Second in the series: old course essays, revisited. The first covered Dennett’s intentional stance. This one goes a year earlier.
A philosophy essay from 2008 analyzing Chapter 12 of Daniel Dennett’s Darwin’s Dangerous Idea. The chapter is called “The Cranes of Culture.” The assignment was to engage with Dennett’s treatment of memes as an extension of natural selection, drawing also on Dawkins’s The Selfish Gene. The essay covers the basics of meme theory: Dawkins’s argument that ideas can function as replicators subject to something like natural selection. Memes spread by leaping from brain to brain, compete for cognitive resources, and persist or die based on factors that have nothing to do with their truth or their benefit to the host. My argument was that Dennett’s account of cultural evolution is philosophically serious, and that resisting it usually involves smuggling in exactly the kind of human exceptionalism that Darwinian thinking has otherwise discredited.
The core tension Dennett identifies is still real. If memes are replicators subject to selection, then successful memes are selected for replication, not for truth. And the observation that a meme’s fitness and its contribution to human wellbeing can come apart is, if anything, more obvious now than it was in 2008. The essay is too faithful to the source, though. I reconstruct Dennett’s argument in careful detail and engage critically only at the margins. The section on whether memes can be identified syntactically barely touches the real difficulty: that “meme” might be doing too much work as a concept to be analytically useful. I also hadn’t yet figured out what I actually thought, as distinct from what I could competently explain.
The meme framing turns out to be genuinely predictive. In 2008 I was writing abstractly about how memes compete for attention in ways that favor replication over quality. Social media is a meme selection environment, and we’ve been watching what gets selected ever since. Dawkins’s passing observation that a best-seller list increases sales of books on the list simply because they’re on it is now a description of algorithmic recommendation at scale. There’s also the AI angle: Dennett argues that our minds are in part artifacts of the memes that shaped them. A version of that applies to language models, whose outputs are shaped by the distribution of text they were trained on, which is a record of which memes won. Whether that’s a useful frame or an overextension, I haven’t settled. But the question of what propagates versus what’s true runs through AI sycophancy as digital priesthood and Interface friction and critical thinking for reasons that connect back to this.