Determining whether a breach of fiduciary duty has occurred is a science. It is an organized way of gathering and analyzing evidence about the behaviors and methodologies of those who make choices. By observing and de-constructing the choices one has made, which reveals the way one thinks or the way one doesn't think, it is a process that evolves during discovery.
As one collects and organizes information in some logical way, patterns and connections that are prerequisite to outcomes begin to appear. Potential breaches of duty cascade as poor methodologies tend to breed outcomes that lead to additional poor methodologies and outcomes. Rather than starting with pre-determined beliefs and precedence, Fiduciary Science is strictly evidence based.
It seems like the most useful way to view fiduciary duty, in prudence and loyalty, is approach it as system composed as parts and processes that may interact and perform interdependently. There should also be a sense as to where there should be an absence of interdependence. The perfect example of this may be the interdependent play that sometimes occurs between administrative/record-keeping costs and the investment returns that participants receive in retirement plans. This relationship, however, is a methodological choice which by birth and review is supposed to be for the exclusive benefit of the participants of the plan free of conflict of interest. A fiduciary must consider if creating an interdependence between plan costs and investment returns, which increases fiduciary burden, is the best choice relative to the alternatives. Fiduciaries should know that a loss in returns reduces gains that should be compounding which decreases the likelihood that a performance outcome will benefit the participant.
ERISA, so it seems, inadvertently recognized "uncertainty" as no specific methodological guidance is given only mandates of intention. It makes perfect sense because creating methodologies is a subjective journey and, like any other science, proves nothing in absolution. In fact, requesting clarification or guidelines in behaviors, for an endeavor that can only be subjective, may indicate a lack of subject matter expertise or an attempt to justify self-interest which is a clear breach. Oftentimes, self-serving economics, predicates intention which "throws a wrench" into Fiduciary Science. That too is a breach that can lead to poor outcomes.