Article Text

Download PDFPDF

Truth telling, autonomy and the role of metaphor
  1. D Kirklin
  1. Correspondence to:
 D Kirklin
 Department of Primary Care and Population Sciences, Royal Free and University College Medical School, Archway Campus, 2nd Floor, Holborn Union Building, 2-10 Highgate Hill, London N19 5LW, UK; d.kirklin{at}


This paper examines the potential role of metaphors in helping healthcare professionals to communicate honestly with patients and in helping patients gain a richer and more nuanced understanding of what is being explained. One of the ways in which doctors and nurses may intentionally, or unintentionally, avoid telling the truth to patients is either by using metaphors that obscure the truth or by failing to deploy appropriately powerful and revealing metaphors in their discussions. This failure to tell the truth may partly account for the observation by clinicians that patients sometimes make decisions that, from the perspective of their clinician, and given all that the clinician knows, seem unwise. For example, patients with advanced cancer may choose to undergo further, aggressive, treatment despite the fact that they are likely to accrue little or no benefit as a result. While acknowledging that the immediate task of telling patients the truth can be difficult for all those concerned, I argue that the long-term consequences of denying patients autonomy at the end of life can be harmful to patients and can leave doctors and nurses distressed and confused.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

When a doctor fails to tell a patient the truth, whether about the diagnosis, prognosis or the potential risks and benefits of the alternative treatment options, the doctor is, in effect, making a unilateral decision to deny the patient the opportunity to exercise his or her autonomy. Higgs1 has written extensively about truth telling in medicine, specifically about the temptation for doctors to be economical with the truth, ostensibly to protect patients from distress. He reasons that although telling the truth in certain medical situations can indeed be distressing, it is often the doctor’s distress rather than the patient’s that is the driving force behind the failure to tell the truth.

The assertion that honesty is always the best policy is controversial.2 Sometimes, competing moral imperatives, such as the duty to do good or the duty to avoid harm, require decisions to be made about which should take precedence.3i Some authors have argued that the priority given to truth telling, and thereby to respect for autonomy, is culturally determined and that one size will not necessarily fit all.4,5 In one study, oncologists were interviewed to try to understand their views about disclosing information about prognosis to patients with advanced cancer. The oncologists acknowledged the importance of being truthful and saw respecting the patient’s autonomy as an important principle. They did, however, report reluctance in being too specific in detailing prognosis, preferring instead to maintain hope. In explaining their hesitation, they used powerful metaphors, such as “hitting you over the head”, to try and explain to the interviewers what effect being truthful in this situation may have on patients. Writing in the context of the care of patients with cancer, Freedman6 has suggested that the truth is something that should be offered to patients rather than imposed on them. Implicit in this suggestion is an acknowledgement that although not all patients will want to know everything, it is nevertheless for the patient, and not the healthcare provider, to decide whether he or she wishes to have access to the truth. Truth telling could of course be taken to mean simply the accurate communication of a set of facts. However, in this context, I would argue that unless those facts are communicated in such a way that the patient is able to interpret them—that is, to understand what they mean for him or her—the communication or offering of the truth will have been unsuccessful. In this paper, “truth telling” should therefore be understood to mean a communication of the facts in such a way that the recipients are able, should they so wish, to understand the consequence of the communicated facts for them.

Communicating complex information to people with neither professional nor personal experience of the matters being discussed is of course a challenging task. With the best will in the world, it can be extremely hard to be truthful in the sense of conveying a true impression of issues such as risk and of what it is like to undergo certain treatments or side effects. If patients have experience relevant to the facts being communicated, perhaps because they or a family member has previously undergone a very similar intervention, then little imagination may be required for them to understand what these facts mean for them. If, however, patients have no relevant experiences on which to draw, then an act of imagination may well be required for them to understand the implications of what is being communicated. In this paper, I examine the potential role of metaphor in helping doctors to help patients engage in this act of imagination and thereby gain a richer and more nuanced understanding of what is being explained.

