engineers are to recognise
What to do with the ascribed ethics First, if engineers are to recognise ascribed obligations in the development of their ethics they need some kind of meta-ethic that modifies or combines these obligations
J. S. Busby and M. Coeckelbergh
374 Science and Engineering Ethics, Volume 9, Issue 3, 2003
with the ethics they hitherto espoused. We have suggested earlier that this could take the form of limits to prior generalisations. If, for example, one has a general utilitarian ethic, and encounters problems, like that with the airbag, where utilitarianism has been followed but a small group of product users do not receive the benefit of protection, then one could modify the general utilitarian principle. For example one could espouse the principle that the best course of action is utilitarian unless there is a group of people who are positively harmed by the action. Or the principle could be stricter, and insist that no actions which provide protection differentially (even though no-one is positively harmed) is acceptable. The former seems more plausible than the latter. There is a danger of moral relativism here, of modifying every universal principle until it fits every situation in a different way.
Second, it seems to us that some concern with ascribed obligations should be introduced into moral maturity models. There are various models of moral maturity in existence – models that are both descriptive of moral development, such as Kohlberg’s, and models that are essentially normative. A normative model used for teaching is, for example, reported by Spier.18 This refers to qualities like the capacity to make a moral determination that is above personal values, and implement a moral decision in the face of forces that militate against it, but it does not refer to moral imagination or empathy. Whether or not it is a descriptive dimension of maturity, there is a good case, based on our earlier arguments in favour of ascribed ethics, that moral imagination and empathy should also be a dimension of normative maturity models. It should be used to judge the extent to which engineers are morally developed.
Third, it seems quite likely that being confronted with others’ expectations could help engineers find imaginative solutions, not just be imaginative about the consequences of their decisions. Whitbeck19 has observed that it is misleading to characterise the ethical problems of engineers as ‘multiple choice’ tests in which one is simply choosing among a small set of pre-defined options. It is more realistic to view ethical problem solving as finding options in the first place. If an engineer were designing an airbag system and had to think about its discriminatory properties (s)he might start to think about what physical constraints have this discriminatory consequence, and might be able to ask how such constraints can be circumvented. There is plainly no guarantee this would yield new possibilities, but perhaps it could. There is a sense in which knowledge about ascribed obligations can be unfavourable, in making the engineering job even harder, but also a sense in which it can be favourable, being a stimulus for thinking beyond the normal run of concerns.
Aren’t engineers part of the ascribing group? It is important not to differentiate too definitively between engineers and the people (the ‘public’) who sometimes bear risk arising from engineers’ decisions. Engineers are product buyers, users, maintainers and disposers in contexts other than their work. They are fully socialised as a rule, despite their image. They would probably also be uncomfortable with compartmentalisation20 and would not deliberately hold one, general moral expectation of designers at home and another at work. Nonetheless it seems to us that compartmentalisation is probably true some of the time, not least
The Social Ascription of Obligations to Engineers
Science and Engineering Ethics, Volume 9, Issue 3, 2003 375
because much human activity is concrete, and a response to particular situations, so is not entirely subject to generalisation and being tested against general rules of conduct. It also seems to us that the moral case against compartmentalisation is not completely compelling. If one is inconsistent in one’s principles there is an evolutionary advantage, in the sense that this allows one to see how different principles work out, provided that one knows that one has adopted different principles. There is also an argument that you cannot perform as well as possible as an X if you cannot subscribe single-mindedly to the values of role X but instead try to take account of all possible perspectives on the actions of an X. Substantial engineering successes would never have come into being if the engineer had not ignored the fact that if he or she were someone who could be harmed by the enterprise then he or she would have objected to it. This seems to us to be a weak argument because, as well as permitting great engineering successes, it would also permit great engineering failures. But the point is that compartmentalisation has certain goods as well as bads, and probably exists to some degree. Therefore the idea that engineers could learn something from ascribed ethics is not vitiated by the fact that in some contexts engineers are also users of engineered artefacts.
Limitations of the work Our treatment of engineers making decisions that create risks for others has dealt with the engineer largely as an individual, although an individual who works in a formal organisation. It has ignored the role of various cultures – for example at the level of firm and profession – and the role of social constraints. It has also concentrated on the engineer as risk creator and ignored the situations where engineers face moral difficulties in the risks they observe others creating. The procedure we adopted to explore the responsibilities that members of the public ascribe to engineers was also limited in reach. It looked predominantly at the ascriptions of those who had recently suffered some harm, and it looked only at the way journalists had written about these. It provided a way of generating possible ascriptions rather than producing a definitive empirical study of them.
