is that word and constructional meaning
directly and inherently influences the way words and constructions interact with their structural surrounding.
In this paper, the aim is to relate the meaning of motion verbs to their use in different structural contexts,
3
discussing how the meaning of motion verbs and the meaning of idiomatic phrases (in which they are used)
as a whole are related.
I will adopt a frame semantical and constructionist perspective on a group of Swedish motion verbs: gå, fara,
åka, sticka, köra and focus on their use in specific idiomatic phrases (constructions such as pseudocoordination: motion verb + och + verb; motion verb + PP; motion verb + particle). The questions addressed
are: How is the motion sense of the verbs brought into the constructions in which they are used? What is the
relation between the constructions and the verbs as far as the verbs (different) meaning components are
concerned? The constructions analyzed include types where the motion sense of the verbs remains, and
types where it does not. The data consists of texts from a set of sub-corpuses from the large text-corpus
provided by Språkbanken (sprakbanken.gu.se). The Swedish FrameNet and the Swedish Constructicon (ibid.)
will be used as tools for identifying and analyzing idiomatic phrases (constructions) including the focused
verb in the corpuses.
Sari Mäittälä-Kauppila (Helsingin yliopisto)
On translating the English negative prefix of non- and its pejorative overtone into Finnish
It is in the nature of human being to exhibit utmost verbal creativity when expressing less favorable opinions
of other people and their ideas. Irony is a convenient tool and can take many forms and produce intriguing
stylistic expressions. In English, a negative construction with the prefix non- is one of the most subtle ways
of conveying irony.
Originally French, the prefix non- entered the English language as a syntagmatic loan. The pejorative use of
what was originally a privative prefix appeared for the first time in the 1960’s, and its use as a means of
expressing a hidden critical opinion has increased considerably ever since. New formations are mainly of an
ad hoc type and highly context-dependent and reflect the writer’s verbal ingenuity. A newspaper describes
a collection of furniture under a playfully vague heading of Non-design Design and, in another newspaper,
some office workers admit that the week before Christmas is actually a non-work week as they only pretend
to be working and instead are planning their holidays and sneaking off to do the shopping. A musical
performance characterized as non-music may refer to a composition with new unconventional additions or
themes, or it may be a rude and critical comment on the performer’s talent.
In my presentation, I will approach the pejorative use of the prefix non- from the framework of translation
theories. Translating irony in non- may prove challenging, since a good translation does not reveal itself to a
translator automatically, but what is required is a thorough analysis of the text and a solid knowledge of
several structural and semantic aspects of the source and target languages. I examine depreciatory
expressions with the prefix non- in journalistic texts and fiction, and discuss the challenges in the translation
process. I will address features of several textual genres and various methods of identifying irony, and suggest
applicable translation theories and strategies. In connection with the question of whether irony embedded
in the prefix non- can fully be conveyed in the translation, I will cite certain sections from my forthcoming
4
doctoral thesis. Examples will include texts from American and British newspapers and magazines and their
Finnish translations by students of English translation at the University of Helsinki.
Ángeles Carrasco Gutiérrez (Universidad de Castilla la Mancha)
Progressive states?
The aim of this study is to explain why the progressive non-stative predicates behave as the stative ones with
respect to a set of stativity tests. More specifically, our purpose is to prove that the behaviour of the Spanish
periphrasis <estar + gerund> in those tests is derived from its nature as a content of Phasal Aspect. So, the
progressive will be shown not to be either a stativizing device, or a variety of the imperfective Grammatical
Aspect.
Jukka Havu (Tampereen yliopisto)
Estar + participle in (Ibero-)Romance languages
The semantic area of estar + part. in Ibero-Romance languages is determined by the aspectual properties of
the predicates. In Spanish, the construction estar + PP has a very wide range of use. Traditionally, it is the
fundamental expression for the resultative passive with transitive predicates and the expression of a resulting
state with unaccusative predicates, ser + PP being the periphrasis of the eventive passive. However, there
are a number of deviating cases: i) unaccusative motion verbs are not compatible with estar + PP in their
basic meanings (?? Paco está entrado ‘P. has come in’), but combine with it in a metaphorical sense (Paco
está entrado en años ‘P. is getting on in years’); ii) many telic predicates do not produce a resulting state and
reject the expression estar + PP (Laura tocó la sonata > *La sonata está tocada ‘L. played the sonata’).
