Highest Standards Aren’t Always Best in Ethics
No one wants low ethical standards, but it’s also a mistake to aim at the highest possible standards — at least it can be, depending on what you mean by “highest.”
See, for example, this useful piece on defence contractors, by Noah Shachtman for Wired: Pentagon Probe Will Review Every Darpa Contract
Since Regina Dugan became the director of Darpa [Defense Advanced Research Projects Agency], the Pentagon’s top research division has signed millions of dollars’ worth of contracts with her family firm, which in turn owes her at least a quarter-million dollars. It’s an arrangement that has raised eyebrows in the research community, and has now drawn the attention of the Defense Department’s internal auditors and investigators…
The story usefully points out that aiming for the highest possible standards of integrity can also cause trouble:
[A former director's] bright ethical guidelines had unintended consequences. If a company allowed an employee to take a sabbatical to join Darpa, the firm was essentially blocking itself from millions of dollars in agency research projects.
The result, of course, is that under the old rules the agency risked cutting off useful sources of expertise. That’s not to say that the old rules were worse. It’s just to point out that there’s a legitimate trade-off here.
There’s a very general lesson to be drawn from this. When thinking about ethics, the goal isn’t always to be squeaky-clean. The goal is to find standards that are high enough to merit the trust of relevant stakeholders, and to do so without sacrificing other, possibly-equally important, values.
Consider this graphic, which illustrates the challenge of choosing experts to make decisions. On one hand, we want people with real expertise. On the other hand, we want to avoid conflict of interest. That is, we want maximum expertise and minimal risk of bias. So the upper-left quadrant of this graphic is the sweet spot:
Note that what we’re looking for here is not the “highest possible” standard of integrity (i.e., the standard that implies the lowest possible risk of bias among decision-makers), but a system that makes the optimal tradeoff between risk of bias, on one hand, and relevant expertise, on the other. The point here is not that we’re trading off ethics and expediency. The point is that we’re trading off competing, ethically-significant values.
The point in thinking about ethics is not to aim at the highest standards, but at the best standards. We do enormous damage both to the functioning of organizations and to people’s willingness to talk openly about ethics when we talk about “high standards” in a way that comes off as unthinkingly pious.