Page 384 - The ISA Handbook in Contemporary Sociology

9781412934633-Chap-24 1/10/09 8:55 AM Page 355

SOCIAL JUSTICE IN LATIN AMERICA 355

population size and growth rate; similar age and the fifth to the actor’s self-perception in
structure, rural-urban distribution, and reli- the context of social change. On the surface,
gious and ethnic make-up of the population.) these five questions address concerns and
On the other hand, as we already pointed out, topics with which all members of social
their political history has followed quite dis- movements are very familiar. In fact, in many
similar paths. We do not intend to explain – or cases, the answers may convey elements
even factor in – these differences here. Our taken directly from the movement’s official
goal in this chapter is to show how social rep- platform. But by giving respondents the
resentations in two neighbouring states with opportunity to elaborate and by telling them,
distinct structures display common patterns. at the beginning of the interview, that we
seek their personal perspective, we elicit
what the respondents apparently believe to
be their own thoughts on these matters. These
A DISCOURSE ANALYSIS particular questions lead respondents to
speak about social conflict and cooperation.
A five-question protocol was designed and They mirror the crucial, timeless questions of
interviews were conducted on a one-to-one collective action: What is to be done? What
basis. The sample was obtained through mul- brings us together? Who is with us and who
tiple snowballing (chain-referral) as well as is against us? We recorded an average of
from contacts with networks of social 4,000 words from each respondent, or
activists. (The interviews were conducted approximately 800 words for each of the five
under the author’s supervision by Rosa questions in our interview protocol. The text
Amelia Maltez, a graduate student at the database contains in all 158,000 words.
Université du Québec à Montréal.) In this Our method of discourse analysis is het-
regard, the sample is obviously statistically erodox as it does not apply the usual criteria
unrepresentative. However, we carried out found in most quantitative or qualitative
quantitative procedures on the respondents’ studies. Quantitative approaches usually
vocabulary in order to detect objective pat- focus on electoral and political data, event-
terns in their discourse. These procedures are analysis (including media coverage analy-
themselves statistically based and allow sis), and opinion polls in order to describe
probabilistic inferences about the subjects’ and explain the behaviour of the mobilized
word choices. The questions in the interview (or non-mobilized) individuals and groups.
protocol were extremely simple and pur- These approaches provide robust data and a
posely broad. Each of them contains one or comparative and cumulative perspective, but
two key words that aim to trigger reactions they fail to grasp the complexity, heterogene-
and associations: (1) What are the main ity, and internal logic of the activists’ repre-
injustices in this country? (2) Who benefits sentations. Qualitative approaches build on
from them? (3) What should be done in order historical research, ethnological observation,
to put the country on the right track? (4) Who and in-depth interviews with key informants
speaks on behalf of the people? (5) What is in order to grasp the way actors define
your role in this movement? The first ques- and experience their actions and their envi-
tion’s focus is on the notion of injustice, and ronment. These approaches provide insight
implicitly addresses the definition of ‘Us’. into activists’ discourse, but they (usually)
The second question encourages the respon- fail to produce robust data and a comparative
dent to elaborate on the issue of social and and cumulative perspective. In-depth semi-
political conflict, by naming the adversary. structured interviews can sometimes strike a
The third question refers to the goals and balance between the quantitative and qualita-
objectives of social mobilization, the fourth tive approaches, but they, for the most part, are
to the issue of representation and cooperation, based on context-specific research objectives.