Chapter 6 - Objectivist vs Praxial Knowledge: Towards a Model of Situated Epistemologies and Narrative Identification

Summary

The final chapter revisits some of the tenets of the narrative paradigm, based on the analyses presented in the preceding three chapters, and suggests ways in which the concept of narrative rationality may be further developed and nuanced. Fisher distinguishes between objectivist knowledge and praxial knowledge and argues that it is the latter type of knowledge that narrative rationality seeks to ‘foster and support’. This distinction is used by the authors as a starting point to develop a model for situated epistemologies based on insights drawn from Fisher’s narrative paradigm, proposals put forward by some of his critics, the work of thinkers such as Heidegger and Kristeva, as well as more recent work on narrativity.

Throughout this book, we have argued that a turn to narrative – specifically, to a modified and extended version of Fisher’s narrative paradigm – can offer new insights into various phenomena that continue to hinder effective healthcare communication. This argument is developed against the backdrop of the growing hegemony of evidence-based medicine (EBM) since the turn of the century and the many challenges it has faced with the spread of Covid-19 since the end of 2019. In medicine and healthcare, the orthodox version of the EBM paradigm has generally contributed to promoting an understanding of evidence as a singular phenomenon that can be ranked on a fixed scale (the so called evidence pyramid; see Chapter 1 for details), with simple observational methods at the bottom and – moving towards the top – increasingly rigorous methodologies, notably randomized controlled trials (RCTs) and systematic reviews of such trials. The basic idea is that as long as allocations to the intervention group and the control group are double masked, RCTs are less likely to yield biased results than other types of research designs. Inherent in this assumption is the view that truth is universal and will eventually emerge once all sources of bias are eliminated. EBM researchers therefore invest in developing critical appraisal tools and checklists to evaluate whether research evidence can be considered valid, unbiased and reliable.¹ The dominance of EBM has been accompanied more recently by a growing tyranny of metrics in all areas of social life, including healthcare and health policy (Muller 2018) – especially in terms of modelling during the Covid-19 crisis. Alongside narrow understandings of evidence as defined by some of the most eager proponents of EBM, over-reliance on metrics and modelling has exacerbated an already problematic divide between traditional scientific rationality and people’s lived experience. This divide, we believe, is unsustainable. One way in which it can be bridged involves appealing to our innate capacity to make sense of happenings by embedding them within narratives we can assess and act upon. Without dismissing the importance and worth of the type of knowledge produced in scientific and medical laboratories, we would therefore agree with Pabst (2021:86) that ‘[t]ransformative policies’ must draw on the best available evidence but their success will ultimately ‘depend on the persuasive power of the underlying narrative’.

In making this assertion our intention is not to devalue rationality or scientific evidence. As we explain in more detail below, our argument is that rationality itself is born out of a prerational experience, and hence the epistemological standards by which science arrives at and assesses knowledge ‘are built on a foundation that they cannot themselves account for’ (Qvortrup and Nielsen 2019:157). In principle, at least in its original formulation, the narrative paradigm does not dismiss traditional rationality or the value of scientific evidence, although McGee and Nelson (1985), among others, have criticized Fisher for creating an unhelpful dichotomy by pitting the rational world paradigm against the narrative paradigm. Warnick (1987:175) argues that Fisher’s attitude changed over time, and that whatever his original intentions, his writings gradually implied a clear hierarchy between traditional and narrative rationality. She also criticizes Fisher for equating traditional rationality with one of its ‘lesser forms’, namely, ‘technical rationality possessed by experts who seek to close off discussion and exclude the public from making decisions on issues of social and moral concern’. While sharing Fisher’s ‘commitment to communities who reason through stories’, McGee and Nelson (1985:140) likewise insist that he paints a ‘misleading portrait of the place of experts in public affairs’. Be that as it may: our own revised version of the narrative paradigm treats narration ‘as a facet of rationality’ (Stache 2018:576). Rather than assessing a particular account of some aspect of the world on the basis of an alleged universal rationality, as the canonical EBM paradigm presupposes, the version of narrative theory we adopt in this book recognizes variation in the cultural, historical and social definitions of rationality and further suggests that we ultimately assess competing narratives of the same event on the basis of the values we believe each encodes. Similarly, but from a different angle, Stengers (2002) has argued that the struggle to define a universal rationality or evidence base beyond political differences is not only impossible but counterproductive. Evidence becomes evidence ‘not because it has been proven by empirical science … but because it has become a crossroads for heterogeneous practices, each with different interests, each of which has required the phenomena in question to be able to relate reliably to their questions and interests’ (Stengers (2002):1; our translation from French). According to Stengers, bias is therefore not necessarily a negative concept; indeed, it is a prerequisite for the production of evidence. Wieringa et al. (2018b:933) further argue that there is not one but at least two different forms of bias involved in evidence-based decisions:

