SciELO - Scientific Electronic Library Online

vol.3 special edition author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Interface - Comunicação, Saúde, Educação

Print version ISSN 1414-3283

Interface (Botucatu) vol.3 Botucatu  2007


The Collective Subject that speaks


O sujeito coletivo que fala


El sujeto colectivo que habla



Fernando LefevreI; Ana Maria Cavalcanti LefevreII

IProfessor titular, Faculdade de Saúde Pública, Universidade de São Paulo (USP). <>
IIInstituto de Pesquisa do Discurso do Sujeito Coletivo. <>

Translated by Cauê Cardoso  Polla
Translation from Interface - Comunicação, Saúde, Educação, Botucatu, v.10, n.20, p. 517-524, July./Dec. 2006.




This paper discusses the Discourse of the Collective Subject as a qualitative-quantitative proposal for opinion polling or research on social representations. The authors propose the presentation of collective opinion in research as an empirical variable of qualitative and quantitative nature. This is achieved by introducing a subject of discourse, who is individual and collective at the same time. This empowers the speaker to express him or herself directly, without the intervention of the researcher's meta-discourse and avoids converting opinion in a mere quantitative variable, mutilating its essentially discursive nature.

Key-words: discourse of the collective subject; methodology; qualitative research; social representations.


Discute-se aqui o Discurso do Sujeito Coletivo como proposta qualiquantitativa para as pesquisas de opinião ou representação social. Propõe-se a apresentação, nas pesquisas, da opinião coletiva como uma variável empírica de natureza qualitativa e quantitativa capaz, pela interposição de um sujeito de discurso ao mesmo tempo individual e coletivo, de se exprimir, diretamente, sem a mediação do metadiscurso do pesquisador e sem a transmutação da opinião em variável quantitativa, com prejuízo de sua natureza essencialmente discursiva.

Palavras-chave: discurso do sujeito coletivo. metodologia. pesquisa qualitativa. representação social.


Se discute aquí el Discurso del Sujeto Colectivo como propuesta cualicuantitativa para las investigaciones de opinión o representación social. Se propone la presentación en las investigaciones de la opinión colectiva como una variable empírica de naturaleza cualitativa y cuantitativa capaz, por la interposición de un sujeto de discurso al mismo tiempo individual y colectivo, de expresarse, directamente, sin la mediación del metadiscurso del investigador y sin la transmutación de la opinión en variable cuantitativa, con perjuicio de su naturaleza esencialmente discursiva.

Palabras clave: discurso del sujeto colectivo. metodología. investigación cualitativa. representación social.




This work aims at reflecting upon the possibilities that are offered to express a collective opinion or thought empirically. Taking into account the fact that collective opinion as an empirical fact is conveyed only indirectly by the researcher's meta-discourse or by means of a given mathematical formula (losing its immanently discursive form) one proposes as a truly expressive alternative the Discourse of Collective Subject (DCS).

The proposal of the Discourse of the Collective Subject (Lefevre & Lefevre, 2003), associated to the Qualiquantisoft software ( is grounded mainly in the presuppositions of the Theory of Social Representations (Jodelet, 1989). It lists and articulates a series of operations on raw data constituted of statemens obtained through empirical polling using open-ended questions, what results in collective statements made up of different extracts of individual statements. Each collective statement stands for a determined opinion or position and is written in the first person singular aiming at producing in the receptor the effect of a collective opinion expressed as an empirical fact through the "mouth" of a single subject of discourse.

The application of the DCS technique to a great number of empirical researches in the field of public health and also in other fields (DCSs's bank) has been showing its efficacy for the processing and expression of collective opinions.

The accumulated experience in using the DCS methodology shows an increasing perfecting of the technique and its various applications. Hence, in general, the most recent works present this methodology at its most.

In this sense we would like to highlight in what regards academic works the following: Valverde (2006), on overweight, Medina (2005) about internet forums dealing with urban violence and Akyiama (2006) on phonoaudiologic intervention in deafness.

In what regards non-academic works, it's worth citing the work that evaluates the Post-Graduation Program of the National School of Public Health (Rivera, 2005), representing an important application of the methodology to institutional assessment. Another work is Levèfre et al.'s (2005) which uses among other resources the discourses of preteens students allowing the detailed description of the subjective representation of the day-to-day relationship between parents and sons affected by the consume of cigarettes by the parents.


Stalemates for expression the collective thought.

