July 19, 2020
Experts need a hippocratic oath.
If people would understand that they are asked their opinion on specific things, not to "help" those they think too stupid to take care of themselves. Busybody do-gooderism is at the root of so many of our problems. If you are an "expert" stick to what you actually know. Don't hedge your bets "just in case".
Anyway, this article is good. A brief snippet:
Perennially, proposals are made that economists, data scientists, and other individuals with influential knowledge sets should, like medical doctors, have to take a Hippocratic Oath: an oath upholding fundamental ethics.
While the original Hippocratic Oath did not, as it does now, require physicians to "First, do no harm,†the modern reach of technically-skilled elites via the media and policy should unquestionably bring that dictum.
A Technocrat’s Oath would not be restrictive of the mere practice of model building, running simulations, or other present day vaticinations. Rather, it would come into force where and when the forecasts of those methods are passed to policymakers. For one thing, at least three requirements should be met; and all preferably publicly:
- At least three major sources of inaccuracy– whether found in deficient sources of data, questionable assumptions, or estimates of sensitivity – must be expressed clearly – must be expressed clearly before and after a prediction is made;
- A discussion of statistical inference must be accompanied by a discussion of causal inference (preferably an interdisciplinary one); and,
- Any prognostication which does not prominently feature uncertainty as a growing factor over time must be summarily discarded.
If professionals whose judgment can directly impact one person are bound to a strict moral code (in addition to bearing legal and reputational risk), shouldn’t technocrats informing the highest levels of government face considerably more stringent standards of practice?
Perhaps another of the many effects of telling two or three generations of young people that a university degree is of unquestionable value is an undue veneration of the expert class. Many of our grandparents and great-grandparents, who had a formidable assemblage of wisdom but only a fraction of our formal schooling, were naturally doubtful in the face of boundless pessimism or optimism – especially when offered free of charge by bureaucrats from lofty heights of the Federal edifice. It was the people who lived through the Great Depression who offered such succinct quips as "You get what you pay for (and less),†and "It’s always darkest before the dawn.â€
Whether by today’s standards or a medieval one, at a certain level of influence, unqualified prognostication is either ignorant, irresponsible, or deceptive. Pithy epigrams alongside basic skepticism, knowledge of personal hygiene, and a propensity to self-isolate in the face of illness have superintended the human relationship with microorganisms more than massive agent-based models ever will.
A tip of the porkpie to E. Calvin Beisner.
Posted by: Timothy Birdnow at
09:58 AM
| Comments (2)
| Add Comment
Post contains 479 words, total size 4 kb.
Posted by: Dana Mathewson at July 19, 2020 02:25 PM (3ESpR)
What needs to happen is that "expert testimony" used to formulate public policy ought to be open to legal recourse; you should be able to sue guys like Fauci or James Hansen for misinformation. It would be hard to prove, but it would at least help keep them at least on planet Earth. If Michael Mann, who is funded by the U.S. taxpayers at Penn State, can sue people like Mark Steyn or Tim Ball then the opposite should be doable.
There should be a review board that examines some of what is done. Of course, the Left will merely stack the board and then use it as a bludgeon to beat more conservative scientists...
The problem will not be solved until the way science is funded changes, and until governments quit using science as a dodge for what they want.
Posted by: Timothy Birdnow at July 20, 2020 08:46 AM (8sgLQ)
37 queries taking 0.2251 seconds, 185 records returned.
Powered by Minx 1.1.6c-pink.








