Friday 10 July 2015

Recovery, paternalism and narrative understanding in mental healthcare

I have cheated and replaced the first version with a second version a week later.

Recovery, paternalism and narrative understanding in mental healthcare

Abstract
There has been a growing emphasis on the idea that recovery in mental healthcare should not be seen as a matter of getting better but instead of successfully living a flourishing life as conceived by the subject herself. Theorists of recovery also stress the importance of narrative understanding for articulating the sort of life that would count as recovery. But surely one of the threats of mental illness is that it can undermine a subject’s autonomy and hence capacity to author a suitable narrative for recovery? Addressing this worry raises the threat of paternalism. Having sketched an abstract model of recovery and compared it to the capabilities model of Sen and Nussbaum, this chapter outlines a minimal account of narrative understanding drawn from Peter Goldie’s book The Mess Inside can address the worry of paternalism.

Introduction: Recovery as the goal of mental healthcare.
In this first section, I introduce recovery as the goal of mental healthcare and sketch an abstract model for it. Recovery aims at a value-laden and person specific conception of flourishing. In the second section, I show how the capabilities approach of Amartya Sen and Martha Nussbaum fits this abstract model but that the difference between Nussbaum’s and Sen’s versions reflects a corresponding difference between substantive and procedural accounts of personal autonomy. This difference is also present in Davidson and Hopper’s more specific claims about the possibilities for recovery from mental illness and leads to a challenge to the recovery model. If mental illness can compromise autonomy and calls for sensitive clinical intervention to recover it, does this not risk the paternalist imposition of others’ values? In the final two sections I argue that a narrative view of a sense of self can address this on either opposed broad view of recovery and autonomy.
Recovery has come to be promoted as a novel and desirable target for mental healthcare. It has become a proud boast that mental healthcare is recovery orientated. Nevertheless, whilst there is agreement that in this context, it does not mean merely getting better or returning to a previous state of health, there remains disagreement as to what recovery is.
There is an increasing global commitment to recovery as the expectation for people with mental illness. There remains, however, little consensus on what recovery means in relation to mental illness. [Davidson and Roe 2007: 450]
The term ‘recovery’ appears to have a simple and self-evident meaning, but within the recovery literature it has been variously used to mean an approach, a model, a philosophy, a paradigm, a movement, a vision and, sceptically, a myth. [Roberts and Wolfson 2004: 38]
In the UK, a policy paper published by the Sainsbury Centre for Mental Health, titled ‘Making recovery a reality’, begins by summarising some key points of emphasis which, it is suggested, characterise any broadly conceived recovery-based approach. These points include:
Recovery is about building a meaningful and satisfying life, as defined by the person themselves, whether or not there are ongoing or recurring symptoms or problems.
Recovery represents a movement away from pathology, illness and symptoms to health, strengths and wellness.
Hope is central to recovery and can be enhanced by each person seeing how they can have more active control over their lives (‘agency’) and by seeing how others have found a way forward.
Self-management is encouraged and facilitated. The processes of self-management are similar, but what works may be very different for each individual. No ‘one size fits all’.
The helping relationship between clinicians and patients moves away from being expert / patient to being ‘coaches’ or ‘partners’ on a journey of discovery. Clinicians are there to be “on tap, not on top”.
People do not recover in isolation. Recovery is closely associated with social inclusion and being able to take on meaningful and satisfying social roles within local communities, rather than in segregated services.
Recovery is about discovering – or re-discovering – a sense of personal identity, separate from illness or disability. [Shepherd, Boardman and Slade 2008: 0]
The Scottish Recovery Network summarises its views of recovery in similar terms:
Recovery is about living a satisfying and fulfilling life.
Recovery is about more than the absence of the symptoms of illness. Some people describe themselves as being in recovery whilst still experiencing symptoms.
There can be lots of ups and downs during the recovery process – some people describe it as a journey.
For this reason people often talk about being in recovery rather than recovered.
Some people consider recovery as being ‘back to the way things were’ or back to ‘normal’ but for others recovery is more about discovering a new life or a new way of being. [Brown and Kandirikirira 2007: 3]
These lists provide a starting point for setting out a theoretical model of recovery. But there is a further structural constraint. To articulate a recovery model which is distinct from or contrasts with, for example, a bio-medical model of mental healthcare, it is not enough to say that recovery (construed in some broad way) is a desirable aim of mental health care. One could hold that whilst holding a broadly bio-medical view of health and illness: for example, as pertaining to biological function versus dysfunction. To count as a distinct model of healthcare, it must offer more than just a broad aim but rather a theoretical conception of what illness, or health, or something like health is. (In the UK, the rise of the recovery movement coincided with both greater optimism within biological psychiatry of the efficacy of medicines but also the kind of theoretical articulation of a novel view of recovery with which I am here concerned. Both elements played a role, complicating the historical story.)
The characterisations of recovery above suggest the importance of two distinctions. First, there is a distinction of focus between pathology and whatever is its relevant contrast, perhaps health. Second, there is the distinction between what is evaluative or normative and what is merely plainly factual. Together these can be used to sketch a distinct although abstract conception of recovery. I will take each in turn.