The challenge faced by the DCS is the searching for the self-expression of the collective thought or opinion, preserving its double condition of being both qualitative and quantitative. In effect, if one considers the framework of empirical research, thought seen as material, as signifying matter, is a previously unknown result (for the empirical researcher), i.e., is a discourse inductively obtained. This thought shows itself undoubtedly as a qualitative variable, as a product to be qualified by the research a posteriori as an output. Nevertheless, being this thought of collective character it is also quantitative to the extent that it must express opinions shared by a certain number of individuals that configures the researched collectivity.

Given this scenario one of the challenges to be defeated in order to allow the collective thought to express itself through empirical research would be the constitution of a subject bearing this collective discourse.

However, how can one verbally express this collective subject as a subject-that-speaks-directly by not using a mathematical expression or a scientific "they say" (even "one says", impersonal subject)?

Now, apparently such a collective subject cannot speak if it is kept bound positivistically to the possibilities offered by the language it speaks (the Portuguese one or others);  we will only have a precarious way to access directly this collective subject which is the pronoun "we" – first person plural. The "collective I" is not an alternative.

Now, a collective subject as understood by us in the Discourse of the Collective Subject is more than a "we" that express only a very particular kind of collective subject that speaks; it is also not less because one single subject can be a collective subject.

In traditional opinion polling, the subject that states his opinions (that who speaks: "in my opinion…", "I think that…" or "I believe that…") is almost always a single subject, or at most a "we". Thus, a collective subject would not find direct forms to express itself and "henceforth" would come to be non-existent. To be more precise it wouldn't be seen as a speaker susceptible only of being indirectly retaken as a "they" about which one speaks or as an artificial non-linguistic subject such as "30% of the users of the healthcare unity think that…".

On the other side, the common sense (and also the common-sense-type-researcher) believes that the subject that states directly his or her opinion  is only the single speaker of the "I" or a limited speaker that speaks on behalf of the "we". These are seen as the only natural subjects of the discourse which express opinions given that for this common sense a subject that express his or her opinion is only speaking when there is "linguistic emission" (or a transcription of it) of a single "mouth" (even when it is about "we", it is only a single "mouth" that speaks).

Hence, once that there is no collective "mouth", a collectivity that express its opinions would not be able to speak directly, would only be talked about (through the meta-linguistic "mouth"), or would be reconstituted non-discursively, e.g., "30% of Brazilian men think that…".

This is the reason why one believes that empirically there isn't such collective speech stating opinions.

Now, this strict positivistic and "naturalistic" behavior needs to be overcome – what is not an easy task –, assuming that scientific and systematic treatment of the object "collective opinion" will require specific methodological constructions that allows one to keep the necessary binding with empirical reality. It is also necessary that the collective opinion can be artificially rebuilt (given that in this case it can only be artificial) as a qualitative object.

Besides, an "I" or "We" subject is also a subject of reconstituted opinion to the extent that one leaves aside the linguistic and psychological illusion that the natural home of opinion is individual consciousness.


The proposal of the DCS

The Discourse of the Collective Subject is an explicit proposal for reconstituting an empirical collective being or entity, stating his/her opinion as a subject of discourse spoken in the first person singular. 

What is the reason for making this choice?

Because the speaking social (speaking structure) or spoken social (structured structure) (Bourdieu, 1990) in individuals, in the first person singular, is the natural working regimen of opinions or social representations. In fact, opinions or social representations are efficient, i.e., work, precisely because individuals believe that these are their opinions, i.e., are generated in their brains.

Thus, DCS as this apparently paradoxical subject of discourse, once it is written in the first person singular though reporting a collective thought, is sociologically possible.

However, collectivity speaking in the first person singular does not only illustrate the regular working regimen of social representation but also is a resource to make feasible these very social representations as collective facts regarding qualitative collectivities (of discourse) and quantitative (of individuals). In fact, no one doubts that individuals share the same opinion(s), but when these very individuals state their opinions individually they only communicate a fraction of the content of a shared idea.

One has been attempting to reconstitute a collective subject in the DCS as a collective subject that is a collective person simultaneously speaking as if it were an individual, i.e., as a "natural" subject of discourse that conveys a representation of  amplified content.


Two Examples of the DCS

First Example

It is here presented a DCS elaborated as an exercise by students (teenagers between 16 and 20 years old), during the course offered by the School of Public Health of the University of São Paulo – Projeto Bolsa Trabalho: formação de pesquisadores juniores. [Convênio] PMSP/Secretaria do Trabalho/Unesco/Faculdade de Saúde Pública – USP, 2003.