When viewed from the perspective of the ideal limit theorem, bias is viewed negatively and unproductively as anything that distorts the comparisons between groups. Thus defined, bias can potentially be eliminated using technical procedures and checklists, but bias can also be defined in terms of a value-driven perspective on what is worth studying or taking into account. This kind of bias cannot be eliminated. It is unavoidable – and potentially productive and even necessary. Indeed, it could be argued that without bias, there would not be any truths at all.

In what follows, we take Fisher’s narrative paradigm as a point of departure, revisit its main weaknesses (including some we discussed in earlier chapters) and draw on a number of complementary theoretical strands to address its limitations. The aim is to outline a more inclusive and socially responsive model for assessing medical knowledge and dealing with sources of controversy around health issues such as Covid-19.

6.1 Limitations of Fisher’s Narrative Paradigm

Putting aside reservations about the rigidity of Fisher’s dichotomy and the version of traditional rationality he assumes, the various controversies analyzed in this book demonstrate how the logic of good reasons often clashes with the rational world logic of science; some of the more extreme versions of the latter claim to have universal validity independently of the way different people experience the world. We have seen, for instance, that many in the Black, Asian and minority ethnic community have been hesitant to wear a mask in public because of the racist fears it evokes and their negative experience with law-enforcement institutions in countries such as the United States (Chapter 3). Even vaccine hesitancy can be explained with recourse to the logic of good reasons rather than being dismissed by appeals to abstract, decontextualized traditional standards of rational thinking (Chapter 5).

At the same time, Fisher’s narrative paradigm is not without its more serious limitations, which have to be addressed in order to render it more productive and more amenable to being complemented with other approaches (see Section 6.2 below). One such limitation is that in asserting that stories not already familiar to and believed by an audience are unlikely to resonate with them, the narrative paradigm rests on an unhelpful tautology that fails to explain how we come to subscribe to certain narratives rather than others in the first place. The concept of narrative accrual, borrowed from Bruner (1991) and further expanded in Baker (2006), was introduced briefly in Chapter 3 as a corrective to this tautology. It suggests that we come to believe in certain narratives and the values they promote through repeated exposure to specific ways of making sense of the world. The values that underpin our decision making are not produced in a laboratory and are not arrived at by applying any logical formulae. They evolve through a long and complex process of socialization (i.e. of narrative accrual) that may span centuries and generations rather than merely years or decades, with powerful institutions such as the media, religious organizations, the family and educational systems playing a major role in this process. Narrative accrual normalizes certain accounts of the world and masks others from view. As Baker (2006:11) explains, the normalized accounts it sanctions eventually ‘come to be perceived as self-evident, benign, uncontestable and non-controversial’, however morally and practically untenable they may seem to those not socialized into the same set of narratives. This is borne out by the fact that earlier generations have largely seen little wrong with slavery, with the burning of those suspected of witchcraft or with policies and customs that marginalized women and castigated gays in ways that strike us as barbaric today. It is precisely this normalizing effect of narratives that requires us to complement the narrative paradigm with an approach capable of accommodating stories that ‘contest social reality’ (Baker 2006:163), that challenge rather than simply reinforce existing beliefs. This brings us to another, more serious limitation of the narrative paradigm as elaborated by Fisher.