The philosophy of psychiatry, and more generally the philosophy of medicine, has tended to focus on the illness end of a spectrum between health and illness. The key concern has been with the notion of illness (or disease or disorder). One reason for that has been the origin of the debate, at least within the philosophy of psychiatry, in the response to Thomas Szasz’ argument that mental illness is a myth [Szasz 1972]. Szasz’ claim that mental illness is an oxymoron prompted responses by biologically minded psychiatrists and researchers who attempted to devise models of illness (or disease or disorder) which accommodated not only physical but also mental illness [Kendell 1975, Boorse 1975]. That in turn has led to an ongoing debate focusing squarely on ill health [Fulford 1989, Pickering 2006, Wakefield 1999]. There was no equivalent Szaszian argument for the mythic status of mental health and hence no incentive for a philosophical defence of that notion.
Against a background focus on the nature of illness, recovery can seem to be simply a return from, or a removal of, that status. Whatever illness or disease is, recovery is its negation. By contrast, concentration first on the health end of the spectrum is at the heart of the recovery approach. This accords with the claim quoted from the document ‘Making recovery a reality’ above that ‘Recovery represents a movement away from pathology, illness and symptoms to health, strengths and wellness’. A recovery model has to do more than just take the aim of healthcare to be the removal of illness.
The second distinction is between views of mental illness in particular, or illness more generally, as necessarily evaluative or merely descriptive. Does the analysis of mental illness contain reference to values or not? Some philosophers and psychiatrists argue that at the heart of the idea of illness is something that is either bad or wrong for a sufferer or is a deviation from a social or moral norm. Both of these are evaluative or normative notions and hence both are views of illness as value-laden.
Others argue that illness is a plainly factual matter. Typically, they argue that illness involves a failure of a biological function and that function – and hence deviation from, or failure of, function – is a plainly factual, biological (and/or psychological) term couched in evolutionary theory.
Having sketched the two distinctions, I suggest that a clue to articulating a recovery model which genuinely contrasts with a medical model is to locate it on the health-focused rather than pathology-focused side of the first distinction and on the values-laden or normative side of the second.
That remains just a clue: more conceptual work has to be done. But it might be objected that the first of these two distinctions is unnecessary; we can articulate a genuine contrast to a bio-medical model simply by using the second distinction.
The idea is that a bio-medical model construes mental illness as value-free, as reducible to plain facts about biological (and/or psychological) function. Adopting the opposite view – that illness is a necessarily evaluative notion – stands in genuine contrast. And a conception of recovery in relation to illness, so construed, might be enough to count as a recovery model because of that genuine, substantive contrast.
Whilst such a position is a genuine contrast to a plausible candidate for a bio-medical model of illness, it does not seem to capture an important element of the recovery approach: a particular goal or aim of therapy which is not defined merely as the absence of illness. It is instead captured in specific terms such as a hope, autonomy and social inclusion.
One specific problem is that even if one thinks of illness as a necessarily evaluative notion, this is not a sufficient reason to think that health is. It might be the mere absence of an evaluatively identified illness state. Health might be conceptualized in merely statistically normal (rather than normative or evaluative) terms, perhaps as the state of most people, or, alternatively, the state one was previously in. If so, whilst the states that individuals have an interest in recovering from are those with particular normative or evaluative properties (whatever those are), recovery itself might be characterized in non-normative non-evaluative terms. And that does not seem to fit the way the recovery approach is characterized by its supporters.
What of the other distinction? Could a recovery model be defined simply as one which focuses on a positive conception of health, or something like it, rather than merely the absence of pathology, however construed? Again, no. As the authors of ‘Making recovery a reality’ make plain, they have a very particular conception of the aim of recovery tied to a conception of hoped for and autonomous life connected to social inclusion. That specific content is not captured merely by a focus on health which, as I have just argued, might be thought of in statistical normal terms.
To capture what is characteristic of a recovery approach in order to frame a recovery model, it seems that both aspects are needed: a) a focus on a conception of health, or something like it, and b) in normative or evaluative terms. A recovery model is thus one which construes the positive aim of mental healthcare to be a state of health or something like it, necessarily characterised in normative or evaluative terms.

The capabilities approach to recovery
The model sketched above abstracts from typical comments made about the nature of recovery in mental healthcare in practical policy documents. The idea that recovery aims at a value-laden conception of a flourishing life, which is tailored to the individual concerned, fits such documents. But that is not to say that that is the most that might be offered by way of theoretical articulation of recovery. More can, and has been, said which is, nevertheless, consistent with the abstract model just sketched. In this section, I will outline an influential theoretical framework for thinking about recovery: the capabilities approach. I will use this to extract a potential tension in thinking about the value-ladenness and person-centeredness of recovery and thus flag the role, in the next section for narrative.
The capabilities approach to recovery in mental healthcare is based on Amartya Sen’s model for welfare economics developed in the 1980s. Rather than focusing on the fair distribution of resources or primary goods, Sen proposes that the focus of welfare economics should be on a fair distribution of the capacity to lead a flourishing life. In this context, capability is a measure of the ability to do the things and to be the ways that amount to a flourishing form of life. Sen uses the word ‘functionings’ for this latter idea.
The expression [‘capability’] was picked to represent the alternative combinations of things a person is able to do or be-the various ‘functionings’ he or she can achieve. The capability approach to a person’s advantage is concerned with evaluating it in terms of his or her actual ability to achieve various valuable functionings as a part of living… Functionings represent parts of the state of a person-in particular the various things that he or she manages to do or be in leading a life. The capability of a person reflects the alternative combinations of functionings the person can achieve, and from which he or she can choose one collection… Some functionings are very elementary, such as being adequately nourished, being in good health, etc., and these may be strongly valued by all, for obvious reasons. Others may be more complex, but still widely valued, such as achieving self-respect or being socially integrated. Individuals may, however, differ a good deal from each other in the weights they attach to these different functionings-valuable though they may all be-and the assessment of individual and social advantages must be alive to these variations. [Sen 1993 :31]
Applied to welfare economics, the approach takes account of the fact that people can differ in the resources they need to achieve valuable ways of being and acting. Hence it delivers different results from simply advocating equal distributions of resources or primary goods.