It was proposed as one of the didactic activities of this course the realization of a research applying the DCS concerning the neighborhood where these students lived in, named Casa Verde. This research was carried out and its results were published in a specialized journal (Lefèvre et al., 2004). Here are some excerpts:

Research: the opinion of the dwellers of Casa Verde about violence against children.

Question: In your opinion, what makes  a parent  beat a child?

Category for the answer: alcohol and drugs.

Key-expressions of the answers:

Subject 5 - ...or if he takes any kind of drug, even being alcoholism.

Subject 9 - ...drug and alcohol.

Subject 12 - ...alcoholism and drugs alter parents  at home…

Subject 14 - ...when he arrives drunk at home or even high. 

Subject 19 - ...drugs, if they are addicted to.

Subject 20 - …when a parent has problems with alcohol and drugs. Then he becomes aggressive and beats his son…

Subject 1 – Alcoholism, drugs...

Subject 8 - ...father or mother that drinks alcohol and take drugs…

Subject 6 - …drinking alcohol, and also taking drugs…

Discourse of the Collective Subject

It is alcoholism, drugs.

When a father or a mother drinks alcohol or is addicted to it, or takes drugs and arrives at home drunk or even high, they become altered, becoming aggressive, beating their children.

One needs to notice that the DCS was composed in the first person singular, with key-expressions from statements of similar meaning, drawn from nine distinct individuals.

This collective person talks here as if it were a single individual, i.e., as a "natural" subject of discourse who nonetheless conveys a representation of various individuals, what allows the emergence of a collective opinion both qualitative and quantitative: qualitative because it is a discourse of amplified and diversified content, and quantitative because nine subjects have contributed to the construction of this DCS.

Second Example

The research reported here (Seragi et al., 2005) aimed at analyzing the current representation of some aspects of the Health Surveillance system by the inhabitants of the city of Águas de Lindóia (Brazil) in order to subsidize capacitating processes, education and development of technical personnel, as well as providing material for communication and marketing plans destined to bring closer services offered and the population.    

The research was carried out in the city of Águas de Lindóia

To realize the interview a semistructured script has been used. The sample was composed of sixty users of the three health care unities of the city: Unidade Básica de Saúde Alexandre Gatoline in the neighborhood Casas Populares; Unidade Básica de Saúde Bela Vista, in the neighborhood Bairro Bela Vista; Pronto Atendimento Municipal, in downtown. 

The research was made choosing at random in each unity a user older than 18 years old, in each working shift (morning and afternoon), totalizing six interviews a day during 10 days. The selected user was approached in the waiting room with a question asking if he/she would like to participate in the research. If the answer was positive he/she was then conducted to a private room previously selected, where the interviewer informed him/her about the mechanism and purpose of the research and fulfilled a form with information given by the user. In the form, interviewees were named in sequence from ÁGUAS 01 up to ÁGUAS 60. Then, the Term of Agreement was read and the user asked to sign it in case of agreement. After turning the voice recorder on, the interviewer began by naming the interview according to the name of the form (interview: Águas n), then asked the first question.

We report here only some qualitative and quantitative results of the question: a person buys food and notices that it is rotten. What could this person do?

The synthesis of central ideas to this question was as follows; the percentage of obtained answer is also given:



A – To make a complaint to the supplier. 20%

B – Give it back, change or be paid back by the supplier. 34.44%

C – Call and make a complaint to the Costumer Service of the supplier. 1.11%

D – Unspecified denounce. 10%

E – To call specific institutions (PROCON, VISA, Police Stations etc) and denounce the supplier.  24.44%

F - To get rid of the product, do not buy, inspect  personally 7.78%

G – To be not afraid to denounce 1.11%

H – Central idea excluded 1.11%

The DCS B – "Give it back, change or be paid back by the supplier" was the most shared idea between the interviewees. The resultant discourse is:

I think that he/she should go back to the supermarket and give it back because it is an abuse against the consumer to sell rotten things, and the supplier must be responsible for what it is selling: you are not going to consume rotten food nor lose your money.

The consumer must contact the owner of this shop and dialogue with him, give the product back and try to reach an agreement so that he/she takes the right measure because we want another product or to be paid back.

This has already happened with me, I went back to the market, complained and asked for another product, because I've paid for that. Why should I buy another rotten thing? Change, give it back and take another one!



The DCS and double representativity

It can be highlighted that the novelty presented by the DSC is the double representativity  - qualitative and quantitative – of collective opinions that emerge from the research: representativity is qualitative because in the researches using the DCS each distinct collective opinion is presented under the form of a discourse that reconstitutes distinct contents and arguments that matches the given opinion in the social scale; but representativity is also quantitative because such discourse has, furthermore,  a numerical expression (that indicates how many statements out of the total were necessary to compose each DCS), this means, statistical reliability, considering societies as collectivities of individuals.