Fisher’s tautology has a more serious flaw than failing to explain how we come to believe in specific stories. Its emphasis on resonance may imply that we can only entertain stories that reinforce our existing beliefs and values; if taken at face value, this would condemn us to live within the limits of our current moral imagination (Kirkwood 1992:34). As Morooka (2002) argues, by ‘appealing to the common sense of audiences’, that is, to their existing beliefs, ‘storytellers may degenerate into what Bourdieu calls doxsophers who do little more than reinforce the doxic submission to the social world’. In perpetuating or appearing to perpetuate the status quo, the narrative paradigm also fails to account for the dynamic movement of narratives as they evolve, multiply, splinter, are repeatedly contested and continually recast in all areas of social life. These dynamics can only be captured by attending to the tension between the normalizing, self-perpetuating aspect of narrativity and the simultaneous ability of stories to disclose the world in original ways (Sadler 2022). Any ethically responsible theory of narrative must be able to accommodate stories that challenge rather than reinforce our established beliefs and biases. Kirkwood (1992) thus calls for a rhetoric of possibility as a central component of moral argument, for acknowledging that rhetors have a responsibility not only to attend to (and reinforce) an audience’s existing beliefs but also to disclose new ways of understanding the world to them. It is possible to do so, we believe, by revisiting the interplay of fidelity and probability. The two dimensions of evaluation are conceived as mutually interdependent, which means that the decision of whether an experience has ‘truth qualities’ and rings true to the reader, whether it has fidelity, cannot be made independently of the internal logic of the story (its narrative probability).

Fidelity does not require the audience to actually share the experiences of protagonists such as Black populations and their reasons for a lack of trust in health authorities during the pandemic. It merely requires that these protagonists’ experiences appear to the audience to be ‘true to life – in principle’ (Fisher 1987:176). Hence our ability to empathize with characters in a film or novel, which merely requires that we can imagine ourselves in their position despite believing the story to be fictional. The story of Frankenstein can likewise be ‘true to life – in principle’ in the sense of accounting for experiences that seem real or credible ‘given the universe in which the characters live and the logic of their story’ (Fisher 1987:176). It is important in this context to note that Fisher’s notion of fidelity is based on a rhetorical concept of truth, as truthfulness in the eyes of an audience, meaning that the truth qualities of a story are understood to be a product of the rhetorical situation rather than of correspondence with an external reality. Hence, it is possible to acknowledge the truth qualities of a given story, provided it is coherent within its own universe, without accepting it as true in any objective sense. This tension between probability and fidelity, we believe, can be exploited to provide an opening for an audience to acknowledge the truth qualities of a new and unknown universe. In other words, the audience can be encouraged to imagine themselves as characters in a story and to accept that, had they been these characters, their experiences would probably have been similar (Fisher 1987). In this lies a possibility for stories to challenge our established world views and introduce an alternative universe. Although the fidelity of a story requires that it resonates with our experiences, a carefully crafted story can also move us to new and unexpected places. The tension between normalization and disclosure, emphasized by Sadler (2022), is thus potentially present in Fisher’s version of the narrative paradigm, contrary to what some of his critics have claimed.

The indirect implication in the narrative paradigm that effective stories ‘cannot and perhaps should not exceed people’s values and beliefs, whether or not these are admirable or accurate’ (Kirkwood 1992:30) has consequences for the way we approach medical communication and policy making. If taken at face value, it may suggest, for instance, that policy making could or should be reduced to adjusting stories to people’s existing beliefs rather than adjusting people’s beliefs to new, evolving stories. We reject such implications, whether or not they are warranted by or intended in Fisher’s approach to narrativity. Instead, we would reiterate that rhetors – including policy makers and those working in the field of healthcare communication – have a moral duty to expand the horizons of their audience beyond their current beliefs and values. This requires acknowledging that incoherence and contradiction, which are considered problematic in the narrative paradigm, can sometimes offer ‘potential entry points for novel ideas and values into the auditor’s belief system’ (Stroud 2002:387). Recognizing inconsistencies and contradictions as potentially productive and revealing of different ways of understanding an issue in turn requires more engagement with cultural variation than can be found in Fisher’s writings. As Stroud (2002:390, n4) argues, the emphasis on coherence and lack of contradiction in Fisher is itself a product of his focus on a Western context (including ‘modern American political rhetoric, modern American literary texts, and a Greek philosophical dialogue from Plato’) in which consistency is highly valued. In multivalent texts such as the Indian Avadhoota Gita and Devi Gita, by contrast, lack of consistency is not necessarily problematic: these narratives articulate contradictory value structures ‘in such a way as to force the audience to reconstruct how they interact with and what the text “means”’ (Stroud 2002:389). When connected to familiar notions, the confrontation with foreign narratives and values can trigger new insights and enable change. In such cases, it is ‘the auditor that rings true to new ideas and values within a foreign narrative’ (Stroud 2002:389). Stroud therefore suggests redefining narrative fidelity as ‘whether or not a story “rings true” with the values that an auditor holds or potentially could hold, given a coherent reconstruction of the narrative in question’ (Stroud 2002:389; emphasis added).