In characterising capability, Sen stresses the role of freedom. The freedom to live different kinds of life is reflected in a person’s set of capabilities. Freedom itself adds value to a life. Even the existence of possibilities not adopted or embraced add, via a sense of freedom, to the value of a life. But the relevant sense of freedom does not range over just any possible life. Having the option to live kinds of life that an individual would never consider is not freedom in Sen’s sense. Genuine freedom has to be assessed relative to what a subject values. Further, what is valued changes what counts as ‘functioning’. Fasting and starving both involve the deprivation of food but because the former is chosen it counts as functioning.
This sensitivity of what counts as capability fits the value-ladenness and person-centredness of the abstract model of recovery outlined in the previous section of this chapter. It also fits the focus not on pathology but on health or flourishing. The aim of recovery, on such an understanding, is to maximise the capability of a person to achieve various valuable functionings as a part of their life. It thus connects the abstract requirements on a recovery model to some pre-existing philosophical and economic thinking. But this connection also highlights a tension in, or challenge for, the recovery model which, in the next two sections, will be connected to the role of narrative.
Despite the importance of freedom there is an important tension in thinking about capability because of a second influence on its initial articulation: the Aristotelian philosopher Martha Nussbaum. Nussbaum draws on Aristotle’s account of flourishing to argue that there is a list of basic human functions that applies to all human beings [Nussbaum 1988: 176]. Not just anything could count as human functioning. Thus drawing on Aristotle, Nussbaum argues for universal standards for assessing human capability. This contrasts with Sen’s liberal or relativist view that with freedom comes proper diversity of choice. This forms the basis in the The Tanner Lectures on Human Values of one of his criticisms of traditional welfare economics based on the fair distribution of resources or primary goods [Sen 1980].
In the co-authored introduction to a collection of papers on the capabilities approach, Nussbaum and Sen jointly set out the difficulties that apply to the choice between pressing a universal or a culturally relative view of human capabilities. It is worth quoting this passage at length.
Should we, for example, look to the local traditions of the country or region with which we are concerned, asking what these traditions have regarded as most essential to thriving, or should we, instead, seek some more universal account of good human living, assessing the various local traditions against it? This question needs to be approached with considerable sensitivity, and there appear to be serious problems whichever route we take. If we stick to local traditions, this seems to have the advantage of giving us something definite to point to and a clear way of knowing what we want to know... It seems, as well, to promise the advantage of respect for difference: instead of telling people in distant parts of the world what they ought to do and to be, the choice is left to them. On the other hand, most traditions contain elements of injustice and oppression, often deeply rooted; and it is frequently hard to find a basis for criticism of these inequities without thinking about human functioning in a more critical and universal way…The search for a universally applicable account of the quality of human life has, on its side, the promise of a greater power to stand up for the lives of those whom tradition has oppressed or marginalized. But it faces the epistemological difficulty of grounding such an account in an adequate way, saying where the norms come from and how they can be known to be the best. It faces, too, the ethical danger of paternalism, for it is obvious that all too often such accounts have been insensitive to much that is of worth and value in the lives of people in other parts of the world and have served as an excuse for not looking very deeply into these lives. [Nussbaum and Sen 1993: 4]
Uncritical relativism versus paternalism is an apparent rather than actual dilemma, however. Each is the criticism that someone taking the corresponding opposed view might make. Universalists fear uncritical relativism but are in turn accused of paternalism by their opponents who emphasise diversity. To see this it is worth flagging corresponding opposed views of autonomy since paternalism is the trumping of another's autonomy.
The idea that universal standards of human flourishing necessarily threatens paternalism presupposes that autonomy really is autonomous of external standards. One such view is proceduralism equates base autonomy with the capacity of a subject to reflect on and endorse, at a second order level, their first order actions and values [eg Frankfurt 1971]. But it remains neutral as to what those first order actions and values are. Substantive approaches, by contrast, argue that the notion of autonomy involves an ability to be guided by the good and the true. And hence a specification of what it is to be autonomous cannot avoid substantive claims about human flourishing [eg Wolf 1990].
Clearly if the very idea of autonomy presupposes tracking some universal standards, then the mere existence of such standards cannot threaten the ethical danger of paternalism, the unwarranted undermining of autonomy. If, on the other hand, one thinks that autonomy is merely a matter of reflective coherence, then the absence of universal standards is not a threat of uncritical relativism, it is just that criticism is a matter of local coherence.
