Why the AMERICAN confidence dial goes to eleven... or how to do  “strong opinions, weakly held” correctly!

Over my last 6 years in the USA, I've found an interesting trend among many debates with my younger American colleagues - say it confidently and never back down. In peer discussions back at Cambridge University I fondly remember when claims were made on limited knowledge – the anticipation that debate would surely lead to an interesting exploration, an ebb and flow of opinions and joint learning, of the new topic. However, now too often the American pattern is familiar - with full volume, confident claim, and brisk certainty, an American will state the answer to whatever the question... yet often they don't see debate as an exploration, but rather competition to enforce their view. While I am a huge fan of the American philosophy of “strong opinions, weakly held” [Paul Saffo, 2008], I have often found the "weakly held" bit often seems to be the first casualty when knowledge is thin. Yet  “strong opinions, weakly held” is also a style I try and foster in myself - so what can I learn for my discussions with Americas, why does sometimes go wrong here, and what should I focus on to ensure my own correct use of this paradigm?

When thinking what might be going on with my "Average American Stereotype", I believe I see three ingredients recurring... First, classic overconfidence effects. People who know least about a domain tend to overestimate their competence [Kruger & Dunning, 1999]. Add the planning fallacy and availability heuristics and confidence hardens quickly [Tversky & Kahneman, 1974]. Second, in its culture the United States scores as a loose, individualist society that rewards assertion and rapid self-expression, not tentativeness [Gelfand et al., 2011; Markus & Kitayama, 1991]. Third, incentives in such and large well connected country - with public discourse typically preferring clarity and speed. Student's seem to learn at school that nuance looks like hedging, so opinions solidify for presentation to an audience - with limited reward for backing down when disproven. All of these factors lead to challenging discourse, where people (ironically, especially smart people) have learnt how to justify, rather than correct and reanalyse their views.

However various studies, such as Tetlock’s long-run study of expert forecasting have shown that the most certain voices also tend to be the least accurate. "Foxes", who update, beat "Hedgehogs", who double down [Tetlock, 2005]. If we forget the “weakly held,” we lose the ability to course-correct as evidence arrives - and very quickly find ourselves defending our pride rather than accuracy. Interestingly, it was this same domain of forecasting that drove Saffo's philosophy in the first place: "the fastest way to an effective forecast is often through a sequence of lousy forecasts. Instead of withholding judgment until an exhaustive search for data is complete, I will force myself to make a tentative forecast based on the information available, and then systematically tear it apart". Or as he later summarised for HBR: "come to conclusions quickly, but do not get attached to them... Be your own worst critic of your own forecasts. Otherwise someone else will."

From this, several lessons arrive for my own personal conduct -  while confidence is good, its important to treat every claim as a provisional design. Internally know the assumptions and the failure modes. Specify (internally, but ideally to others too) the evidence that would change your mind - and then actually update your opinion if this arrives. I believe with intent, and effort, I can keep the energy of American argument while turning the confidence dial from eleven to an adjustable knob.