However, these predicates are often compatible with the periphrasis when they are accompanied by an
expression of an agent (la sonata está interpretada por un solista famoso ‘The sonata is played by a famous
soloist’), in which case they are more or less equivalent to the expression ser + PP; iii) with many degree verbs
that have an activity and an accomplishment reading (e.g. engordar ‘to get fat / fatten’), the periphrasis estar
+ PP combines much more naturally with the accomplishment interpretation (el presupuesto está engordado
‘The budget has been fattened’ vs. ?Paco está engordado ‘P. has become fat’); iv) the compatibility of many
activity predicates with estar + part. gives rise to slightly surprising phenomena; for example, the participle
of the verb ayudar ‘to help’ is very rare with estar (estaba ayudado por un médico ‘He/she was being helped
by a physician’), but verbs like asistir, auxiliar, secundar, practically synonymous with ayudar, combine more
freely with estar; v) the geographical variation is considerable; in European Spanish, expressions like está
atropellado ‘He/she has been run over’ or está fallecido ‘He/she is deceased’ are practically non-existent, but
more common in some Latin American countries (whereas está muerto ‘He/she is dead’ is normal in all
varieties).
5
A few remarks will be made of parallel structures in some other Romance languages; in Italian and Romanian
STARE + part. il museo sta aperto fino a tardi; Rom. muzeul stă deschis până târziu ‘The Museum is open
until late’). There are also other uses, especially of posture (It. sta seduto/-a and Rom. stă așezat/-ă ‘He/she
is sitting’), and cases where the construction is accompanied by a locative expression (It. la figura sta dipinta
nella volta ‘The figure has been painted on the vault’; Rom. o hartă stă lipită pe perete ‘A map has been glued
on the wall’).
Alexandre Nikolaev, Marja Nenonen, Juha Mulli and Esa Penttilä (Itä-Suomen yliopisto)
How light can a light verb be? The constructional nature of verb ‘take’ in four languages
This study investigates the predication patterns in verb phrase constructions consisting of the verb ‘take’ and
its NP complement in four languages, English, Finnish, German, and Russian. ‘Take’ is a typical light verb but
at the same time one of the most idiom-prone verbs as well. In order to test how the predicative meaning of
VP constructions containing ‘take’ is distributed in the construction, we collected randomly selected samples
of 600 instances of these constructions for each language in two textual genres, newspapers and fiction.
The main aim in this investigation is to look at how light or heavy the verb ‘take’ in the actual constructions
is. At the same time, we aim to test and discuss the methodological applicability of the criteria that we have
chosen to use as the basis of our data analysis. The idea is to define the weight of the verb on the basis of
how the constructional meaning is distributed. If the meaning can be regarded as mainly residing in the verb,
the verb would be regarded as heavy, but if the meaning can be located in the complement, the verb would
be regarded as light. In between the two extremes are the constructional expressions where the meaning is
divided between the verb and the complement and neither of them can be regarded as the main carrier of
the constructional sense. In this case, the verb would not be light, and the meaning would rather be
constructional than lexical. One of our central ideas thus is to test whether this type of operationalization
offers reliable tools for analyzing the semantic distribution within VP constructions.
Ciro Imperato (Helsingin yliopisto)
A cognitive linguistic-based contrastive analysis of the Italian verb prendere and the Finnish verb ottaa
The aim of this paper is to compare the multiple senses of two highly-polysemous verbs, the Italian verb
prendere and the Finnish verb ottaa, both corresponding to English verb to take. The analysis will be carried
out within the theoretical framework of Cognitive Linguistics.
In accordance with this linguistic approaches, the array of distinct senses of these two light-verbs will be
accounted for by showing how each sense is an extension of another and how these senses can be described
on the basis of a proto-scene consisting in the concrete event of taking performed in the physical world.
6
A proto-scene of the prototypical act of taking implies the idea of manipulation, force dynamic, possession
and transfer of control. Hence, these concepts will be appealed to in describing the various metaphoric
expressions of the two verbs under consideration. The analysis aims also to organize the multiple senses of
the verbs into a coherent and well-structured radial network.
Unlike the most studies inspired by Cognitive Linguistics, which seek to account for the meanings associated
with words in a single language, the present study will evaluate possibilities of a cognitive linguistic-based
analysis to be applied for contrastive purposes. Two basic assumptions underlying the study are: 1) the protoscene can provide a departure point for comparison of figurative senses generated by two distinct verbs in
two different languages, 2) different senses can be described in terms of ‘profiling’, that is, some elements
of the proto-scene can be foregrounded or backgrounded, giving rise to cognitively motivated metaphoric
extensions deriving from the central mining of the verb.
The examples forming the corpus of the study are extracted from dictionaries, translations and the Web.
7