This opposition within accounts of autonomy and versions of a capabilities approach to welfare economics also has an echo in its application to the recovery model in mental healthcare. In their paper ‘A Capabilities Approach to Mental Health Transformation: A Conceptual Framework for the Recovery Era’, Larry Davidson and colleagues follow Sen in stressing the connection between a capabilities approach to recovery and the proper diversity of choices that will be made:
It is in the very nature of choice to result in variability, otherwise choice would not really be free but would refer only to changes in the quantity of some basic universal. While smoked salmon and french fries are, in fact, both foods, to say that a person who prefers smoked salmon to french fries has no real preference because they are both foods is to miss the point of having preferences to begin with. It is to gloss over the issue of choice, but this is precisely where our primary interest lies. Without choice there is no freedom, and therefore no justice; with choice there inevitably will be differences and diversity. [Davidson et al 2009: 42]
The view could be described as ‘liberal’ or ‘procedural’. It stresses the role of freedom and the proper diversity of responses to its exercise. By contrast, Kim Hopper argues in his paper ‘Rethinking social recovery in schizophrenia: what a capabilities approach might offer’ that:
Any application of capabilities must therefore first define/defend a (full or partial) list of valued functionings... [Hopper 2007: 876]
Hopper follows Nussbaum following Aristotle in assuming some objective and universally applicable limits to the proper exercise of choice and freedom. An objective and substantive list of valued functionings would be an objective standard independent of individual choices and that might, in principle at least serves as standards of correctness for them. That is, it makes sense on Hopper’s view to think that someone might be in error about the nature of their own flourishing. They might be able to follow the correct procedures of second order reflection on first order values but be in error about objective values at both levels. That possibility is ruled out on a liberal or procedural view.
A similar contrast is also evident concerning the authenticity of choice. Davidson et al claim that mental illness does not affect the status as an agent of individuals.
There can be no recovery without self-determination… Mental illness may pose an obstacle to the person’s achievement of the kind of life he or she wishes to have, may make it more difficult to live that life, and, at its most extreme, may even deprive the person of life altogether. In none of these cases, though, does mental illness fundamentally alter the basic nature of human beings, which is that of being self-determined agents, free to choose and pursue the kind of life they as individuals value. Mental illness does not rob people of their agency, nor does it deprive them of their fundamental civil rights. [Davidson 2009: 4-1 italics added]
By contrast Hopper warns that the choices made people with mental illnesses may lack authenticity. Their choices may be affected or distorted as a consequence of illness itself or their treatment as a result of that illness.
Deprivation and disgrace can so corrode one’s self worth that aspiration can be distorted, initiative undercut and preferences deformed. Sensitive work will be needed to recover that suppressed sense of injustice and reclaim lost possibility. [Hopper 2007: 877]
Whichever view one takes of the local and diverse or universal view of capabilities and the corresponding opposition between procedural versus substantive view of autonomy, Hopper’s point is surely empirically correct. The stigma of carrying a mental illness diagnosis is often reported to be as disabling as the mental itself [Corrigan and Watson 2002]. That raises the possibility of an indirect connection – mediated by medical and broader societal treatment – between illness and aspiration. But there are also direct connections. In his study of the phenomenology of depression, Matthew Ratcliffe reports that in severe depression not only is motivation undermined but awareness of the very possibilities for action diminish. Objects are no longer imbued with the possibilities for action. So it is not just that there are possible actions which the sufferer feels incapable of taking up. Rather, the sense of such possibilities also vanishes. In extreme cases, this undermines an understanding of other people’s purposive actions.
[M]ore profound losses involve an inability to comprehend the possibility of anything being practically significant for anyone:
But in among the bad and worse times, there were also moments when I felt, if not hope, then at least the glimmerings of possibility… It was like starting from the beginning. It took me a long time, for example, to understand, or to re-understand, why people do things. Why, in fact, they do anything at all. What is it that occupies their time? What is the point of doing? During my long morning walks, I watched people hurrying along in suits and trainers. Where was it they were going, and why were they in such haste? I simply couldn’t imagine feeling such urgency. I watched others throwing a ball for a dog, picking it up, and throwing it again. Why? Where was the sense in such pointless repetition? [Brampton, 2008, p.249]
This description of the ‘return of possibility’ serves to make salient what was previously diminished or lost: a sense of what it is for someone to act purposively, to find things significant and respond to them accordingly. Activities such as playing with a ball or hurrying to a destination had become strange, unfamiliar, bereft of meaning. The depressed person therefore experiences her situation as something she cannot act upon. [Ratcliffe 2015: 167]
The examples Ratcliffe describes suggest that Davidson et al are wrong to deny that mental illnesses can rob people of their agency. Such a connection to agency seems, to the contrary, to be a key element in the way that mental illnesses cause harm. This may not be so obvious in in schizophrenia, for example, but it is still plausible to think that delusions disrupt the formation of intentions for action [Fulford 1989]. Perhaps the reason Davidson makes that claim is a confusion with, or assimilation to, the claim that follows: ‘nor does it deprive them of their fundamental civil rights’. This claim might express the following warning. One should not assume that just because someone has a mental illness that they therefore forfeit fundamental civil rights premised on the idea of being an autonomous human agents. But even this claim has to accommodate the complication of legal detention and compulsory treatment under mental health legislation. Such a widely held legal principle suggests that mental illness can, in a limited and nuanced way, even alter fundamental civic rights.
Hopper’s comment that reclaiming a lost sense of possibility will require ‘sensitive work’ suggests, however, the danger that Nussbaum and Sen flag concerning an objective view of human flourishing: the danger of paternalism. If the central aim of the recovery model is to articulate a conception of a life worth living which fits the values of the person concerned, but if mental illness can corrode their sense of possibility and undermine their agency, how can the right endpoint for healthcare for them be selected without external imposition? Although highlighted within conceptions of recovery drawn from a capabilities approach, this general problem seems also to affect the more abstract account of recovery developed in the previous section. If recovery is aimed at a conception of flourishing articulated by the mentally ill patient or client him or herself, and if mental illness can affect his or her ability authentically to choose, what should guide the right conception of recovery?
