Narratives: Use in Qualitative Health-related Research

The bibliographic review followed a path through several chains of thought concerned with studying narratives. Some classical studies on narrative structure with origins within literature, history, communications theory and psychoanalysis were analyzed with the aim of exploring whether their categories and concepts would be methodologically applicable to qualitative health-related research. In the conclusions, the potential for using narratives to study situations in which there is interest in mediations between experience and language, between structure and events, between subjects and collective groups or between memory and political action are highlighted. These are questions that traditionally are of interest within Brazilian public health with regard to the fi eld of " Policy, Planning and Management " .


INTRODUCTION
Qualitative research has always occupied a prominent position within Brazilian public health studies, 20 considering the transformation of public health into collective health at the meeting point with the social sciences and their research methods.Previous studies observed that, while initially subservient to a structuralist bias, the hegemony of this reference point has been surmounted by means of studies on topics such as social and day-to-day representation. 1 In more recent published papers, 1,11,15 it has been seen that topics linked to subjectivity and narrative approaches have acquired greater importance, through anthropological and ethnographic studies. 5,19vertheless, few collective health studies have come close to narrative with regard to its aspects of narrative structure 13,16,21 and communication.However, it needs to be taken into account that such studies form part of the Brazilian healthcare scenario, under the auspices of the national health system (Sistema Único de Saúde, SUS) and the new healthcare services, in which new practices need to be evaluated and studied.Thus, traditional methodological approaches are insuffi cient for understanding the dilemmas and impasses of these new healthcare practices.
The objective of the present article was to comment on classical studies with narrative structure coming from the literature, history, communications theory and psychoanalysis, in order to investigate whether their categories and concepts would be methodologically applicable to qualitative health research.For this, the strategies of bibliographic review are differentiated from the normal ones that are used when the aim is to ascertain what recent production exists within a given fi eld.Thus, exhaustive searches in relation to recently published papers are important when seeking the state of the art within a given topic, but they may be inadequate as systematic reviews for supporting the broadening of our theoretical fi elds.
Here, it involves resorting to tradition 11 as suggested by Gadamer's hermeneutics.However, Gadamer 7 said that tradition always speaks in multiple voices.Therefore, attentiveness is required in order not to lose the sound of the weaker and more modest voices of history.How can this be done with modern search tools?In this search, the inclusion and exclusion criteria end up being delimited over the course of the study and not always at the outset, as would happen in a search for updates with the normal tools.The poet's adage: "caminante no hay camino, se hace camino al andar" (A.Machado) would be applied to the production of knowledge.
This investigation was started with the principal question of the need to set up studies to evaluate mental health services. 6Furthermore, the fi elds of activities of the researchers also infl uenced this starting point.In the case of the fi eld of "Policy, Planning and Management", the ethical-political motivation for producing syntheses that might be used to formulate recommendations for SUS initially stimulated a search for qualitative approaches capable of facing up to the challenge of interpretation that is always composed of two moments: analysis and construction. 6,14,18Moreover, within the possibilities for constructive delineation, the topic of narratives stands out.

NARRATIVE AS A MEDIATION CATEGORY
Ricoeur 17 situated the origins of narrative in Aristotle's Poetics and in the aporia of being and non-being of the time of St. Augustine (354-430 A.D.).According to Ricoeur, narrative would be a mediating operation between live experience and discourse.It would connect the explanation to comprehension: "overcoming the distance between comprehension and explanation" (p.11).
In the work of St. Augustine, Ricoeur investigated the aporia of experiences of time to draw attention to the temporal nature of human experience.In his view, speculation about time is an inconclusive rumination that only replicates narrative activity.Narration implies memory and prediction implies expectation.It is because of expectation in the present that future events appear as such.Therefore, the way to proceed is not to abolish temporality but to deepen it.The diversity of time should be admitted in order to do justice to human temporality: between distension and tension, the possibility of an intense time.
