I’ve had this book on my radar ever since The Literary Escapade talked about it (the site is currently private but I’ll link to the post in case it ever comes back) but somehow, it escaped my notice when I was doing my binge reading on social media! But it’s never too late to read a book so I finally grabbed this one.

The Chaos Machine is a systems-level look at how social media platforms’ drive for increasing engagement don’t just increase polarisation and encourage extremism, they may be creating these views too. That is, essentially, the book in a nutshell. If you’re familiar with dark design, the controversy over algorithms, the problem with filter bubbles, the problem of moderation, and how our attention has been changed (there’s been many excellent books on these issues, also cited here), then a lot of the book won’t be new to you. In fact, I thought the first few chapters covered old ground and a lot of it wasn’t new to me.

However, I appreciated the fact that Fisher took the effort to look at how social media platforms have been affecting countries outside of the US. In particular, the chapters on Myanmar and Sri Lanka were eye-opening and horrifying, a clear indictment of how platforms like Facebook’s refusal to take action had very real consequences on minority groups. I wonder if this inaction is caused by the fact that Facebook was willing to subsidise the Facebook experience to gain a monopoly in the country, but because the countries weren’t rich enough for advertisers, did not want to actually pay for moderators or have a dedicated local team.

In general, I thought that The Chaos Machine gave a good overview of the problems with the social media platform. However, during the second half of the book, I couldn’t help but think that the discussion might have been richer if the following issues were also considered:

• The role of bias in traditional media platforms and if/how that leads to a reactionary response on social media. One case study Fisher examines is the rise of anti-immigrant sentiment in Germany and mentions a case where an immigrant accused of groping a teenage girl was turned into a story that a group of refugees had raped an eleven-year-old girl. However, I noticed that in his examination of this problem in Germany, including the superposters who spread fake news, he never covered the 2015-2016 New Year’s Eve sexual assaults in Cologne and other areas of Germany. How would the coverage of news like this or the Rotherham child exploitation scandals reduce the trust of people in mainstream outlets and drive them towards more extremist views or make them more inclined to believe fake news? That’s a question that I never had answered; instead the book seems to suggest that people are led down a path of fake news towards extremism, rather than perhaps having a trigger in the coverage of something they felt was unfair.
• The difference in community behaviours between different platforms. The four major platforms covered are Instagram, Facebook, Twitter, and Reddit. However, I think that there wasn’t much differentiation made between how the platforms are run and the norms of their respective communities. For example, reddit has many unpaid moderators, which influence the discussion on the site, as well as paid staff. How would discussion then differ from platforms like Facebook, which outsource moderation to third-party companies?
• Including non-American social media companies. I would have been very keen to read about the impact of Tiktok, and perhaps a comparison of TikTok USA and Douyin in China. Or perhaps a comparison between Douyin/Xiaohongshu/Weibo in China and Facebook/Twitter/Instagram/Reddit in America to see if there are similar issues, how the platforms deal with it, etc. It would also have been a good place to further explore the relationship between governments and platforms.
• Discussion of how American platforms influence the discussion in more Western-facing countries. Fisher discusses in detail how social media works in other countries, but those are cases in non-English speaking countries. I was also interested in how social media helps American discourse influence the political and cultural conversations in non-American contexts. For example, the concept of “Chinese privilege” has been introduced into Singapore and is the source of a lot of confusion – many younger folks seem to use it in a manner similar to “White privilege”, despite the fact that we do not have the same history as America, while older folks tend to compare us to our neighbour and their explicit privileges for the majority group which results in a lot of cross-talking.
• More discussion on the role of platforms and perhaps even ISPs as free speech hosts. I’m in a country that actively moderates hate speech to prevent another religious/racial riot, so while I do not have an issue with platforms doing the same, I was intrigued by the mentions of platforms’ relationships with government influencing their moderation activities. Fisher gives us very obvious examples of speech that should be curbed because of their devastating real-world effects. But what about speech that governments disagree with? That individuals disagree with? How do we avoid the filter bubble effect and allow for robust debate without tipping into hate speech? This was a question that was touched upon quite a few times in the book but never discussed in sufficient detail for me.

These were just a few of my thoughts which arose while I was reading the book.

Speaking of examples, since I mentioned a few earlier, I think that overall Fisher tried to be fair with his examples. I’ve seen him cite the Blood Heir controversy in a series of others on how people on the left can also jump to conclusions. However, the horseshoe theory (i.e. that the far left and the far right are more similar than you might expect, for example, see this article on anti-semitism in the left since Oct 7) doesn’t really come into play here, which is a pity because that would have been another excellent launching point for a discussion on what the role of platforms is to be. How can they be “neutral” in the face of hate speech on both sides and considering that algorithms can be written with unconscious bias in them (here’s an article that talks about this)?

I know my review talked a lot about what I wished to see in this book, but I have to emphasise that it’s a good introduction to the systemic problems that social media has (and is a good encouragement for me to continue staying off them!). I actually think this would be a great book to read as a book club since there’s lots to discuss, or perhaps with the other books on social media like Get Rich or Lie Trying.