I am not of course suggesting that the deployment of metaphor is either the only or the preferable way to communicate the truth. Nor am I making any claims about whether healthcare professionals are, as a group, inclined to understate (rather than overstate or accurately state) the severity of the patient’s condition in communications with them. Instead, I suggest that as the use of metaphor has the potential to markedly affect the effectiveness of any attempt to either offer or obscure the truth, an understanding of how metaphor functions should be of interest to both medical ethicists and healthcare practitioners.


The suggestion that metaphors can and do play a useful part in caring for and communicating with dying patients has been made by palliative care nurse Hutchings,7 who writes that “the creative and judicious use of metaphor provides health care practitioners with many veils—veils that shield the dying from the glare of their prognosis, veils particularly valuable and suited in communicating with our palliative patient population.” Rather than making communication more transparent and honest, the intention is to partially obscure the painful truth and thereby to avoid distressing the patient. I will argue that in addition to using metaphor in this way—that is, to actively obscure the truth—doctors and nurses who fail to deploy appropriately powerful and revealing metaphors are thereby failing to tell the truth to patients. I suggest that this failure to tell the truth may partly account for the observation by clinicians that patients sometimes make decisions that, from the perspective of their clinician, seem unwise. I make this argument with specific reference to patients with advanced cancer who choose to undergo further, aggressive treatment despite the fact that they are likely to accrue little or no benefit as a result. Before beginning such an examination, I outline some of the different understandings of metaphor and indicate the way in which metaphor may be functioning in this setting.


 Metaphor consists in giving the thing a name that belongs to something else.
 Aristotle, Poetics.8

Scholarly examination of metaphor has a long and distinguished history stretching back to Aristotle. I will focus on two major strands of the contemporary study of metaphor: the first draws on the seminal analysis by philosopher Riceour9 of what he calls The rule of metaphor, the second on the work of cognitive linguists as exemplified by the work of Lakoff and Johnson,10,11 who discuss the embodied nature of metaphor. The work of Sontag,12 who argues that the type of metaphor used about an illness reflects the degree to which it is perceived as a threat to society, is interesting but will not be detailed here.

According to Riceour,9 Aristotle’s definition of metaphor combines three distinct elements: deviation from ordinary usage; borrowing from an original domain; and the substitution of an absent but available ordinary word. According to this definition, metaphor is little more than decorative or ornamental: a more interesting, witty or pleasing way of saying exactly the same thing. Sometimes, however, as Aristotle made clear, there is no existing word for the object that the metaphor designates. In these cases, one of the functions of metaphor is to fill a semantic lacuna. Riceour9 (p 23) expands on this idea by arguing that

 ... it always takes two ideas to make a metaphor. If metaphor involves taking one thing for another by a sort of calculated error, then metaphor is essentially a discursive phenomenon. To affect just one word, the metaphor has to disturb a whole network by means of an aberrant attribution.

Riceour9 goes on to say that categorical transgression, giving the thing a name that belongs to something else, is only interesting because it creates meaning. He argues that metaphor bears information because it redescribes reality in a way that is made possible by the use of the calculated category—mistake. The deployment of metaphor therefore requires the creation of a new order by the creation of a rift or disturbance in the old order. For example, if I were to describe Florence Nightingale as the mother of modern nursing, then you, the reader, would need to engage in an act of imagination to try to understand what I mean. According to Riceour’s analysis, by redescribing the reality of who Florence Nightingale was or is in this way, I am creating meaning, or enriching your understanding of who, from my perspective, Florence Nightingale is. Now although Riceour’s analysis provides a practical and intellectually pleasing way to understand metaphor, it still leaves unanswered a very important question at the heart of much of the contemporary debate in this area. If metaphor requires a deviation from the ordinary usage of a word, what exactly is meant by “ordinary” usage? What, in the example I have given, is the ordinary usage of the word mother? The theory of metaphor proposed by Lakoff and Johnson10,11 offers one answer to this question.