A related critique concerns some implications of Fisher’s assertion that narrative rationality ‘is a capacity we all have’ (Fisher 1984:9) and, more specifically, that ‘the people’ have a natural capacity to judge stories that are told for or about them. They can misjudge stories; they can be wrong, but so can experts and elites. The problem is that Fisher goes on to argue – following Aristotle – that ‘the people’ ‘have a natural tendency to prefer the true and the just’ (Fisher 1984:9). In other words, from a narrative paradigm perspective, we all ‘have a natural tendency to prefer the true and the just’ because we all possess the capacity of narrative rationality. But as Warnick (1987) contends, such assertions ignore the widespread success of Nazi propaganda and a host of other highly unjust and destructive narratives that plague our societies. Such stock political narratives (Bennett and Edelman 1985) persuade – they have resonance – precisely because they offer people attractive scapegoats that absolve them of responsibility for various social ills and allow them to maintain the ‘best conception’ of themselves and their immediate communities. In other words, they rank high on narrative fidelity. Warnick thus criticizes Fisher for acknowledging that ‘the people’ can be wrong but remaining silent ‘on the question of how they can avoid being deluded, given the absence of traditional rationality’ (Warnick 1987:177). Rowland (1987:272) similarly argues that traditional rationality need not be elitist; at the same time, ‘narrative modes of argument are not necessarily democratic. There is nothing inherent in storytelling that guarantees that the elites will not control a society’.

We suggest that some of the limitations in Fisher’s narrative paradigm can be addressed by acknowledging the importance of opening people’s minds to ‘creative possibilities’ that they may not be alert to, and by constructing narratives that ‘provoke intellectual struggle … and the creation of a more workable human order’ (Bennett and Edelman 1985:162; Baker 2006). To sensitize audiences to the self-perpetuating, conservative aspect of narrativity, it is important to enhance their critical skills; to encourage them to adopt a critical stance towards all narratives rather than accept dominant conceptions that circulate in their environment without scrutiny. This is, after all, the ultimate goal of education, especially at university level.

6.2 Revisiting and Extending the Narrative Paradigm

Public health is strongly linked to communication and persuasion, in that efforts to change behaviour are necessarily communicative acts. In order to design and communicate effective public health measures, we propose, health authorities must acknowledge and engage with stories like those we have documented in earlier chapters. The concerns of those who object to various restrictions such as wearing face masks or who are vaccine hesitant can only be addressed and contested by understanding and engaging with the logics of the stories to which they subscribe. Despite the limitations of the narrative paradigm as acknowledged above, and with the various caveats we have outlined to temper its basic dichotomy (traditional vs narrative rationality), our claim remains that public health discourse is too concerned with facts and not sufficiently concerned with stories. The crucial question for the success of health policy interventions is not only ‘what are the facts’ but ‘how do these facts make sense to people, and why’. This does not mean that establishing and communicating scientific facts is not essential to successful public health work. Rather, it means that we do not get anywhere with science unless it makes sense to people. Therefore, scientific facts need to be presented in a manner that either resonates with people’s current values and experiences or is capable of alerting them to new possibilities they can potentially make sense of and buy into. Facts cannot make sense in a vacuum: they only make sense as stories that reinforce or productively challenge the narratives that make up our existing moral universe.