In his aporia of the experiences of time, St, Augustine rained some fundamental questions.The fi rst of these was the aporia of being or not being within time.He asked: "How can time be, if the past is no more, if the future is not here yet and if the present is not always here?" (p.23).Expectation is analogous to memory.In narration, it is always the language, experience and action articulated that will endure.Memory (past), attention (present), expectation (future) exist in people as an ever-present trio.
St. Augustine's second preoccupation was about measuring time.In his view, to pass by is to transit and the measurement of time lies within this: not the future that is not here, not the past that is no more and not the present that does not extend onwards...It is in this passage, in this transit, that the multiplicity of the present and its dilacerations needs to be sought.Augustine said that in the trio of the present there would be a distension of the spirit.
The third question presented now is that of intention and distension.If something is only measured while it is passing by, it can only be done when this fi nishes: for a measurable interval to exist, a start and end are needed.Now, what it important is no longer the passage (transire) but the permanence (manet): and thus the question of the spirit returns.According to Augustine, there would not be any future that diminished or past that grew if there were no spirit to perform this action.This is no longer a matter of imprinted images or anticipatory images but a matter of actions that shorten expectations and lengthen memories.Expectations and memories are extended: present intensions make the future pass into the past.The spirit carries out an intension but undergoes distension.This is a living metaphor that keeps together the ideas of passing by (ceasing) and making something pass by (transitioning).
Lastly, Augustine sets out the contrast between time and eternity.In eternity, nothing passes by.Eternity is taken to be the limiting idea of time.According to Ricoeur, this affi rmation has consequences relating to the possibility of "dechronologizing" the narrative, and the way out would be far from "logicizing" the narrative but, rather, to admit the diversity of time and do justice to human temporality, not by abolishing it but by deepening the levels of temporalization.Hence, there would be less distension and more intention.Thus, Ricoeur investigated the times of Augustine and Aristotle.In Poetics, he sought its forms and styles: epopee, tragedy, comedy and dithyramb.These are ways of arranging actions, but they differ with regard to the means (through which means?), object (what?) and manner (how?).The objects are always human actions.Their agents can be represented by showing them to be better, the same or worse then they are, i.e. tragedy or comedy.
With regard to Aristotle, Ricoeur highlighted the notion of weaving a plot or a web of intrigue or scheming (muthos).He designed a layout of events in a system; the art of composition.The activity that produced the plot would need to be superimposed in any structure.Ricoeur emphasized this function of composition as an operating system (i.e.something that makes it work).From Poetics, Ricoeur also emphasized the concept of mimetic activity: mimesis signifi es imitation or representation of an action or negotiation of events: the "what", "by which means" and how (the way in which the events pass by).This would always be praxeôs mimesis: representation of action.For this reason, it would always be about men who act.
The action is the construct from which the mimetic activity is constituted: the weaving of the plot.The relationship between muthos and mimesis leads to refl ections about the ethics, author and characters.Homer is lauded for self-effacement behind his characters: leaving them to speak in their own voices.However, in ethics, the subject always precedes the action in the order of moral qualities.In all narratives, there is a relationship between the characteristics of the character and the plot, but the representation is more of actions than of men.
For the entirety of the fi eld of narratives, some other questions are posed.Is this order susceptible to extension and transformation?All stories have a beginning, middle and end.Where should one start to tell the story?What is the starting point?According to Ricoeur, in his reading of Aristotle, the start was not the absence of antecedents, but the absence of any need for succession.The end came after other things either through necessity or through likelihood.The middle was a simple succession: it came after one thing and before another thing.This is an interesting argument for supporting periodization in study designs.
Ricoeur examined the conditions of validity and truth in narratives.In his view, to think of a causal link was to universalize.The internal connection would be the condition for universalization and its coherence would be sought through the plot.Composing a plot would make something intelligible come from accidental events: the necessary or likely elements of the episode.Mimetic activity would make up the action; the universal nature would not be seen but would arise.If it is understood that to invent is in reality to rediscover, a prospective conceptualization of the truth is therefore attained.
