The choice of sample size and the method of maximum expected utility—comments on the paper by Lindley

This commentary deals with the choice of utility function and, briefly, the role of coherence, in the context of the choice of sample size. I suggest, through the description of two applications, that one can be a Bayesian without necessarily using utility theory. The paper also provides an example of a sample size determination problem where a maximum expected utility approach is equivalent to other Bayesian procedures.