Epistemologically, we may follow Fisher in distinguishing between information, knowledge and wisdom (Fisher 1995:172–173). Information, or what Fisher also refers to as ‘objectivist knowledge’ (Fisher 1995:173), is often linked to the idea of data as self-interpreting ‘facts’, in contrast to ‘knowledge’, which is assumed to have ‘semantic value’ and thus to require interpretation (Fisher 1995:173). Wisdom, finally, is about ‘knowing whether’ and is fundamentally concerned with values and ‘life as it ought to be lived’ (Fisher 1987:73). Facts are the cornerstone of the rational world paradigm, which proceeds by considering ‘whether the statements in a message that purport to be “facts” are indeed “facts”’ (Baker 2006:152). The narrative paradigm, on the other hand, considers all facts to be value-laden and assumes that assessing whatever is presented as fact always involves considering ‘the explicit or implicit values embedded in a message’ (Baker 2006:153). All facts then become knowledge that has to be interpreted and require wisdom to be evaluated and acted upon.

Writing in The Conversation in July 2021, Manuel León Urrutia draws attention to how Covid-19 data have proved to be complex and changeable. As an expert in data literacy, he reflects on how the visibility of data ‘has assumed a central role in determining the degree of society’s freedom since March 2020’ (Urrutia 2021). Highly specialist statistical jargon and data visualizations now pervade public discourse about the pandemic. But as the author argues, increased knowledge of specialized terms such as ‘flattening the curve’ do not necessarily contribute to better understanding, and even less to increased consensus about the need for various types of intervention. On the contrary, ‘this data deluge can contribute to the polarisation of public discourse’ rather than resolving controversies. Although data are ‘supposed to be objective and empirical’, Urrutia argues, they ‘assumed a political, subjective hue during the pandemic’. This is understandable given that people can only make sense of data by incorporating it into larger narratives of the pandemic. It means that rather than trying to resolve controversies by providing more data, which is the standard public health approach, health authorities need to engage more actively with people’s values and experiences – that means, with the stories that circulate in our communities.

As discussed in Chapter 2, Fisher stresses that while the philosophical ground of the rational world paradigm is epistemology, that of the narrative paradigm is ontology (Fisher 1987:65). Stroud also acknowledges the ontological nature of Fisher’s project, pointing out that narration, according to Fisher, ‘is fundamentally linked to the ontology and practices of human society’ (Stroud 2002:372). The narrative paradigm is concerned with the primary mode of being in the world, with the way in which we instinctively and pre-reflectively embed an experience within a story or the set of stories that constitute our world in order to make sense of it. To foreground the ontological grounds of the narrative paradigm, Qvortrup and Nielsen (2019) suggest exploring an implicit but less developed part of Fisher’s theory: the concept of dwelling. Fisher (Fisher 1987:94) refers specifically to this concept and to his indebtedness to Heidegger:

Particularly helpful to me is Heidegger’s view that ‘man is a thinking, that is, a mediating being’. This concept was put forth as an antithesis to the idea that ‘man’ is, or should be always, a ‘calculative thinker, a person who “computes”’ – weighs, measures and counts – possibilities, benefits and outcomes but does not ‘contemplate the meaning which reigns in everything that is’ … In another essay, Heidegger celebrates a line from a poem by Friedrich Hölderlin: ‘Poetically Man Dwells’. I would alter the line to read: ‘Narratively Persons Dwell’.

By introducing homo narrans as the root metaphor to describe the primary nature of human beings, Fisher suggests that ‘symbols are created and communicated ultimately as stories meant to give order to human experience and to induce others to dwell in them in order to establish ways of living in common’ (Fisher 1987:63; emphasis added). Stories are not merely modes of discourse or objects of inquiry but modes of living; we do not use narratives as we use an argument to support a predefined rational purpose. We live by and within stories in the sense that our rationality, our purposes and the arguments we use to support them are always already framed by and embedded in a narrative (or a range of narratives) within which they make sense. Fisher further insists that narration is not restricted to the mythical or fictional aspects of human communication. Similarly to Heidegger, he refutes interpretations of logos as ‘reason, judgement, concept, definition, ground’; these interpretations build on an epistemology that regards truth as a question of ‘accordance’ or ‘correspondence’ (Qvortrup and Nielsen 2019:147). Instead, Fisher evokes the ‘original conception of logos’, which he traces back to Isocrates, for whom logos was consubstantial with discourse. Discourse is not understood here simply as the form that an expression takes but is rather assumed to encompass ‘outward and inward thought’ as well as ‘reason, feeling and imagination’ – an understanding Fisher traces back to pre-Socratic times, when a clear distinction between logos and mythos had not yet been drawn (Fisher 1987:6). At that early stage, all communicative behaviour was deemed rational, though in a variety of different ways, suggesting that it is not only philosophical and technical discourses that exhibit logos, but rhetoric and poetics too (Fisher 1987:24). Fisher proposes a return to this early conception of logos and to treating narration not as distinct from but as a type of logic, a fundamental interpretation of the world that is articulated through all forms of discourse and inhabits our thinking.