Ricoeur correlated these questions through his conceptualization of the mimetic trio.Mimesis I would defi ne the arrangement of the events (the "what"; the object) and actions, which would have purposes and motives, since to act is always an action with an interaction.This implies presupposition and transformation.To imitate or represent an action is to have prior comprehension of what will occur in the human action.The prior comprehension is shared by the author and reader.In Mimesis II, the author would understand the "as if", i.e. the mimetic creation: the metaphorical transposition from the practical fi eld to the muthos.This position is intermediate because it has a mediation function.Thus, in the events of the story, the relationships between tradition and innovation are left to be followed, from which the poíesis of the work should be sought.Finally, Mimesis III would designate the meeting between the text and the reader or spectator.The narrative has its full meaning when it is restored to the time of the action.This moment characterizes the application.There will be an intersection between the world of the text and the reader's world.
According to Ricoeur,17 the dialectics between aporia and poiesis consists of the relationship between time and narrative.Narratives would be no more than "stories that have not (yet) been told" (p.115).If an action can be narrated, it is symbolically mediated.If an action is symbolic, it is within the culture and therefore shared through a linkage with the public: an interaction.
This is what leads to the question of agents (individuals who act).Agents have characteristics containing ethical qualities.There are no actions that do not give rise to approval or reproof, in accordance with a series of values in which goodness and evil constitute the poles.For example, Ricoeur recalled the pity that is felt for some characters because of their undeserved misfortune, which he called esthetic pleasure associated with empathy.Thus, he highlighted a trait that is inherent to action: it can never be ethically neutral.From this reference point, narratives would be mediations between action and language.
In studies on historical narrative, Burke 2 saw narrative as mediation between structure and events.In his opinion, all histories represented particular points of views."Contrary to most novelists, historians do not intend to read the minds of their heroes, but just their letters" (p.335).Narrators of history would need to fi nd a way of making themselves visible: by declaring who they were and what their points of view were (like collective health researchers).This would be an ethical conditioning factor and it would have consequences for knowledge production.Methodologically, Burke proposed to "densify" the narrative, i.e. to construct narratives that could be read not only with the sequence of events and the conscious intentions of the players, but also with the structures (which he translated as institutions and ways of thinking), thereby arguing whether they delayed or accelerated the events.This would be a way of mediating structures and events.Furthermore, there would be the possibility of exploring the narrative in various ways: micro-narratives, back-to-front narratives (inversion of the chronological order) and narratives that present the same events from multiple points of view.
In a way resembling what is done in qualitative evaluative health research with stakeholders, Burke drew attention to multiple voices (heteroglossia).This was tried in a recent study in which focus groups were transformed into narrative focus groups. 22This approach is also of interest for studies on SUS, because it would make it possible to explore the relationships between the structure and observed and/or recorded events in health services, thereby escaping from the undesirable dichotomy between macropolicies and micropolicies in such studies.
Going now into the fi eld of communications, other views can be mapped out.Guimarães 8 and Leal 10 positioned narrative as a mediation between the individual and society: a porous interlocution device.In situations of subjects that are communicating, there is always a relational bias, produced through the action of affecting and being affected by the other subject in the narrative mediation.Thus, according to these authors, narratives offer resources for dealing with events (whether they are small or large, or intense or insignifi cant) that emerge in the (only apparent) repetition of experiences of day-to-day life.
"Without neglecting the strength of structures and the presence of power, but also without being unaware of the creative power of man and the strength of establishment within narratives (...) that men invent and experiment with, we seek another type of logic with which to ascertain the complex nature of social life -not the logic of determination but the logic of mobility" (Guimarães, 2006: 8).
Knowledge of day-to-day life and communication experiences requires attention to narratives.However, such narratives are not "given" factors: they need to be devised through an "eye for creating narratives" 10 that establishes links between the different fragments that are circulating.This eye for creating narratives gives rise to the everyday forms of articulation.This would a preferred method for studying new practices.Attention needs to be given to how this collocation approximates to Ricoeur's, with regard to narratives as stories that have not (yet) been told and which are symbolically mediated.Narratives emerge as a result of interrelations between social forces and they characterize possible balances between historical and social fl ows.