In the final two sections of this chapter, I will suggest a role for narrative in addressing this problem. In the next section I will highlight the connection between recovery and narrative and draw on one particular view of narrative for a narrative sense of self. In the final section I will argue that this helps address the risk of paternalism whichever view of capabilities and whichever view of autonomy is adopted.

The link to narrative
To begin the section, I will first motivate the idea that narrative is a helpful place to address the tension between two different versions of the capabilities-based approach to recovery. There have been a number of claims that there is a close direct connection between recovery and narrative understanding. One indication of this is the proliferation of ‘recovery stories’ as part of the promotion of the recovery approach. These explore:
the personal and existential dimensions of recovery, taking the form of subjective and self-evaluated accounts of how an individual has learned to accommodate to an illness. These accounts have become the founding stories of the recovery movement [e.g. Chamberlin, 1978; Lovejoy, 1984; Deegan, 1988, 1996; Leete, 1989; Unzicker, 1989; Clay, 1994; Coleman, 1999; Ridgeway, 2000], and anthologies of these personal stories have been used by governments and professions as a means of combating stigma and reasserting a focus on personal perspectives [Leibrich, 1999; Lapsley et al, 2002; Ramsay et al, 2002]. [Roberts and Wolfson 2004: 38-9]
More generally it is claimed that people who have suffered mental illness can be helped towards recovery through a narrative based theory. Pat Bracken and Phil Thomas, for example, cite both Larry Davidson and Glen Roberts.
In their work on recovery, both Davidson and Roberts choose not to use an approach grounded in traditional descriptive psychopathology, but turn instead to narrative theory and methods, seeing this as providing a rigorous empirical and clinical methodology in helping people suffering from chronic psychosis to move to recovery. [Bracken and Thomas 2009: 245]
On Bracken and Thomas’ view, clinical work is informed by narrative theory which is distinct from traditional Jasperian descriptive psychopathology. Roberts himself makes the connection between recovery and narrative even closer (or at least even more explicit). By contrast with Bracken and Thomas, he sees a narrative view as consistent with Jaspers’ view of understanding (by contrast with explanation).
A narrative view values content, and in seeking to understand delusions and hallucinations, as opposed to explaining them [Jaspers 1974], one is engaged in re-contextualising the illness in the life experience of the individual. This in turn may inform the rehabilitation process and give insight into the complexities of recovery, which for some will include the loss of the compensations of delusional beliefs and re-engagement with the implications of having a severe mental illness and what preceded it [Roberts 1999]. [Roberts 2000: 436]
But he goes on to suggest (or at least to make explicit the idea) that subjects or patients – rather than only clinicians – possess a narrative understanding and it is this which can help or hinder them in recovery.
Patients with self-sufficient, unelaborated, dismissive narratives need to be encouraged to break open their defensive stories and consider other possibilities. Conversely, those who seem unable to find a narrative thread and to be drowning in the chaos of their experience need help to find a shape and pattern that enables them to fit things into place [Holmes, 1999]. In therapy, patients learn to build up their storytelling capacity, their “autobiographical competence” [Holmes, 1993]. [Roberts 2000: 436]
Across the literature there is evidence of an idea that narratives structure subjects’ lives in such a way that partially determines what seems possible to them. Therapists can propose new life ‘plots’ and help map out new possibilities in the face of mental illness and hence new possibilities for recovery. To take a non-mental health example, therapists can propose new plotlines to spinal cord injured patients for whom there is, sadly, no going back to their past able-bodied plots. They have suffered a kind of ‘narrative wreckage’ from which they need rescuing in a dialogue with therapists [Frank 1997: 53-56].
Whilst that direct connection between narrative understanding and recovery is potentially an important clinical one, it is not a necessary connection. Exploring the options for a flourishing life through the idea of stories may in itself be directly therapeutic for many people, but it is possible that such an approach might fail because, for example, it may remind people of what they cannot do anymore.
What is the conceptual or logical connection between a recovery model and narrative understanding? Narrative theorists who apply the idea of narrative to social phenomena face a strategic choice. Either, they offer a specific detailed account of narrative in which case it narrows the range of application of narrative theory since few social phenomena will fit all the characteristics of narrative so defined. Or, they stress the universal application of narrative theory and hence are forced to offer a more general, thinner characterisation of narrative.
In the former approach, they may, for example, divide narratives into: abstract, orientation, complicating action, evaluation, resolution and coda [Labov and Waletsky, 1967]. Or, alternatively: temporality, people, action, certainty (or not) and context [Clandinin and Connelly 2000]. In practice there is a great deal of overlap between the accounts. Nevertheless, there is no obvious necessary connection between a recovery model and narrative understood in this detailed concrete way. Whilst narratives may often have the elements narrative theorists favour, and whilst this may contingently be true of the kind of ‘recovery stories’ often published in support of the recovery approach, there is no need to link the very idea of recovery in mental healthcare to the provision of a narrative with all of the parts of a favoured detailed theory of narrative.