Narration, then, is an expression of a ‘pre-thematic’ and pre-reflective relation to the world, in the sense that ‘access to reality is not to be established; it is always already established because the primary mode of being in the world is to engage with it or to dwell in it’ (Qvortrup and Nielsen 2019:147). The problem with the rational world paradigm is that it tends to reduce the ontological (ways of ‘being in the world’) to the ontic (‘being as brute facts’) and practical problems to scientific ones (Heidegger 2010; Sadler 2022). The overall aim of phenomenology, as outlined by Heidegger and adopted by Fisher, is to ‘establish a method that transcends what is known or given to modern man, science, or history of philosophy’ (Qvortrup and Nielsen 2019:146). As such, the narrative paradigm is an attempt to capture the ‘basic experience of the world of which science is the second-order expression’ and on which science is established (Merleau-Ponty 1962:ix; Qvortrup and Nielsen 2019:148). At the same time, and equally important, it is a response to the phenomenological call to ‘de-structure’ or deconstruct the history of ontology by making the fundamental structures of this tradition explicit. Fisher’s ambition, as we recall, was not solely to acknowledge the role of narratives in making sense of the world, but also to provide a framework that can explain how we assess narratives in order to decide whether or not we should adhere to them as a basis for belief and action (Fisher 1987). While Fisher presented his project as descriptive, he has been criticized for borrowing from the rational world paradigm when introducing narrative rationality as a normative standard. Conceptualizing the narrative paradigm from the perspective of narrative dwelling partly addresses this ambiguity by insisting on the fundamentally situated character of narrative rationality, which ‘follow[s] the internal flows of a given narrative toward its goals rather than a detached evaluation of its external traits’ (Qvortrup and Nielsen 2019:152). The truth qualities of a given story can never be assessed from a safe place outside and beyond the story itself, through reason as such, because reason always already dwells within a story. It follows that a story may be evaluated based not only on the situated principles defined within the story itself, but also with reference to the situated principles and values of the stories that its audience brings into the assessment. The latter may resonate or compete with the situated principles and values elaborated within the story being evaluated. Moreover, as Qvortrup and Nielsen (2019.:153) point out, while we ‘dwell narratively, we rarely do so alone’. Like Qvortrup and Nielsen, who argue that the legitimacy and relevance of a given narrative is contingent on communal dwelling rather than reason and argument, Sadler (2022) maintains that narrative understandings are not first produced by individuals and then shared by communities; instead, they are always produced ‘within an environment already structured by, and saturated with, other stories’ (Sadler 2022:19). Stories are thus communal dwelling places. In inviting others to inhabit their stories, individual members of a community create the ground for identification and conscientia. In this sense, narrative rationality makes it possible for us to ‘feel at home (dwell) in multiple stories’ (Qvortrup and Nielsen 2019:159), allowing us to entertain various possibilities and narratives ‘without being hindered by what constitutes a good argument’ (Qvortrup and Nielsen 2019:160).