Since narrative is language, it should fi t within the textual economy as dialogue with other texts and as a set of sociohistorical characteristics that locate a context.Narrative is open to interpretation at the same time as conditions for its circulation, reception and production are established.Thus, it links relationships of power, policies and identities within the context that are perceived both diachronically and synchronically, thus denoting a complex relationship between narratives and social discourse.In the relationship between text, narrative and discourse, the conditions for the introduction and circulation of social utterances, ideologies and the realities of day-to-day life can be seen.
Lyotard 12 also indicated some fundamental functions of narrative forms.Narratives contain positive or negative formations describing successes or failures of the heroes that give legitimacy to institutions, thereby defi ning criteria for competence.In the evaluative research conducted by the present authors, in which several focus groups of professionals within the new mental health services were constituted, narrative constructions were demonstrated.In these, systematics for how these services function could be identifi ed, in which the agents themselves pointed out what should be considered successes or failures.This research experience shows yet another question indicated by Lyotard, among the functions of narrative forms: the question of accepting that there is a plurality of language plays, 12 with several types of enunciation.This mixes the reported competences into a dense fabric, from the perspective of the entirety.In dealing with social practices in the way that health service practices function, this potential seems to be of fundamental importance.Furthermore, Lyotard deals with another property of narrative forms relative to the means of transmission that gives them a pragmatic note: narrators only have competence through having been a listener, and the person now receiving the narrative becomes elevated to the same authority.The "narrative positions" are distributed in such a way that there is compliance with the sender's position, since this sender was previously a receiver and was placed as a diegetic reference particularly because of bearing a name: "(...) the tradition of the reports is at the same time the tradition of the criteria that defi ne a trio of competences: knowing how to say, listen and do something (...) What is transmitted with the reports is the group of pragmatic rules that constitutes the social link" (Lyotard 12 p.39).
Narratives determine the criteria of competence and/or illustrate their application.They defi ne what can be said by right or can be done within the culture and, since they are part of this culture, they are legitimized.According to Lyotard, the reports are language games articulated in a narrative manner and are the minimum relationship required for their to be society, given that even before humans are born, they are already placed as reference points for stories told by people that surround them.This positioning of humans within the universe of history from the earliest times now makes it possible to come to the work of Kristeva, 9 a psychoanalyst and linguist.In a very careful reading of Arendt, Kristeva stated that life was a narrative and would be specifi cally human under the condition that it could be represented by a narrative and shared with other men.Thus, a life full of events that can be narrated becomes a biography.From Arendt's work, Kristeva highlighted the possibility of a bios life versus the possibility of a zoe life and took the view that a bios life was práxis (Aristotle).Kristeva believed that Arendt, unlike Ricoeur, would undertake rehabilitation of praxis more than poetics.This would be because only action as narration and narration as action would characterize this "bio" life that is specifi cally human.Thus, this type of life would weld the relationships between life, narrative and politics, since life in the polis is always mediated by words.Narrative, the ability to enunciate a biography, becomes as necessary as it is problematic, since, no matter how brilliant a narrative may be, it would not be able to save a life.According to Arendt, narration tells the story but action wins if it is a narrated action.
In distinguishing poíesis as a technically produced activity from práxis as an activity of social action, narrative would be close to the latter.Action in the pólis would not be a "fabrication" in the sense of being constituted by activities aimed towards certain purposes, but acts full of signifi cance.Furthermore, the place of action is always the polis.In Kristeva's view, the pólis is the place of interest, i.e. a place between two people.Narrative would thus constitute a political model grounded in actions and words, but never one without the other.In surveys on SUS and public policies and/or health services, this lack of dissociation of the discourse of action takes on fundamental value, and would be an important contribution towards escaping from the dichotomy between discourse and practice.
However, how could the word "poetics" be manifested in the pólis to reveal the virtue of its heroes?To answer this question, Kristeva analyzed the relationship between phronesis (practical knowledge) and sophia (theoretical knowledge).Instead of asking, "what do you know?", it would be necessary to ask, "who are you?"(which would mean "what do you do?" and "what do you say?").The relationship between true history and recounted history would also have to be questioned.The existence of interest, the generator of memory and witnesses, draws attention to the distancing of what is experienced ex-post facto; the thinking can be divided by means of a plot.The memory of the action makes it specifi c action.From Kristeva's reading of Arendt, his view was that what was essential was not the internal coherence of the narrative (as Ricoeur's reading of Aristotle indicated) but, rather, to identify the agent of the story.In this approach, the art of narrative would consist of condensing an action into an example interval, extract it from the continuous fl ow and reveal the subject of the action.
