Author: Kyle Polich

Millions of People Can't be Wrong

I've heard debates in which the fallacy of argument from popularity is used as if it was logic. I must admit, I'm tempted to use this one from time to time. When I haven't the resources or interest in verifying something, the fact that many others have arrived at that conclusion does seem like a reasonable way to adopt a position, even if it isn't strictly logic based.

I think some of that tempting line of thinking derives from an assumption that while I might not have time to check something, certainly other people do. If a large enoug population of people have set out to verify something and found it to be contrary to popular opinion, it seems some of those verifiers would be working to get the correct information out.

At the start, these people might be indistinguishable from the types of cranks that argue against fundamental science. The difference is, with truth on one's side, minds are more likely to be swayed. Science especially is well designed to achieve eventual consistency. Yet even within the scientific community and certainly outside it, it's still entirely possible for popular opinion to be deeply wrong. That's probably uncommon, but certainly possible.

With that grain of possibility, I see many believers in pseudo-science attempt to use the "argumentum ad populum" in debate. I recently encountered a UFO believer who shared an unsourced statistic with me: "20,000 Americans have seen a UFO since 2010. 20k people cannot all be wrong!"

I could craft a response from many angles. Most noteably, an object in the sky which happens to be unidentified (i.e. a UFO) does not mean that unidentified object doesn't have an ordinary explanation. We need not jump to the extra terrestrial hypothesis.

On this occasion, I thought it might be better to attack the logic more directly. I asked the person to think of all the presidential elections that occured since they were of voting age. For this person and many others, that includes both the election of Barack Obama and Donald Trump. I suspect the intersection of people that supposed both these candidates is a small subset. Even for those in that intersection, surely few of them supposed every single candidate in every race since they began voting.

That being true, in any election for a candidate won whom a person doesn't favor is an example of a case where millions of people "got it wrong". I presented this point with a bit of melodrama, and the person conceeded that line of arguement. Anecdotally, I think this can be an effective rhetorical device for steering people away from this logical fallacy.

Author: Kyle Polich

Kyle studied computer science and focused on artificial intelligence in grad school. His general interests range from obvious areas like statistics, machine learning, data viz, and optimization to data provenance, data governance, econometrics, and metrology.

Enjoy this post? Sign up for our mailing list and don't miss any updates.

Have a word to say? Propose a specific change to the blog post.