Applying this extended version of the narrative paradigm to medical decision making implies a need to incorporate a situated epistemological approach into EBM, one that recognizes and explains different types of rationality, and hence plural conceptualizations of evidence. It also calls for acknowledging the pre-reflective and practical nature of any experience of truth. This need not be seen in a negative light, for as Qvortrup and Nielsen explain, ‘the experience of truth is tacit’ and constitutes ‘an opening that prompts engagement rather than a deterministic thought’ (Qvortrup and Nielsen 2019:158). Finally, it suggests that we would do well to know together and dwell together by exchanging ‘plots that are always in the process of re-creation rather than existing as settled scripts’ (Fisher 1987:18). In the next section, we will look at how some of these extensions to the narrative paradigm might be conceptualized through the notion of narrative identification (McClure 2009).

6.3 Narrative Identification in the Age of Fragmented Narratives

In proposing the concept of narrative identification, McClure attempts to expand the narrative paradigm to better account for the fragmented, intertextual and syncretic character of personal and social narratives of identity, subjectivity and ideology, as emphasized by poststructuralist thinkers (McClure 2009:193). As we have seen, Fisher’s narrative paradigm draws heavily on Kenneth Burke’s notion of identification, treating it as the operative principle of narrative rationality (Fisher 1987:66). However, Fisher’s intentions are undermined by the fact that the two concepts that are central to the narrative paradigm – probability and fidelity – are too dependent on ‘normative notions of rationality’ and too tied to the question of assessment ‘to be fully descriptive of narrativity in general, especially in light of poststructuralism’ (McClure 2009:193). By reducing identification to probability and fidelity, the narrative paradigm fails to account for how narratives interact, and how they may contain contradictory, unstable and implicit layers of meaning. As such, Fisher’s use of the concept of identification is based on a paradox: on the one hand, he explicitly develops his theoretical alternative to the rational world paradigm by drawing on the notion of identification. On the other hand, he reintroduces the rational world paradigm by narrowing the process of identification to probability and fidelity (McClure 2009). Although Fisher includes in his definition of narration all ‘symbolic actions – words and /or deeds – that have sequence and meaning for those who live, create and interpret them’ (Fisher 1987:58), he also indirectly limits the concept to discourses that can measure up to the normative criteria of the rational world. As we saw earlier, Stroud argues that this limitation must be addressed in order for the narrative paradigm to accommodate and account for non-Western narratives such as those elaborated in ancient Indian didactic texts, which are multivalent in nature and do not persuade through the kind of consistency or coherence specified in Fisher’s concept of narrative probablity (Stroud 2002:370). A similar argument can be made in terms of the narrative paradigm’s failure to explain the more general discursive shift towards fragmented narratives as a prominent feature of postmodern consumption, both within and outside the West (Sadler 2022). Firat and Dholakia (1998), for instance, argue that fragmented televisual marketing communication deliberately lacks a coherent story and instead relies on the use of images that are only meant to ‘leave the audience with a heightened sense of excitement about the product being marketed’ (Firat and Dholakia 1998:80). Similarly, Sadler demonstrates how fragmented narratives on social media complicate the assessment of coherence. To assess the coherence of a fragment such as a single Twitter post in isolation is meaningless; the same fragment, moreover, may be understood by different audiences as part of both coherent and incoherent narrative wholes (Sadler 2022:137).

McClure argues that rather than being defined by appeal to the normative rationalities that underpin the concepts of probability and fidelity, we need to acknowledge that identification ‘constitutes probability and fidelity’; that it is identification that ‘makes possible the symbolic processes by which probability and fidelity are constituted” (McClure 2009:195; emphasis added). He thus distinguishes between Fisher’s rationalistic understanding of identification and Burke’s original definition; the latter implies that rationality itself is a rhetorical act that is dependent on the use of symbols to create meaning, and hence that ‘all forms of rationality are composed via processes of identification’ (McClure 2009:198). This approach to narrative identification is not normative: it is intended as a descriptive framework for assessing narratives critically to explain how they deploy symbolic processes of identification to appeal to audiences and secure their adherence (McClure 2009:201). McClure further insists that narrative identification is not achieved simply by engaging with a single narrative but involves mediation between several narratives in an intertextual exchange, recalling studies of intertextuality that demonstrate how ‘multiple texts (narratives) intermingle in ways that are more akin to the processes of identification than traditional conceptions of narrative on which the narrative paradigm is constructed’ (McClure 2009:199). Julia Kristeva’s seminal work on intertextuality asserts that ‘any text is constructed as a mosaic of quotations; any text is the absorption and transformation of another’ (Kristeva 1980:66). A narrative therefore can never present a clear and stable meaning because it embodies societal conflicts and negotiations over meaning, in which utterances taken from various texts ‘intersect with one another and neutralize one another’ (Kristeva 1969:52; our translation). This suggests, too, that authors and receivers can never control the process of communication; they only contribute as mediators between recycled citations in an ongoing process of textual and intertextual productivity (Kristeva 1968). Narrative identification likewise evolves through a process of attending to an internarrative productivity rather than dwelling within a particular narrative.

