Couldn’t find the language – the positive counterparts of risk and hazards

Continuing my recent theme of the impact of language on ethics and decision making I’m presently writing a paper on the use of claims based on justice to object to new technologies such as human enhancement or synthetic biology. In the process of writing this paper I’ve encountered a rather odd gap in our language. It is the case when discussing future technologies that the harms involved are more properly thought of as risks (that is uncertain harms where the probability of the harm is predictable) and hazards (that is uncertain harms where the probability of the harm is not predictable). The same applies to the benefits of future technology but we don’t have the language to directly describe them thusly, ie there is no obvious counterparts to risk and hazard in regards to benefit. This seems odd given that benefits can be just as uncertain as harms.

And while we might think of this as just a linguistic oddity it may actually have an impact on decision making. For example in research ethics it is common to talk about the risks and benefits of research – but this creates

Any suggestions of appropriate terms would be gratefully received!

  • John Coggon

    Hi, David.

    I think Soren Holm and Tuija Takala dealt with this question very well in their paper in the JME a few years ago: “High hopes and automatic escalators: a critique of some new arguments in bioethics,” (2007) JME 33, 1-4:

  • franz_porzsolt

    you are right.
    When we consider harm there are some sitations where the outcome is quite certain, e.g. 99,999% and other situations where the outcome is rather uncertain.
    The first case is called “hazard”, the second “risk”.
    When considering benefit you can NEVER be sure that you will experience the expected benefit. Success rate of 99.999% are extremly rare. In any case there is a “chance” to find what you expect. Murphy's law is talking only about wrong but not about the favourable outcomes.
    The answer to your question could be that there is no reason that a language should provide an approriate correlate because the appropriate correlte does not exist.
    Best wishes