But if not, what is the role of narrative understanding and recovery? A more plausible approach to answer this question is to take a more modest view of the necessary and sufficient elements of narrative understanding. The philosopher Peter Goldie articulates just such a minimal account in his last book The Mess Inside [Goldie 2012]. His account of narrative starts with the general claim that:
A narrative is a representation of events which is shaped, organized, and coloured, presenting those events, and the people involved in them, from a certain perspective or perspectives, and thereby giving narrative structure – coherence, meaningfulness, and evaluative and emotional import – to what is narrated. [Goldie 2012: 8]
This definition emphasises that a narrative is a representation. It is constructed from the perspective of a narrator even when the narrative is not actually written or told and is merely an exercise of thought by a subject. Thus the narrative is fundamentally distinct from what it represents: the events or, for example, the life of a subject being narrated. The three characteristics of the structure of the narrative thus pertain to the narrative and not to what is narrated.
The first characteristic feature of a narrative is that it has coherence, in the sense that it reveals, through the process of emplotment, connections between the related events, and it does so in a way that a mere list, or annal, or chronicle, does not. [ibid: 14]
In fact, Goldie does not attempt to say very much about the nature of narrative coherence. Surprisingly, given that narrative structure might be thought to be distinct from the nomological or lawlike structure of the physical sciences, he uses a comparison with causal explanation. Such explanations cite factors that are typically neither necessary nor sufficient for the explanans [Mackie 1993]. But they are selected from the total set of causal factors for an event on the basis of the interests of the author and audience of the account. Although he denies that there is a simple relation between causal explanation and narrative understanding, Goldie suggests that this is common with narrative accounts.
A further clue to the nature of narrative coherence comes from a remark on the epistemological of narrative construction. Borrowing and modifying Paul Ricoeur’s notion of ‘emplotment’ itself based on Aristotle’s Poetics, Goldie suggests that it involves shaping, organizing, and colouring events, the raw material of the narrative, itself always already described in significant ways. Arriving at this is a matter of tâtonnement or trial and error, feeling one’s way to the real significance of what is narrated.
The process of emplotment is often a tâtonnement, a tentative, groping procedure: one might begin with an idea of how the narrative should be shaped, and, once one has developed it somewhat, one might be able to see saliences that one could not see before, and then find it appropriate to go back and reshape the narrative in this new light. More than that, the tâtonnement typically involves a groping search for the appropriate evaluative and emotional import of what is narrated. [ibid: 11]
The second general characteristic set out is meaningfulness. Goldie suggests that there are two aspects to the meaning of narratives.
First, a narrative can be meaningful by revealing how the thoughts, feelings, and actions of those people who are internal to the narrative could have made sense of them from their perspective at the time—that is, from their internal perspective. And, secondly, a narrative can be meaningful by revealing the narrator’s external perspective: his or her thoughts and feelings that throw light on why the narrative was related (or just thought through) in that particular way. Bound up with these two kinds of meaningfulness are the two ways in which a narrative can have evaluative and emotional import. [ibid: 17]
This connects to the third key element: the evaluative and emotional import of narratives. In a nutshell.
Things matter to people, and a narrative involving people can capture the way things matter to them. [ibid: 23]
These characteristics are supposed to apply across the board to narratives. Goldie has, however, a particular focus: autobiographical narrative thinking. In particular, he discusses in depth autobiographical thinking about one’s past and one’s future. Both illustrate the claim that an autobiographical narrative sense of self is an essential aspect of human subjectivity.
In the case of narrative thinking about the past, the difference between internal and external perspective on both meaning and evaluative and emotional import plays a role. In thinking, now, about one’s past, an autobiographical narrative presents a view of the meaning and significance of past events and actions through the emotional lens of the present. A mismatch between past and present perspectives can take the form of regret in cases where one realises that had one not acted in such and such a way, such an effect would not have occurred.
In narrative thinking about the future, imagination plays a role analogous to memory in thinking about the past. Goldie suggests that it plays a key role in thinking through the ‘branching possibilities’ of different ways in which events may come to pass. In future directed thinking, there is again a difference between internal and external perspective on both meaning and evaluative and emotional import. One can imagine not only different future actions but also their emotional effects. The emotion imagined for the future can have an emotional effect in the present. Goldie suggests that narrative thinking is also involved in developing virtues. One connection is through fictional narratives. An emotional response to the lives and actions of fictional characters enables an understanding goes had in hand with an imagined response to other possibilities. But the same applies to future planning for one’s own life based on responses of shame and guilt to mistakes made in the past and hence the adoption of counter-factual and causal thinking about how to act in the future.

Narrative, paternalism and Hopper’s ‘sensitive wok’
Kim Hopper’s mention (above) of the ‘sensitive work’ needed to recover a suppressed sense of injustice and reclaim lost possibility reflects the dual dilemma for a clinician of either paternalistically imposing an entirely external view of what someone in recovery from mental illness ought to want or failing to challenge a view which may be impoverished by mental illness, stigma and even, in some cases, experience of psychiatric treatment. If the central aim of the recovery model is to articulate a conception of a life worth living which fits the values of the person concerned, but if mental illness can corrode their sense of possibility and undermine their agency, how can the right endpoint for healthcare for them be selected?
Although it does not offer a way out of the fundamental tension articulated by Nussbaum and Sen, a narrative sense of self offers some practical help. Narrative, understood along the lines that Goldie suggests, has a balance of subjectivity and objectivity. The objectivity lies in the need for narrative coherence and structure and in the implicit contrast between internal and external perspectives in narratives. The subjectivity lies in the essentially perspectival nature of the narratives. Autobiographical narratives are both narrated from a particular perspective but also concern the life of the very same person as the narrator. These come together in the tâtonnement, the tentative, reaching for the right narrative structure of events.