In the case of the research produced by the present authors, six narrative focus groups were initially designed, formed by workers from each of the Psychosocial Care Centers, with varying composition of professionals, and another four groups composed respectively of users, family members, local managers and administrators (Figure).Following the second stage of the focus groups, the possibility was highlighted that new voices and certain destabilizations of the recently (but not less) instituted way of working the equipment were among the voices that in the original design were weakest: nursing auxiliaries, users and family members.
The revelation of this "who", of which Kristeva spoke by means of Arendt, allowed new openings to be added to the original design: a focus group just for auxiliaries and greater participation (in numbers) by family members and users in the fi nal workshops for agreeing on indicators.Furthermore, the presentation of narratives, which were constructed based on the material from the narrative focus groups, in a second round of focus groups (given the name hermeneutic groups, because they would have the task of interpreting and legitimizing the narratives) not only allowed questions of the legitimacy of the narrations to be addressed, but also made it possible to work on the "narrative capability effects".The groups confi rmed the arguments, deepened the discussions and expressed changes in relation to the initial situation.The narratives themselves became actions, since: "it is through narrative and not through language in itself that political thought is achieved" (Kristeva 9 p. 87).Thus, all narrative would be political narrative.
It is acknowledged that this review has limitations, such as its non-exhaustive nature.Nor was it intended to be exclusive, in the sense of not accepting other readings.Even less so was the aim to make an extensive survey of the most recent published papers in the methodological fi eld.Nonetheless, it was sought to undertake a systematic review, indicating possible applications of certain concepts and theoretical bases that, starting from a common origin with the social sciences, would make it possible to delimit a particular fi eld for research within "policy, planning and management" in the context of Brazilian collective health studies. 14

FINAL CONSIDERATIONS
In the present review, it was possible to compare how narratives might be understood, coinciding with the various currents reviewed as mediations: according to Ricoeur's literary critique, as mediation between discourse and action; according to the historiographic currents, as mediation between events and structures; according to communication currents, as mediation between individuals and society; and according to Kristeva's psychoanalysis, as mediation between memory and political action.
The emerging problems posed within SUS itself will be put into contact with and will be considered by the fi eld of collective health, with its history marked by controversy between structuralism and phenomenology and recent restoration as a subject fi eld. 1,4Large numbers of new services like Psychosocial Care Centers, Family Health Units, Reference Centers for sexually transmitted diseases (STDs) and others face the day-to-day need to recreate their practices and give them social basis and legitimacy.Would it not be of interest to have methodological tools to research the mediations between what is said and what is done (discourse and action); or between occasional events and more structured questions; or between individual and collective subjects?Furthermore, is there not a pressing need for a text to be brought to the world, in relation to these new practices?What happens in homes during household visits?How can a team be defended in proximity with madness?
Narratives, outside of their scheme of traditional linkage to ethnographic studies, in which they take on a more descriptive and chronologically arranged nature, may contribute towards qualitative collective health research if their capacity as a porous communication device is explored; if they can be "densifi ed"; and if new eyes for creating narratives are brought to bear.Nonetheless, it would be necessary to fi t them within participative research strategies, involving the specifi c players in various resumptions of their narrative (in the manner of circular hermeneutics 7 ), and to provide contact with narrations in other groups of interest, as was attempted by the authors of the present review.In this way, the narrations would themselves be transformed, as stated by Kristeva, into political action.Such research would meet some of the ethical-political imperatives, by intervening in the universe investigated and providing strengthening of their own agents.
This would be an invitation to experiment with a type of research intervention in which the directions and pathways would not be designed at the outset by the researchers.It is a strategy to be explored that might be considered to be a little more open to new happenings within the world of life.

Figure
Figure.Research fl ow diagram.