the three emergences

june 17, 2026

“emergence” is doing three different jobs in popular science writing and only one of them is honest. the other two ride the first one’s credibility.

the honest version: a property at one level that is derivable in principle from properties at a level below, but easier to think with at the upper level. temperature emerges from average kinetic energy. fluid dynamics emerges from molecular interactions. the higher-level concept doesn’t add anything ontologically; it’s a tractability concession. we work in terms of temperature because doing the molecular bookkeeping for a coffee cup would take longer than the heat death of the universe. fine. nobody’s pretending we couldn’t, in principle.

the second version comes from philosophy of mind. strong emergence: higher-level properties that cannot in principle be derived from lower-level ones. genuinely new, not just computationally hard. consciousness, sometimes; semantic content, sometimes. this is a serious philosophical position with defenders. they take on the burden of explaining what “in principle irreducible” means and why the gap is metaphysical rather than epistemic. that’s a heavy lift and most people deploying “emergence” are not doing it.

the third version is the one that bothers me. the X emerges from the Y, deployed when:

“consciousness emerges from the complex interactions of neurons.” okay. so does that mean we have the derivation? no. so does it mean you’re claiming consciousness is in-principle irreducible? probably not, you’d want to leave open the possibility that future neuroscience explains it. so what does the sentence say? it says “i don’t know how consciousness happens, but i want to gesture at neurons being involved, and i don’t want to say i don’t know.”

“the murmuration of starlings is emergent and self-organizing.” biologists actually studied this. each bird tracks something like its seven nearest neighbors and adjusts heading. that’s a mechanism. you can simulate it. calling the result “emergent” is fine in the first sense, the flock-level pattern is easier to think with than the per-bird interaction rules, but the second word, “self-organizing,” carries extra rhetorical weight the mechanism doesn’t underwrite. nobody is organizing anything. there are local rules and they produce a pattern. “self-organizing” suggests an agent where there isn’t one.

“the economy is a complex adaptive system.” sure. now what? if the framing licensed specific predictions or interventions, fine. usually it licenses humility about prediction, which is a real thing, but you don’t need the complexity vocabulary to get there. “the economy is hard to predict because lots of people are doing lots of things and the system has many feedback loops” gets the same conclusion without the implication that there’s a science you’ve consulted.

the test: does the word do anything in the sentence besides label a hole? swap it out for “we don’t fully understand how” and see if the sentence loses content. “consciousness, by mechanisms we don’t fully understand, arises from neural activity.” that’s the honest version. it’s also less impressive-sounding, which is why people don’t write it.

the move isn’t to abolish “emergence.” the first sense is useful and the second is a legitimate philosophical position. it’s to notice when the word is buying you a sentence you haven’t earned. the rhetoric works because the first sense is real, and the third sense gets to borrow that reality without earning it. if you can’t tell which kind you mean, you don’t mean any of them. you mean “i don’t know.”

if it stayed with you, write to me.