Discoursive strata and infinite semiosis

The social representations expressed by the DCS need to be regarded in the perspective of percian semiotics (Peirce, 1975) as successive strata of discourses seen as interpretant signs based in a primary entity that we could call as the thought of collectivity.

The Discourses of Collective Subject shapes a panel of social representations under the form of discourses (as social empirical researches based in a series of methodological procedures) which seek to recover the collective thought in a less arbitrary way (Bourdieu & Passeron, 1970), in contrast with what has been happening in researches of this sort, be it qualitative or quantitative.

Evidently, the DCS does not intend to interpret social representation as infinite semiosis, neither work as "the last word" in what concerns these representations or its meanings and senses: it is only an interpretant sign (Peirce, 1975) that seeks out to reconstruct representations at a determinate level.

Henceforth, DCSs are not the social representations themselves, but only try to reconstitute a stratum out of them; upon this stratum another stratum can be directly added, constituted by one or many discourses or discursive formations or ideologies (Verón, 1980) in action in the DCSs.

The problem resides in defining the methodological procedures that can guarantee rigor and the standardization for these procedures in order to adequately recover this discursive stratum or interpretant sign.



AKIYAMA, R. Análise comparativa da intervenção fonoaudiológica na surdez: com a família ou com os pais? 2006. Dissertation (Master Degree) - Programa de Pós-Graduação em Ciências: Fisiopatologia Experimental, Faculdade de Medicina, Universidade de São Paulo, São Paulo.        

BANCO DE DSCs. Available at:  <>. Access in: May 2006.        

BOURDIEU, P.; PASSERON, J. C. La réproduction: éléments pour une théorie du système d'enseignement. Paris: Minuit, 1970.        

BOURDIEU, P. Coisas ditas. São Paulo: Brasiliense,1990.        

FOUCAULT, M. Microfísica do poder. 17.ed. Rio de Janeiro: Edições Graal, 1979.        

JODELET, D. Representations sociales: un domaine en expansion. In: Jodelet, D. (Org.). Répresentations sociales. Paris, PUF, 1989. p.31-61.        

LEFEVRE, F.; LEFEVRE, A.M.C. Depoimentos e discursos. Brasília: Liberlivro, 2005. 

LEFEVRE, F.; LEFEVRE, A. M. C. O discurso do sujeito coletivo: um novo enfoque em pesquisa qualitativa. Caxias do Sul: Educs, 2003. (Desdobramentos).        

LEFEVRE, F.; LEFEVRE, A. M. C.; FERNANDES, E.; CAZZUNI, D. H.; MEDEIROS, I. Y.; OLIVEIRA, N. G. S. Uma experiência de formação de pesquisadores juniores: discursos do sujeito coletivo sobre a violência contra a criança. Rev. Bras. Crescimento Desenvolv. Hum., v.14, n.1, p.76-89, 2004.        

LEFEVRE, F.; PEREIRA, I. M. T. B.; LEFEVRE, A. M. C.; STEWEN, G. T. M.; OLIVEIRA, N. G. S.; SUTOM, A. P. C.; MEDEIROS, I. Y. Criança: fumante passivo sem opção. Bol. Epidemiol. Paulista, v.8, p.5, 2004.        

MEDINA, A.G. Fórum de discussão na web e violência urbana: estudo de caso por meio da análise do Discurso do Sujeito Coletivo. 2005. Dissertation (Master Degree) - Faculdade de Saúde Pública, Universidade de São Paulo, São Paulo.

PEIRCE, C. S. Semiótica e filosofia. São Paulo: Cultrix/Edusp, 1975.

RIVERA, F.J. U.; ARTMAN, E.; FREITAS, C. M. Relatório cenário da pós-graduação strictu sensu. Rio de Janeiro: Escola Nacional de Saúde Pública, 2005.        

SOFTWARE QUALIQUANTISOFT. Availble at: <>. Accessed in: October 2005.        

VALVERDE, M. A. Cognições e atitudes de mulheres com excesso de peso e obesas: alimentação de controle de peso. 2006. Thesis (Doctorate) - Programa de Pós-Graduação Interunidades em Nutrição Humana Aplicada, Universidade de São Paulo, São Paulo

VERÓN, E. A produção do sentido. São Paulo: Cultrix/Edusp, 1980.