Opinions

Quite a number of articles, interviews and books are shared in group chats and forums I’m part of. This non-fictional content can range from business best practices to macroeconomics to geopolitics. Typically, the person sharing the piece of content used it to form an opinion about the world, and wants to share that opinion and its justification with the group. This opinion sharing is sometimes just phatic or for peer social support. But a lot of the time the opinions are presented as models of the world that should be adopted because they are more useful than the status quo.

I try to be deliberate about how I form a new opinion since being rigorous with my opinions will improve the communities I’m a member of. If we’re precise about how we think and what we mean, we enrich the communal pool of language and meaning. Words matter. What we say matters. How rigorous we are matters.

So I’m often explaining why I don’t form the same opinion someone else does from reading an article and why I recommend disregarding the source material since it isn’t robust.

It’s not that I think the viewpoints presented are necessarily wrong. It’s just that I’d rather have no opinion about a topic than fool myself with one that sounds “truthy” or because it confirms my priors. There’s a lot of content out there so I can always wait for material that’s more carefully reasoned before I form an opinion, even if that takes awhile. I don’t need to have an opinion about everything anyway. It’s enough to have a few key rigorously-formed opinions about what I need to know.

“The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge.” — Daniel Boorstin

How I evaluate content

This note lays out three checks I use before allowing a piece of content to form the basis for a new opinion.

A: Epistemic check

Words are the raw material we use to construct and share opinions. But words are leaky abstractions. What we write and say can only be an approximation of what we’re trying to convey. There will always be gaps between what is meant and what can be said.

To complicate matters, our individual sense-making apparatus is also lossy. Our consciousness needs to provide a stable interface to a dizzyingly rich sensory world. Due to various limitations, shortcuts and biases, we end up creating a subjective social reality that deviates from objective reality. And so there will also be gaps between what is and what can be known.

All this means there are many pitfalls to navigate when creating a new opinion about the world. The epistemic check is to see if the author understands the limits of how they make and share knowledge. Specifically I’m looking for authors or speakers who have:

1. Humility that shows they they understand the bounds of how much they can know and the fidelity of their opinions. They Steelman, i.e. they try to understand and address the strongest form of the opposite viewpoint even though it’s in conflict with their own opinion. They understand that all they see is not all there is.

2. A tendency towards concave worldviews. If the author assumes they know everything there is to know about all viewpoints, they will push you strongly towards their own position. Instead we’re looking for some acknowledgement that the best path might be a mix of multiple viewpoints.

3. A sensitivity to how they might be misunderstood due to the limits of language and so include as much context and nuance as they can. Our modern media environment gives content a boost in distribution and engagement if it removes nuance and adds conflict. A good sign is if the content creator is willing to handicap their own reach to help the audience get a better understanding of the world.

B: Fact check

If facts or past events and speeches are used as the basis for an opinion in the content, these can be used to gauge the author’s impartiality. If the event is interpreted inconsistently from other witnesses, it could be a sign that the author is being influenced by their bias.

1. If a speech is quoted, try to find the original transcript and check if the quote is taken out of context.

2. If an event is referenced, check if every aspect of the event was considered by the author or if parts of it were selectively excluded to skew the spirit of what actually happened.

C: Believability check

All opinions are inadvertently going to make some assumptions about the underlying domain that the reader needs to take on as a leap of faith. A question to ask before doing this is: why is this author believable in this domain? Since the ultimate arbiter of an opinion is the real world, I want to get a sense of what reality thinks of their opinions. What actual outcomes has the author achieved from their actionable insights.

1. The best case is when I myself have personal experience with the domain. I can check if their claims are consistent with my own observations, experiences and outcomes.

2. If I can’t rely on my own experience, then doing a check on the author’s background gives me a sense of what they’ve accomplished already. I can also look to see if they are respected by other prominent practitioners in the domain.

Skipping content that takes too much effort

Trying to validate an article via these checks will require more time and effort from me as the reader. If I get any sense that there’s a lack of epistemic awareness or the author is colored by their bias, I’m fine with stopping right there and foregoing the time I’ve already sunk. I’d rather skip a piece of content mid-way through than spend more time just to form an opinion of it or figure out if it’s “wrong”.

It’s worth re-iterating that if some content fails the checks or is abandoned mid-way, it’s not that I think the opinion presented is incorrect or not useful. It just means that I’d rather ignore the content and hold off forming an opinion based on it since I can wait for something better to come along.

“The trouble with the world is not that people know too little; it's that they know so many things that just aren't so.” — Mark Twain