Paid Blog Post - The Epistemic Argument for Going Outside

Sorry for the delay in getting this out. Also I should have some more bookshelf tour videos soon :). It’s been a bit of a mad dash before Christmas!

——

Recently on my channel, I discussed the many and varied ways engaging with the world primarily online can distort our thinking by making us angrier, less patient, and more hostile to one another (importantly, this is a can, not a definitely will). However, something I did not have the space to go into there is that the internet is, in many ways, the ultimate skewed sample.

If you share your opinions online, or regularly post comments, or make posts of your own, you are in a small minority of users on most platforms. This is most pronounced on X/Twitter, where around 90% of the content is generated by just 10% of the userbase. Precise figures are disputed, and vary across platforms. Reddit has a fairly high lurker rate by most accounts, whereas over a third of Facebook users do post at least occasionally. The most comprehensive data I could find on this was this Gallup poll from 2022 (https://news.gallup.com/poll/467792/social-media-users-inclined-browse-post-content.aspx).

All of this means that if you are looking at social media, you are largely seeing the perspectives of a very small number of people. The particular breakdown will vary according to your past usage, but you know they are the kind of people who actively (and perhaps regularly) post their viewpoints on social media, and this automatically means they belong to a rather odd group. This becomes even more pronounced if you mainly look at the viewpoints of people whose job it is to put content online (including yours truly). That is an even smaller group of the population, and are subject to particular economic and social pressures such as the need to “stay relevant” and “maximise engagement”. All of which compromises their (our?) perspective on the world. No one, no matter how self-reflective, is immune to the incentives they swim in, and people online are no different.

Even among those who do not earn their living from “creating content online” are still influenced by the built in success-metrics of social media platforms, such as likes, shares, views, and subscribers. Humans are weird creatures, and will often be motivated by seemingly arbitrary things so long as they are socially respected in a particular context. My dad used to be a teacher, and kids were so motivated by his “Folley dollars” that he had to keep them in a box under his desk because children were mugging each other for them in the corridors. The tangible rewards available to those who collected these Folley dollars were pretty modest (I think the highest level of reward was a positive letter home, not something that you’d think would motivate 15 year old boys), and yet they pursued the faux-currency with relentless vigour. All of this is to say that we may think we are not motivated by such arbitrary metrics as likes, shares, and views, but we often are.

All of these factors, the skewed sample of posters, the influence of influencers, and the incentivising metrics, create what I like to call “opinion inflation”. In the fitness community, “physique inflation” refers to the ever-increasing standards that social media has brought on, because only the best physiques make it onto visual-based social media platforms like Tiktok and Instagram. This creates the impression that there are huge numbers of people with unattainably great physiques out there, when in actual fact we are just over-exposed to the top 0.0001% of them. Likewise “opinion inflation” refers to the phenomenon of only seeing the most attention-grabbing opinions out there. The uncertain and the intellectually cautious don’t get views, whereas the bombastic and the overly-sure-of-themselves do. Someone tweeting out a complex, nuanced, and fair critique of a country’s foreign policy will be unlikely to take off, except within a niche community, whereas a hot take that is provocative and witty, if ill-considered, will probably be engaged with far more, from both fanatics and haters, and so will spread further. Thus, the over-confident nutter makes it onto your feed, whereas the cautious thinker does not.

It has become cliche to say things like “the internet is not real life”. If nothing else I think it is over-hasty. The internet is now so intertwined with our daily existence that it can sometimes be difficult to tell where this “real life” is supposed to start. However, that does not mean that the content you see on your screen is going to be a good representation of the world beyond its four borders.

This is why you must go outside and talk to people. Any people. Random people. This is something I initially started doing last year out of grim necessity. I work alone, save for my lovely video editor who lives half the world away in the US, and I spent most of 2023 and 2024 in a slightly dingy flat, by myself. This was sending me absolutely stir crazy, which meant I would quite often leave my house with the sole objective of talking to a bunch of strangers. We would chat, I would ask their opinions about life, or politics, or what they were up to.

Of course, leaving the house and talking to the people in your area is also a skewed sample. It is geographically limited to the people who happened to live within walking distance of you. But it is skewed in a different way, and I think that is why it is such a good antidote to too much time spent online. There is not the same incentive structure that pushes people towards opinion inflation. The first thing I immediately noticed was that people’s opinions offline were often more moderate, and less confident, than the opinions I heard online. This stretched over almost every domain, from philosophical topics to politics to literature to whatever was that day’s news item. Even if their position initially seemed dogmatic, the fact we were chatting in person meant we could explore it a bit more, and oftentimes they would acknowledge far more nuance than I have seen almost anyone do online, possibly because they weren’t immediately bombarded with umpteen snarky comments.

My point is not that the people in your local area are representative of the world at large. My point is more that because the selection pressures are different, it brings the skewed sample of the internet into stark relief. And the residents of your local area are also far more representative of the kind of people that are actually, physically around you. The kind of people that, should you choose to, you will make friends with, have conversations with, and work with.

I also think it is a good counter to a mental habit that I have caught myself in, and I suspect is fairly common. If we spend a lot of time online, we begin to adjust our expectations for in-person behaviour around what we have experienced in the digital sphere. That is, we use our experiences with people on the internet, and assume things about how people in-person around us will behave. I think this is both unjustified and unhelpful. It is unjustified because the internet is a weird place, and while your local area will also be a weird place, it will be weird in entirely different ways. And it is unhelpful because the incentive structure of a lot of content produced online means that it will likely be angrier, snarkier, more shocking, and more arrogant than the interactions you have in person.

So, if you are looking for a New Year’s Resolution, try striking up a conversation with just one stranger per week. Not all of them will be pleasant, and not all of them will restore your faith in humanity. But as a whole, I do think they will help hammer home the incredible difference between online and in-person behaviour.

And even if you don’t decide to undertake this experiment yourself, I hope this brings some small comfort the next time someone is a total cock online.