Using metaphors to recount one’s life: A corpus-driven study

Tony Berber Sardinha (PUCSP)

The aim of this paper is to look at the metaphors that people employ when they tell their life story. This should help us better understand how people reflect upon their

lives as they recount their experiences using metaphor, as well as aiding in assessing the role of metaphor in helping people make sense of their lives. The corpus consisted of 60 narratives spoken in Portuguese (each by one different person) taken from the archives of the Museu da Pessoa, an organization based in São Paulo (Brazil) that gathers narratives told by ordinary people as a means for preserving oral history, founded in 1991. It spread to other countries, and the network now comprises other museums in the US, Portugal and Canada. It ‘celebrates the lives of people from all walks of life - those who give heft to the fabric of our community through their daily work, play, traditions and dreams.’ (US Site). Its aim is to keep the memory of communities alive, which otherwise would not be recorded. The corpus amounts to about 63 hours of recording, 617,371 tokens. Most of the interviewees were retired working class (farm hands, factory workers, seamstresses, housewives, etc.) in their 60’s-70’s, with little or no schooling. Finding metaphor in corpora presents a problem for researchers, because corpus linguistic tools such as concordancers require that researchers know which words are metaphorically used in the corpus in advance. The corpus was first processed by the CEPRIL Metaphor Candidate Identifier, an online tool that screened the corpus for words with metaphoric potential and suggested possible metaphorically used words. The program output was then hand-analyzed for linguistic and conceptual metaphors. A linguistic metaphor is a stretch of text in which it is possible to interpret incongruity between two domains from the surface lexical content (Cameron 2002: 10); an example from the corpus is ‘no fundo dos olhos’ (‘deep in one’s eyes’), which is a linguistic metaphor because it includes ‘fundo’ (‘bottom’), a word relating to the field of containers, to describe the eyes (which are not literal containers), thus turning the eyes into  metaphorical containers for emotions. Next, conceptual metaphors were identified. Conceptual metaphors are abstract conceptualizations that underlie linguistic metaphors,  such as life is a load (‘vida pesada’: heavy life), life is an object (‘vida toda’: whole life), life is a battlefield (‘vida de luta’: life of struggle), life is a journey (‘levar a vida’: lead one’s life), and life is a building (‘fazendo a minha vida’: making one’s life). There are three major findings in this study. The first is that metaphor is sparse: 1698 metaphorically used words (MUW) were found in the corpus, thus giving a metaphor density of 1 MUW every 364 words. Metaphor density in previous research was considerably higher: in Cameron (2003), 1 every 37 words (classroom discourse), and in Berber Sardinha (2008), 1 every 22 words (investment conference calls). Low density indicates people can ‘do without metaphor’, even when speaking about complex issues such as telling their own life stories. This has not been adequately reflected in the literature, which tends to reinforce the notion that metaphor is indispensable in all situations. Difference between speakers was noted, with more politically inclined, such as catadores de lixo, using metaphor more extensively; those with more formal education, more metaphors; and those with less formal education, fewer metaphors. Most metaphors may have been acquired in particular specialized contexts, such as schools or political organizations. The second finding is that the source of the metaphors (ie, in terms of what metaphors are created) are biased toward two particular ones: journey and machine, which together account for 54% of all metaphor instances. Examples of conceptualizing life metaphorically along the journey domain appear when interviewees talked about: life events going forward, life events ‘walking’ ( = going), catching a boat in life, following a path, going on the right/wrong path, points in the journey (‘from there’ = and then). And in terms of the machine domain, they talked about life as: people/things giving support, creating/having links between people/events, assembling events, connecting events and setting events in motion. This finding suggests that the range of domains for conceptualizing metaphor though potentially vast is actually quite restricted. Speakers could have chosen out of a huge number of domains, but for the most part they picked only two. The third finding is similar to the previous one; in terms of target / topic domain (that which is being metaphorically conceptualized), speakers show a restricted repertoire: three domains account for 62% of all metaphors. These three domains are time, work and life. Time is conceptualized as, for instance: TIME PASSING IS ADVANCING IN A JOURNEY; TIME IS A LONG, SHORT OBJECT; PERIODS OF TIME ARE PORTIONS OF A JOURNEY. Work is metaphorized as (for instance): STARTING A JOB, WORK, CAREER IS GOING INTO A CONTAINER, HAVING A CAREER IS STICKING YOURSELF TO IT,

HAVING A CAREER IS BEING WRAPPED UP IN IT. And life is metaphorized as LIVING LIFE IS GOING ON A JOURNEY, LIVING LIFE IS CARRYING AN OBJECT, and LIVING LIFE IS FIGHTING A BATTLE. This finding suggest that when talking about one’s life, life itself is not the concept that gets metaphorized more often. Time is actually more frequently talked about using metaphor. The conclusion of the study is that empirical corpus-based findings of metaphor use may contradict previous literature, which claimed that metaphor is ubiquitous, and that implied that source and target domains can be chosen from a wide range of choices. These findings indicate that at least in this particular setting of recounting one’s life in unstructured interviews, people tend to use few metaphors and restrict the range of domains that are activated for metaphor production.

Palavras-chave: Metaphor, corpus-driven methodology.