In Where the Ethical Action Is, 1 we argued that medical and ethical modes of thought are not different in kind but different aspects of a situation. One of the consequences of this, we argued, is that the requirement for or benefits of normative moral theorising in bioethics are undercut. We suggested that in its place medicine would be better served by a programme of exploring the practical and evaluative richness of medical situations in which moral choices and dilemmas arise. It is this secondary argument that seems to have raised the hackles of some in the professional bioethics community. Following on from the reply we received from Emmerich,2 and our response to that,3 we have received a reply from Wagner,4 seeking to defend normative moral theorising. Wagner’s challenge is based on an argument, drawing on an article by Roussos,5 to the effect that we should reconceive normative moral theories as (multiple) models. In this way, the models serve as simplified descriptions which, in Wagner’s view, may be of significant practical benefit to clinicians.

While we appreciate the reply, we find little in Wagner’s article that directly addresses the arguments we make in Where the Ethical Action Is. Instead, we find arguments implicit in what Wagner proposes about the reconceiving of normative moral theories as models. Therefore, our response will address these arguments.

Wagner argues that moral theorising should be reconceived as modelling. In proposing that moral theories are models, he hopes to import the arguments supporting modelling in other domains, such as the natural sciences, with a view to marshalling these to the cause of defending professional bioethics. We do not think this is a strong argument and we are sure it does not work.

In the philosophical literature on modelling, there are, broadly speaking, two senses in which the term is used, which we might call ‘methodological’ and ‘everyday’. Models in the methodological sense are pared-back, often-but-not-exclusively mathematical, representations as employed in the natural sciences; for example, ecologists studying a fish population in a pond (to invoke Roussos’ example). Models in the everyday sense are any representation or anything that implies a representation; for example, a conversation between parent a child about future educational choices that implies a representation of the hoped-for future state of affairs. In the latter, everyday sense, the argument is that modelling is a pervasive, perhaps even pre-eminent, mode of human engagement with the world and that modelling in the natural sciences is a kind of refinement of this. Harré6 was a proponent of the everyday view, cognitive psychology is predicated on this view.

What is the argument of Wagner? On the one hand, if Wagner is invoking models in the methodological sense, then we require an answer to the question of how this might be relevant for ethics. Scientific explanation of the natural world and moral judgement are different endeavours with different logics and objectives, based on being responsive to the world under different aspects. It is thus difficult to see how methodologically conceived models might be useful. On the other hand, if Wagner is invoking models in the everyday sense, then it is difficult to see how this is more than a rebranding exercise, in which we take any practice which involves or implies the representation of states of affairs and reconceive it as modelling. Indeed, the talk of modelling itself in this everyday sense can be seen as a rebranding exercise, to give what hitherto has been a philosophical discussion of representations (cognitive, linguistic, mathematical, etc) an update which confers a kind of scientific feel and respectability. What is gained by referring to representations as models?

In reading Wagner, we were reminded of an anecdote variously attributed to the sociologists Roy Turner and Aaron Cicourel, which is relayed in print by Cuff.7 The anecdote was targeted at the practices of mainstream sociologists preoccupied with theorising and structural analyses, while eschewing the detailed analysis of naturally occurring interactions. Cuff writes:

[T]here is something odd about a sociology which, in purporting to study the social world, can only do so by refining and reducing … naturally occurring occasions in order to produce more manageable and pliable ‘data’… [I]t is like an archaeologist refusing to be transported in a time machine back to the ancient civilization he has been painfully and laboriously studying, even though two days spent in the everyday life of that civilization would be worth more in terms of human knowledge than many lifetimes of normal archaeological endeavour. Similarly, we feel that the many sociologists who would look askance at considering [naturally occurring situations] as basic data for analysis are in effect 'archaeologists by choice'; they prefer to operate at several removes from daily life, rather than attempting to study and analyse the rich materials that surround them.

In Where the Ethical Action Is, we suggested a similar tendency holds in bioethics. Bioethics and sociology are not like archaeology. Accessing data in bioethics, like in sociology, does not require a time machine or special methods. What we need is the will to look closely at the naturally occurring situations that are readily available to us.

The burden of the advocate of modelling is to explain why, when you have the naturally occurring situations to explore, you would forego those situations to focus on models, which are partial by design. Models might be of use in some natural sciences, but it does not follow that they will be so in ethics. Take the case of modelling a pond of fish to study fish population fluctuations over time, as a team of ecologists might, and as Roussos depicts by way of providing an example of modelling. While we can grant that such modelling might serve useful specific explanatory purposes, this relies on a consensus among the ecologists (both in the research team and those accepting the explanation) as to what is explanatorily relevant and what can be legitimately excluded. In ethics, the question of what is morally relevant is precisely what is in question and thus is not something we can obtain consensus on as a preliminary. So, a bioethicist, like a surgical oncologist considering a prostatectomy or a sexual health consultant considering whether to offer PREP, has no more authority on the question of moral relevancy than other members constituting the situation in question. The very act of modelling, in being partial representation, potentially introduces, by stealth, a substantive evaluative framing of the situation. It pre-empts the ethical debate. It begs the ethical question. Therefore, we propose privileging situations and not their formal, idealised, representation as models.

The dispute here might be depicted as between abstraction and particularism. We argue that the particulars are what is important and that one attends to particulars by attending to situations. Moral particulars have a situated sense, which is to say, the moral sense they have is related to the situation in which they are encountered. In contrast, Wagner argues that it is useful to abstract from the particulars to learn from the idealisations encompassed by models. Wagner writes: ‘Each theory sheds light—via simplification and/or idealisation—on a different aspect of a given ethically complex situation, and the theories’ diverse contributions are clearly complementary, capable of together yielding a deep understanding of the key considerations in that situation.’ We contest that light is not shed by the theories of ethics (even if rebranded as models) but by the details of the situation.

There are two points counting against Wagner’s proposal. The first is the Turner-Cicourel challenge (TCC) and the second the question begging challenge (QBC).

The TCC highlights that models are always of something and that the choice to focus on the model when you have that something which is being modelled to-hand and available for investigation directly, seems absurd; it is to be archaeologists by choice.

The QBC draws attention to how pre-emptive ethical judgements about moral relevancy are front-loaded into the practice of modelling moral situations, importing by stealth moral arguments and begging the question to be investigated.

Not applicable.