Nunberg, Sag and Wasow (1994: 498)
departed from the earlier understanding of compositionality as a binary
phenomenon, which defined constructions either as idioms or as non-idioms; in
contrast, they defined compositionality as “the degree to which the phrasal
meaning, once known, can be analyzed in terms of the contributions of the idiom
parts”. Various studies have provided empirical support in favor of this
“compositional view” of language.
To name just a few, Cacciari and Tabossi (1988),
Glucksberg (1993), Peterson and Burgess (1993), and Titone and Connine (1994)
observed that even in highly non-compositional constructions, the semantics of
the component words are activated at least to some degree. Specifically, this
observation suggests that the literal meanings of words facilitate the
comprehension of idioms to the extent that they semantically overlap with the
idiomatic meaning (Gibbs and Nayak 1989; Gibbs, Nayak, Bolton & Keppel
1989).
According
to Baldwin (2004: 1) multiword expressions (MWEs) (including English phrasal
verbs) are “idiosyncratic interpretations that
cross word boundaries”. MWEs are sensitive to morphological and syntactic
flexibility, which enable the language user to conceive their meaning (ibid:
1). Baldwin (2004: 7) further proposes that multiword expressions, whose degree
of idiomaticity ranges from the literal to the idiomatic axis, can be
interpreted by virtue of “decompositionality”, termed as “the degree to which
the semantics of an MWE can be ascribed to those of its parts”.
At this point, I should
mention that both traditional and Cognitive approaches to idiomatic expressions
consider fixedness as an important factor studying multiword expressions like
phrasal verbs. To be more precise, “fixed expression” is a very general but
convenient term, adopted from Alexander (1978, 1979), Carter (1987) and others,
and used to cover several kinds of a phrasal lexeme, a phraseological unit, or
a multi-word lexical item. As Moon (2003: 2) observes these units include: i)
frozen collocations, ii) grammatically ill-formed collocations, iii) proverbs,
iv) routine formulae, v) sayings and, vi) similies.
Moreover,
other scholars assumed that fixed-form expressions, whether literal or
figurative in interpretation, are simply lexical entries precisely like those
assumed for standard literal interpretation of individual words (Swinney &
Cutler 1979; Di Sciullo & Williams 1987). Finally, in its simplest
description, a fixed figurative expression is a string of words for which the
interpretation is not (entirely) derived from the individual meanings of the
words comprising the string (even if there can be seen some historical linkage of the literal words to the overall
expression); this non-literal interpretation has become “fixed” in the language
by use (Hillert & Swinney 1999: 2).
The latter
remark suggests that an idiom compound's meaning may be: a) entirely
independent from the “literal” meanings of the individual words in the compound,
b) partially related to one of the “literal” meanings of the compound via
either structural or semantic analysis, or c) entirely ambiguous, having both a
“literal” and a “figurative” meaning (Hillert & Swinney 1999: 3). For
example, the English compound “redhead” is composed of: i) the structurally
decomposable literal meaning involving a head that is red in color (perhaps
from sunburn or dye), ii) the partially decomposable meaning of “having hair
that is red”, and iii) the non-decomposable figurative meaning of “hot tempered”,
a meaning derivable only by knowledge of personality features stereotypically
associated with persons who have red-hair (ibid: 3). The latter meaning has
come to be fixed by use, and is thus idiomatic. In similar manner, if we
consider the case of the English compound “horse-laugh”, we will observe that with
the exception of the small class of individuals who deal closely with horses
(and who might attribute the human descriptor of laughter to a horse), there is
no literal (decomposable or otherwise) interpretation of this compound; Yet the
fixed idiom is easily understood to mean a loud annoying (braying) laugh by a
human (ibid: 3).
Furthermore, according to
Wray (2002) phrasal verbs once conventionalized belong to the class of fixed expressions.
Fixed expressions refer to “specific combinations of two or more words that are
typically used to express a specific concept. […] The defining feature of a FE
is that it is a word combination, stored in the Mental Lexicon of native
speakers that as a whole refers to a [linguistic] concept. This makes FEs
non-compositional in the sense that the combination and structure of their
elements need not be computed afresh, but can be retrieved from the Mental
Lexicon. However, the degree of lexical and syntactic fixedness can vary” (Sprenger
2003: 4).
In this regard, the
notion “fixed expression” encompasses more than idioms. As Booij (2009: 221)
points out idiomatic expressions have one or more non-compositional properties,
whereas a fixed expression may be completely compositional, but nevertheless
stored because it is a conventional name for a particular concept. This view
argues that linguistic knowledge encompasses both knowledge of the grammatical
system of a language, and knowledge of the conventions involved in using that
language (ibid: 221).
Moreover, Gibbs (1995:
99) claims that the most important problem concerning the analyzability of
idioms derives from the traditional view according to which idioms are
non-compositional and hence, their meaning cannot be predictable from an
analysis of the meanings of their parts.
Gibbs (1995: 103) attempted to approach idiomaticity by relating it to
fixedness; in particular, he claims that it is of utmost importance to consider
that most languages have many idioms with similar figurative meanings. For
instance, American English has many idioms referring to the concept of getting
angry, such as “blow your stack”, “hit the ceiling”, “blow off steam”, “bite
your head off”, “get pissed off” (ibid: 103).
Gibbs (1995: 104) further
criticizes traditional approaches to idiomaticity stating that there is no
particular reason why we should create and use so many different expressions to
convey roughly the same idea or concept. He also disagrees with the view that
each phrase’s meaning is supposedly determined by separate historical
situations that have evolved into pragmatic conventions of use, implying that
the link between an idiom and its figurative meaning is arbitrary and cannot be
predicted from the meanings of its individual words (ibid: 104).
Contrary to the
traditional view, Gibbs criticizes traditional approaches by pointing out that the
figurative meanings of idioms might well be motivated by people’s conceptual
knowledge that is itself constituted by metaphor (ibid: 104). For instance, the
idiom “John spilled the beans” maps our knowledge of someone tipping over a
container of beans to that of a person revealing some previously hidden secret;
English speakers understand “spill the beans” to mean “reveal the secret” because
there are underlying conceptual metaphors, such as THE MIND IS A CONTAINER and
IDEAS ARE PHYSICAL ENTITIES, that structure their conceptions of minds,
secrets, and disclosure (Lakoff & Johnson 1980). Even though the existence
of conceptual metaphors does not predict that certain idioms or conventional
expressions must appear in the language (i.e. “spill the beans” vs. “spill the
peas”), the presence of these independent conceptual metaphors by which we make
sense of experience partially explains why specific phrases, like “spill the
beans” are used to refer to particular events (such as the revealing of
secrets) (Gibbs 1995: 104).