On a broadly procedural view of autonomy, any intervention to impose values on a subject who already enjoys a reflective balance between first and second order values and policies is a form of paternalism. But there may still be a need for what Hopper calls ‘sensitive work’ is a subject’s self narrative is incoherent. For example, is there is a lack of coherence between past regret and suitable future directed policies to avoid similar errors in the future. Clinical narrative-based intervention could take the form of working with a client to enable their understanding of their past and plans for the future to fit their current conception of themselves. Narrative self-understanding introduces a temporal dimension and the explicit possibility of divergence of value between what one valued in the past, values now and may come to value in the future. Narrative coherence provides a standard by which present values and habits can be subject to criticism by the subject her or himself. In the initial apparent dilemma, an extreme of relativism is avoided insofar as a conception of flourishing and hence recovery will have to have a form of narrative coherence for the subject. Whilst the notion of objectivity within narrative understanding is not one of external
On a substantive view of autonomy, answering to external, universal standards for human flourishing need not involve paternalism. Hence a clinician’s ‘sensitive work’ may involve persuasion that some forms of life are more valuable than others and their neglect by a subject is the result of his or her illness. Whilst external standards need not seem to a substantivist to threaten paternalism they could. Given the connection, to which both sides of the debate about capabilities agree, between flourishing and freedom a brute imposition of an external view of flourishing could be paternalistic. But if externally proffered conceptions of the good life are suggested and adopted as part of a narratively structured set of policies and plans for future life reconciled with the subject’s narrative account of their past and present too then that is not a brute imposition. Even if a clinician has to work to introduce neglected values, if they are adopted they will have to made to cohere by the subject him or herself with a narrative account. This avoids the charge of crude paternalism.
Emphasising the normative standards implicit in the idea of narrative coherence is not a quick fix for a balanced conception of recovery. The severely depressed person quoted by Ratcliffe (above) was, at the time of the experiences described, incapable of the kind of narrative understanding just outlined. In such cases, a recovery model can neither be adopted nor can it coherently be imposed. But a narrative sense of self suggests a rationale for a middle way between paternalism and uncritical relativism whichever broad view of recovery and of autonomy is adopted.

Conclusion
In this chapter I have attempted to shed light on three related issues. First, in the light of its recent popularity and promotion as the goal of mental healthcare, what is meant by 'recovery'? Second, how can the harm that mental illness can do to people's autonomy be reconciled with a recovery approach which is based on patient or client choice? Third, does the existing connection between recovery and narrative shed light on the first two issues?
Drawing on typical statements of the aims of the recovery movement in mental healthcare, I have articulated an abstract specification for a recovery model. It is health rather than illness based and it is essentially value laden. Recovery aims towards the goal of a conception of flourishing tailor-made to individual patients and clients. This is consistent with one elaboration of a theory of recovery based on Sen's capabilities approach to welfare economics. On that model, the valued 'functionings' that underpin capability are person-specific. But as two of the key architects of that policy suggest, there is a tension between two rival versions of it. On one (Nussbaum's), there are universal standards for human flourishing. On the other (Sen's), the emphasis on freedom as a key aspect of capability implies diversity. These two views give rise to two opposing perceived dangers: uncritical relativism versus paternalism.
Paternalism is a threat to human autonomy and the two approaches to capabilities correspond to two broad approaches to it. On a proceduralist account, autonomy is a form of internal coherence of beliefs and values. On a substantivist account, the idea of autonomy presupposes successfully tracking some particular conception of the good and the true.
The two view find expression within recovery on a stress on human diversity (Davidson) versus the idea of some universal standards coupled with the idea that mental illness may cloud people's judgements of flourishing so understood or otherwise undermine their agency (Hopper). This raises a threat, however, that in responding to this clinicians will inevitably be paternalist.
On the assumption that mental illness can indeed distort people's views of their own flourishing, the final section has suggested that a narrative sense of self can play a role in undermining the threat of paternalism. On a proceduralist view of autonomy, and Sen's version of capabilities, a clouding of judgement is a lack of coherence in a narrative sense of self and hence the sensitive work to correct this involves helping the patient or client to repair their own narrative. On a substantive view of autonomy, which corresponds to Nussbaum's view of capabilities, a clouding of judgement may also involve failing to observe universal forms of human flourishing. And hence the sensitive work to correct this involves enabling a patient to see the rightness of these universal forms. But the role of narrative ensures that this is not a brute imposition from without. Only if the subject can integrate the universal values into his or her own narrative will the work have been successful.

Bibliography
Boorse, C. (1975) ‘On the distinction between disease and illness’ Philosophy and Public Affairs 5: 49-68
Bracken, P. and Thomas, P. (2009) ‘Authors’ response. Invited commentary on… Beyond consultation’ Psychiatric Bulletin 33: 245-246
Brampton, S. (2008) Shoot the Damn Dog: A Memoir of Depression. London: Bloomsbury
Brown, W. and Kandirikirira, N (2007). Recovering mental health in Scotland. Report on narrative investigation of mental health recovery. Glasgow, Scottish Recovery Network
Chamberlin, J. (1978) On Our Own: Patient Controlled Alternatives to the Mental Health System, New York: McGraw-Hill
Clandinin DJ & Connelly FM (2000) Narrative Inquiry: Experience and Story in Qualitative Research. Jossey-Bass, San Francisco, CA.