The approach taken by Lakoff and Johnson and others in the field of cognitive psychology is aimed at understanding the ways in which the human mind processes and interprets language. They argue that the successful deployment of metaphor depends on an elaborate and neurologically determined—that is, embodied—categorisation of the concepts associated with language. Each person possesses a cognitive map made up of a web of concepts organised in such a way that abstract concepts can be grounded or understood relative to the cognitive agent’s physical experiences and the agent’s relationship to the external world. Metaphors play an important part in the assimilation of novel abstract concepts by providing an experiential framework within which the concept can be thought about. Key to this idea is the contention that humans tend to conceptualise the world through idealised cognitive models. If this theory is correct, then in processing the metaphor above you will refer to the idealised cognitive model of “mother” that has been neurologically laid down for you. If you and I share a similar social, historical and cultural background, then our idealised concept of a mother will probably be similar: that of a nurturing, supportive, caring and generous woman who, literally, gave birth to us. If, however, you had a different experience of what mothers are, then the phrase “Florence Nightingale was the mother of modern nursing” may invoke for you images of a dominating, ruthless, determined and competitive person. Even if this is so, if we do share a similar cultural background, you may well share my idealised cognitive model of “mother” while recognising that your personal conception of your mother is itself a deviation from this ideal. In trying to work out what I mean by my metaphor, you will therefore have to draw on both this idealised cognitive model and the various partial cognitive models that reflect your understanding of the fact that not all mothers conform to this ideal. Furthermore, by engaging with my metaphorical use of the word mother you will expand your own partial cognitive models for this term and thereby create new, neurologically determined, ways of understanding what is meant by “mother”.

This system of categorisation therefore views metaphor as a cognitive device that does just as Riceour proposes: it expands the conventional use of language to create new ways of understanding. If these views of metaphor are correct, and one of the functions of metaphor is to redescribe reality and thereby to gain new insights into that reality, then the judicious use of appropriate metaphors by doctors and patients may offer new perspectives into what it means to be ill and to undergo treatments. It is perhaps worth noting that although this theory sounds superficially similar to Quine’s web of belief13 theory (in which an inter-related network of revisable propositions form the basis of all intellectual activity), it is in fact incompatible with the embodied characterisation of meaning that cognitive psychologists like Lakoff and Johnson propose. This confusion may, ironically, arise because both theories use the term “web” metaphorically.


Practitioners sometimes feel great uncertainty about the ability of patients to make fully informed decisions about treatment—for example, as stated earlier about the continuation and commencement of invasive treatments—at a time when there is little or no hope of cure or remission.14ii In this context, I intend the term “unwise” to reflect the perspective of the healthcare professionals who do not think the patients’ decisions are in their own best interests. The wisdom or otherwise of any particular decision is of course a matter of judgement and not what primarily concerns me here. For the sake of the argument I wish to pursue here, let us assume that the healthcare professional’s judgement is sound and the decision made by the patient is indeed unwise. One possible explanation for why the unwise decision has been made is that although the patient has been given all the appropriate information, he or she has not managed to assimilate it. Unless there is a physical or psychological reason for this, this is effectively a straightforward failure of communication. When patients make decisions that doctors consider unwise, this possibility—that they have failed to understand the information they have been given—is often addressed through further discussion and explanation.

An alternative possible explanation is that the patient has fully understood everything that has been explained, including the unpleasant consequences of treatment that may do little to slow the disease, but has nevertheless decided to go ahead. To use a gambling metaphor, this is a high-stake and high-risk gambler, ready to risk all on a last throw of the dice. From the perspective of the doctors and nurses, with only a theoretical understanding of the patient’s perspective, this decision may seem misguided. Distinguishing between poor decisions that patients make on the basis of poor understanding and decisions that patients make that seem poor from the professional’s perspective, but not from the patient’s perspective, is a daily task for those who deliver end of life care. Appreciating the subtle but fundamental distinction between the two is essential for any ethicist examining autonomy in end of life care.

A third possible explanation exists as to why patients sometimes make unwise decisions despite being in possession of all the relevant facts, and this explanation brings us back to the role of metaphors in medicine.