McClure’s critical analysis of Young Earth Creationism offers a good demonstration of this process. Despite substantial scientific evidence to the contrary, Young Earth Creationists maintain that the Earth is no more than 10,000 years old. McClure argues that it is not possible to account for widespread adherence to this narrative without taking into account a whole range of interrelated Biblical narratives as well as other religious and social narratives that intersect with them (McClure 2009:205–206). Because the unity of the text and the autonomy of the subject are illusions, McClure argues that fidelity is not necessarily produced by the narrative itself ‘as if it was an isolatable attribute’ of it. Instead, fidelity is to be understood as ‘an act of constancy and personal attachment produced by agents to a collection of narratives’. Together these multiple narratives enable relations among members of an audience and produce ‘texts and subjects in a sticky swirl that creates and sustains a community’ (McClure 2009:207–208; emphasis added).

The Covid-19 controversies are similarly situated at the crossroads of multiple and conflicting stories. Rather than engaging in a detached, considered assessment of the coherence and fidelity of one specific narrative, we are continually negotiating our way through a multitude of narratives from a variety of medical and non-medical sources, often vascillating between conflicting accounts and reassessing their plausibility as we encounter new narratives. The purpose of the model of narrative analysis we have presented in this book is not to assist the reader in verifying a given story or stories. Instead, we hope that it will alert readers to the need to understand ‘the strains that make alternative narratives inevitable’ and encourage them to recognize ‘the diversity of human frustrations, aspirations, satisfactions, and imaginative constructions’ (Bennett and Edelman 1985:171).

6.4 A Final Note on Critical Appraisal in the EBM Model

To clarify the main argument we put forward in this book and guard against misunderstanding our claims, it is necessary to return briefly to the subject of critical appraisal in EBM. Burls (2009) offers a useful summary of the role of this process in EBM:

When critically appraising research, it is important to first look for biases in the study; that is, whether the findings of the study might be due to the way the study was designed and carried out, rather than reflecting the truth.

It is also important to remember that no study is perfect and free from bias; it is therefore necessary to systematically check that the researchers have done all they can to minimise bias.

Critical appraisal, as Burls’ definition makes clear, is principally conceived as a methodological endeavour. It is intended as a tool for evaluating whether a given study is designed and conducted in a way that reduces (rather than fully eliminates) bias. Although Burls acknowledges that no study can totally escape bias, and hence no study can capture the absolute truth, the very idea of minimizing bias implies that there is an objective truth out there waiting to be discovered. The ontological presupposition that there is a world of hard facts that we can collect with varying degrees of success is not questioned. The whole idea of critical appraisal is therefore embedded within a rational world paradigm in which the world is conceived as a set of facts and logical puzzles that can be solved through appropriate analysis and the application of reason. We see the narrative paradigm as adding an ontological dimension to the concept of critical appraisal by casting the world as a set of stories that must be chosen among rather than facts to be discovered. In doing so, we do not set out to challenge the idea of appraising evidence from a methodological or epistemological perspective. Our claim is only that such appraisal is incomplete. The question ‘What are the facts?’ must be supplemented with another one: ‘How do these facts make sense to people, and why?’. The latter is not about appraising the facts but about appraising the stories within which they are woven and acquire meaning.

Ultimately, we maintain, it is through narratives that knowledge about medical and other phenomena is communicated to others, enters the public space, and provokes discussion and disagreements. Importantly, effective narratives can enhance the reception of that knowledge and reduce some of the sources of resistance and misunderstanding that continue to plague public communication about important medical issues such as pandemics.