Clay, S. (1994) The wounded prophet. In Recovery: The New Force in Mental Health, Columbus, OH: Ohio Department of Mental Health
Coleman, R. (1999) Recovery: An Alien Concept, Gloucester: Hansell Publishing
Corrigan, P.W. and Watson, A.C. (2002) ‘The Paradox of Self-Stigma and Mental Illness’ Clinical Psychology: Science and Practice 9: 35–53
Davidson, L. and Roe, D. (2007) ‘Recovery from recovery in serious mental illness: one strategy for lessening confusion plaguing recovery’ Journal of Mental Health 16: 459-470.
Davidson, L., Ridgway, P., Wieland, M., & O'Connell, M. (2009). A capabilities approach to mental health transformation: a conceptual framework for the recovery era. Canadian Journal of Community Mental Health (Revue canadienne de santé mentale communautaire), 28(2), 35-46
Deegan, P. E. (1988) ‘Recovery: the lived experience of rehabilitation’ Psychosocial Rehabilitation Journal, 11: 11-19
Deegan, P. (1996) ‘Recovery as a journey of the heart’ Psychiatric Rehabilitation Journal 19: 91-97
Frank, A.W. (1997) The Wounded Storyteller, Chicago: University of Chicago Press
Frankfurt, H. (1971) ‘Freedom of the Will and the Concept of the Person’ Journal of Philosophy
68:5-20
Fulford, K.W.M. (1989) Moral Theory and Medical Practice, Cambridge: Cambridge University Press
Goldie, P. (2012) The Mess Inside: narrative, emotion, and the mind, Oxford: Oxford University Press
Holmes, J. (1993) John Bowlby and Attachment Theory. London: Routledge
Holmes, J. (1999) ‘Defensive and creative uses of narrative in psychotherapy’ In Healing Stories: Narrative in Psychiatry and Psychotherapy (eds G. Roberts & J. Holmes) pp. 152–180. Oxford: Oxford University Press.
Hopper, K. (2007) Rethinking social recovery in schizophrenia: what a capabilities approach might offer. Social Science & Medicine 65: 868-879
Jaspers, K. ([1913] 1974) ‘Causal and “Meaningful” Connections between Life History and Psychosis’, trans. by J.Hoenig, in S.R.Hirsch and M.Shepherd. in Hirsch, S.R., and M. Shepherd, Themes and Variations in European Psychiatry, Bristol: Wright: 80-93
Kendell, R.E. (1975) ‘The concept of disease and its implications for psychiatry’ British Journal of Psychiatry 127: 305-315
Labov, W. and Waletsky, J. (1967) Narrative analysis, in J. Helm (ed) Essays on the Verbal and Visual Arts. Seattle: University of Washington Press.
Lapsley, H., Waimarie, L. N. & Black, R. (2002) Kia Mauri Tau! Narratives of Recovery from Disabling Mental Health Problems, Wellington: Mental Health Commission
Leete, E. (1989) ‘How I perceive and manage my illness’ Schizophrenia Bulletin, 8: 605-609
Leibrich, J. (1999) A Gift of Stories: Discovering How to Deal with Mental Illness, Dunedin: University of Otago Press
Lovejoy, M. (1984) ‘Recovery from schizophrenia: a personal Odyssey’ Hospital and Community Psychiatry, 35: 809-812
Mackie, J.L. (1993) ‘Causes and conditionals’ in Sosa, E. and Tooley, M. (eds) Causation, Oxford: Oxford University Press: 33-50
Nussbaum, M. (1988). ‘Nature, Function, and Capability: Aristotle on Political Distribution’, Oxford Studies in Ancient Philosophy, suppl. vol.
Nussbaum, M. and Sen, A. (eds) (1993) The Quality of Life, Oxford: Clarendon
Ramsay, R., Page, A., Goodman, T., et al (2002) Changing Minds: Our Lives and Mental Illness, London: Gaskell
Ratcliffe, M.R. (2015) Experiences of Depression: a study in phenomenology, Oxford: Oxford University Press
Ridgeway, P. A. (2000) ‘Re-storying psychiatric disability: learning from first person narrative accounts of recovery’ Psychiatric Rehabilitation Journal 24: 335-343
Roberts, G. (1999) ‘The rehabilitation of rehabilitation: a narrative approach to psychosis’ In Healing Stories: Narrative in Psychiatry and Psychotherapy (eds G .Roberts & J. Holmes), pp. 152-180. Oxford: Oxford University Press.
Roberts, G. (2000) ‘Narrative and severe mental illness: what place do stories have in an evidence-based world?’ Advances in Psychiatric Treatment 6: 432-441
Roberts, G. and Wolfson, P. (2004) ‘The rediscovery of recovery: open to all’ Advances in Psychiatric Treatment 10: 37-49
Sen, A., 1980, ‘Equality of What?’ in McMurrin (ed.), Tanner Lectures on Human Values, Cambridge: Cambridge University Press.
Sen, A. (1993) ‘Capability and Well-Being’ in Nussbaum, M. and Sen, A. (eds.) The Quality of Life, New York: Oxford Clarendon Press
Shepherd, G., Boardman, J. & Slade, M. (2008) Making Recovery a Reality London: Sainsbury Centre for Mental Health
Szasz, T. (1972) The Myth of Mental Illness, London: Paladin
Unzicker, R. (1989) ‘On my own: a personal journey through madness and re-emergence’ Psychosocial Rehabilitation Journal 13: 70-77
Wolf, S. (1990) Freedom within Reason Oxford: Oxford University Press