Narrative vs. Theory

Narratives, Tod Chambers claims, are not mere companions to or illustrations of moral theory; instead, he says, narrative theory "should be thought of as being as vital to the discipline [of bioethics] as moral theory itself." (Chambers 2001) Chambers complains that the usual understanding of the role of narratives in bioethics is too marginal and too superficial. To combat this tendency, he suggests that we should take a serious theoretical look at the components of narrative and at the ways that narratives function in bioethics.

The fact that Chambers speaks of narrative theory reminds us that narratives are not non-theoretical raw data to be used for the illustration or testing of theories; narratives themselves are richly theoretical constructs. One of Chambers's key claims is that narratives "filter" some content out and some in. There are no complete, objective narratives; every description of a medical case leaves some features of the situation out and emphasizes other features. Therefore, setting out a description of a particular case in narrative is a practice that deserves critical attention. What we put into such stories, what we leave out, what we emphasize, how we phrase things, and more, all make a difference to what moral theories seem to be important to the case when it is described in a narrative.

This insight is familiar in the philosophy of science, where some time ago it was argued (and later widely accepted) that observations are "theory-laden"; that is, the observation of an experiment and any report of that experiment are necessarily filtered by certain background assumptions that can shape the way the experiment is described (e.g., Kuhn 1970; Feyerabend 1975). This idea is refreshing and illuminating when turned on narratives in bioethics. Chambers critically turns to various aspects of narrative cases and asks how these narratives are put together and why, demonstrating how some aspects of cases are systematically understated or ignored in a way that alters the moral meaning of the narrative.

Chambers is right. The way a story is told does matter to how it is assessed. How, then, should we tell these stories? Given that we still want to tell stories in bioethics, what is the best way to tell them? What features should we emphasize? To answer these questions, we might turn back to moral theory. Moral theories include what Barbara Herman calls "rules of moral salience," which pick out features of situations as worth attending to, and which filter out other elements because they are supposed to be morally neutral (Herman 1985). Consequentialist theories instruct us to emphasize the outcomes of different possible choices; Kantian theories lead us to look carefully at the agency and autonomy of the persons involved; feminist theories tell us to attend to the relationships between persons and to their gender roles, and so on. So the way we tell a story may already reflect a particular theoretical bias. Moral theories are often already incorporated into narratives.

Chambers concludes this precis by claiming that narrative theory "permits us to be critical of the evidence used to test moral theory." We cannot simply use narratives as tests that moral theories either pass or fail; this much seems right. But this should not be taken to imply that moral theories cannot help us to understand and evaluate narratives, or that narratives cannot help us to understand and evaluate moral theories. Narratives can and should still play a role in discussions about moral principles and theories, just as moral theory can play a role in constructing and analyzing narratives.

Any case can be described in a number of ways, and each moral theory can suggest a different way of describing the same case. How each case is described depends on what features of the case the moral theory says are morally salient. Imagine then that we produce a number of narratives, all describing the same case, but each emphasizing different aspects of that case. Each narrative reflects and reinforces the theoretical filter used to construct it. But now we can look at these different narratives, step back, and discuss which of them seem to leave important elements out...