Corpus-Based Approaches to Metaphor and Metonymy 225be none,‖ the use is much more metaphorical because here oases is used as an abstract term.
There are no shared semantic properties between control and the conventional use of oasis.
Hanks also notices the close relationship between ―syntagmatic patterns‖ and the
metaphorical use of lexical items. Unfortunately, this is not fully elaborated.
The chapter ―A corpus-based study of metaphors for speech activity in British English‖
examines how metaphor is used in written contexts to present verbal communication. Elena
Semino bases the research on a manually annotated corpus of ― 20 th century written British
English‖. The investigation is narrowed down to the ―Narrator‘s Representation of Speech
Acts‖. In this category, the focus is on expressions relating to two well-known conceptual
metaphors for communication—AUGUMENT IS WAR and the CONDUIT metaphor. Concerning
the former, the observations show that Lakoff and Johnson‘s formulation is simply too
narrow in terms of the source domain as well as the target domain. Therefore, Semino
presents a proposal for refining their mapping, namely, ―ANTAGONISTIC COMMUNICATION
IS PHYSICAL CONFLICT‖. Concerning the latter, the findings support Grady‘s refinement.
That is, the CONDUIT metaphor is best accounted for in terms of ―a set of partly independent
primary metaphors‖ (55). In the concluding section, Semino briefly also touches on some
metaphorical patterns not covered in the chapter. Together, the authentic data reveal that
verbal communication is dominantly realized by metaphorical expressions to do physical,
concrete interactions (57). In addition, Semino hints at the complementarity of corpus
research and experimental investigations.
In the chapter entitled ―Words and their metaphors: A corpus-based approach‖, Anatol
Stefanowitsch succeeds in demonstrating the superiority of metaphor pattern analysis over the
traditional introspective approach in identifying, extracting and quantifying metaphors. The
chapter is lucidly written, highly-organized, and well-argued. Stefanowitsch introduces the
reader to the data retrieving method of metaphor pattern analysis. There are two steps in the
procedure. First, instances of a certain target domain were extracted. Second, of all the
instances, metaphorical patterns are exhaustively identified (66). This was followed by a
systematic comparison between metaphor pattern analysis and the introspective method in
collecting data. Kövecses is targeted because his work on ―emotion-specific metaphors‖
represents the traditional method. Centering upon the five basic emotions of ANGER,
DISGUST, FEAR, HAPPINESS, and SADNESS, Stefanowitsch extracted a sample of 1000 hits for
the most frequent emotion term for each of the five emotions from the British National
Corpus (71). His findings are surprising and convincing. In comparison with metaphor pattern
analysis, the introspective method is handicapped in several ways. Typically, it fails to
include coverage of many important metaphorical manifestations for each basic emotion.
Statistics show, for instance, the introspective method misses 20.03% of the metaphorical
expressions from the domain ANGER (75), even 23.26% from the domain of SADNESS. After
the corpus illustration of the preciseness and much wider coverage of metaphor pattern
analysis, Stefanowitsch readdresses the issue of emotion-specific metaphors. Here he sets the
―frequency of occurrence‖ as the criterion because it is objective and ―relatively exhaustive‖
in identifying central metaphors for each emotion domain. The analysis reveals, for example,
that ―the most strongly associated metaphor for ANGER is EMOTION IS HEATED LIQUID‖ (92).
In the final section of his discussion, Stefanowitsch highlights the advantages of his corpus-
driven strategy through metaphor pattern analyses of representative lexemes within and
across target domains of emotion. Stefanowitsch concludes his chapter with an optimistic
prediction of the practical and theoretical significance of metaphor pattern analysis.