If metaphor obliges the reader, or listener, to imagine, might it therefore require the judicious use of appropriate metaphors, rather than simply the imparting of factual information, to bring the perspective of the patients, who can necessarily only imagine their own death, closer to the perspective of the professional, based on years of caring for dying patients?

And if there are suitable metaphors to help the patient imagine the horrible consequences of, for example, excessive use of treatments, is it that practitioners are unable to access metaphors that are powerful enough, or is it instead that they hesitate to bring about such a painful act of imagination on the patient’s behalf? If the second option is true, then this has important implications for consent, with the failure to use a powerful enough metaphor resulting in the patient not being fully in command of the necessary information about the consequences of the choice he or she is making. The patient’s choice is, as a result, not autonomous. The desire to avoid harm—in this case, the distress that may be caused to a patient by a powerful act of imagination about the end of his or her life—results in an arguably far greater harm whereby the imagined harm becomes a reality.

The hesitation I suggest that a practitioner may experience when deciding whether to use a powerful metaphor to provoke an act of imagination in the patient is understandable and often born out of concern not to cause patients any unnecessary distress, and a wish not to deny the patient hope.15 It may also be a consequence of an attempt to provide non-directional counselling. Nevertheless, it does raise the possibility that the natural instinct of healthcare practitioners to protect patients from unnecessary psychological suffering and the desire to remain neutral when imparting the information a patient needs to make an informed decision may, inadvertently, make it impossible for the patient to make a truly autonomous decision, as well as resulting in avoidable harm to the patient.


It is tempting at this point to tell you a few anecdotes, to tell you about specific cases I have been associated with where communication went well or poorly, and to share with you the part that metaphors played—for better or for worse—in those communications between patient and doctor. I am also tempted to offer you examples of metaphors whose use I consider could obscure the truth and of others whose use I consider could make the truth clearer. The examples I gave would of course be highly context specific with their appropriateness dependent on a complex mix of variables including, but not limited to, the patients’ medical and personal situation at that particular time, their cultural and social background, the extent to which they wanted to access the truth on offer and my own situation at that time, including my cultural and social background. All these factors would come into play as the patients and I tried to find a way to reach a shared understanding of what the truth was for them. So although by providing anecdotes and suggestions of helpful metaphors I may seem to be offering something useful to clinicians, what I would instead be doing is ignoring the fact that it is only in the unique context in which each clinical encounter occurs that those seeking to offer the truth can decide the extent to which they want the language, including metaphors, that they use to enable that process or hinder it. So rather than offering clinicians a detailed guide on how and which metaphors to use, I instead invite them to become aware of and thoughtful about the metaphors they choose or choose not to use in each of the unique clinical encounters to which they are party.


Telling patients the truth can be difficult for all those concerned, but it is also essential if respect for patient autonomy is to be paid more than lip service. If telling the truth is the gold standard in doctor–patient communications, then I have reasoned that the careful and imaginative deployment of metaphor could help achieve this goal. For some, perhaps many, healthcare professionals, the suggestion that doctors deploy powerful metaphors to enlighten patients about what may lie ahead will be considered unnecessarily cruel and harmful. But, as I have argued, the consequence of such well-intended kindness, although perhaps protective in the short term (of patient, family and professional carers), may in the long term return to haunt all three. If, at the end of life, healthcare professionals fail to tell patients the whole, unedited, often messy and sometimes disturbing truth, then patients are effectively denied the opportunity to make an informed decision about how they wish to spend the rest of their lives, and doctors and nurses can be left feeling distressed and confused about why patients sometimes make unwise and ultimately self-harming decisions.


I thank Professor John Harris and both reviewers for their helpful comments about the ideas explored in this paper.



  • i In this paper, the authors describe a case in which a mother was covertly giving the drug to her son, who had schizophrenia, in his soup. The authors ask whether the doctor should continue to prescribe the drug, thus becoming complicit in by-passing the patient’s autonomy.

  • ii This supplement was devoted to edited transcripts from the Swartz Rounds, and commentaries about the process, content and outcomes from these healthcare professional-centred oncology rounds.

  • Competing interests: None.

  • Funding: None.

Other content recommended for you