I found your history very interesting (I was familiar with much of it but I don't think I've seen so much collected in one place) but I had some issues with your conclusion, mostly because I don't really see the phrase "cargo cult" or the verbed form "cargo culting" to be inherently pejorative. I think the concept of someone going through the motions without a necessary understanding of their purpose to achieve the desired effects is very useful one, especially given the ever increasing layers of abstraction that exist in our society.
Have you thought about an alternative concept or word that describes this phenomenon that could be used instead?
I read the whole history and it only affirms my belief that the phrase is spot on. I'm not concerned of its pop culture origin, etc.
Look, these people were seriously believing some ridiculous junk. Most Europeans once believed some ridiculous junk. There's millennia old ridiculous junk still being believed. It's all "cargo cult".
> Have you thought about an alternative concept or word that describes this phenomenon that could be used instead?
I think "magical thinking" would be an appropriate term for what Feynman characterizes.
However, one of the post's important points is that we're not even using Feynman's mischaracterized explanation of cargo cults: it's become a generic negative descriptor for anything the user considers insufficiently justified, even if the underlying rationale is not "magical."
Magical thinking is a bit different than cargo-culting. Magical thinking is the belief that unrelated events are causally connected. For example, "I survived the car crash because I had my lucky charm in my backpack".
Cargo-culting is the belief that specific best practices which are causally connected to an outcome in one context will produce those outcomes in other contexts where the chain of causal reasoning no longer holds. For example, "I survived the car crash because I was wearing a seatbelt. Now I'll install a seatbelt on my bicycle too."
Cargo-culting is an important concept in its own right in the tech industry, because those best practices do get often blindly shared, recommended and even enforced into codes and standards, even when the context that made them a good idea is lost. Without a concept like "cargo-culting" to label the fallacy, it can be hard to argue against that proposal, because the side recommending the change has lots of out-of-context data in their favor. For example, "car-crash survival rates are much higher when drivers are wearing seatbelts. Therefore, we're now requiring bicycles and motorbikes to have seatbelts."
Clearing an airstrip isn't magical thinking though, you'd need to do that for planes to land. It's just not sufficient, because the planes also need to want to land there. I don't think magical thinking covers that same concept.
It appears that there are some people (like you and the post author) that are constantly confronted with someone screaming "cargo cult", which has to be exhausting. But it's absolutely not the world others live in. In my world, it comes up every now and then (I'd guess every 3-4 months), and I've experienced it multiple times that I mention it and the other person hadn't heard it before but absolutely LOVED the term after an explanation because it describes certain behaviors so well.
The magical thinking in the case of the airstrip is the lack of a causative connection: you need to clear the airstrip for the plane to land, but clearing an airstrip does not make a plane land.
Or in other words, magical thinking doesn't imply the lack of a conceptual connection: airstrips and planes landing are definitely conceptually connected. Magical thinking is the drawing of illogical causative connections from conceptual ones.
(I don't run into "cargo cult" that often. But I think TFA is a great writeup of why it's not the best term; as engineers, I think we should aspire to use the best terms available to us.)
Magical thinking explicitly lacks the connection -- that's why it's magical after all. Jupiter and Mars are aligned, therefore I missed my bus, that sort of thing.
That's not what I usually encounter. I see people imitating something they've seen others do (who appear to be successful) without understanding the full concept. Their efforts are in vain because clearing the brush for an airstrip doesn't make the planes land, and neither does one deliver projects more successfully by changing nothing except asking everyone to join a daily meeting each morning.
I always found the use of the phrase mildly racist and an easy low effort way to take someone down.
Thank you for the amazing thoroughness in your research. I just read aloud the entire article with my kid. So many tributaries of history and science to explore later.
Things that I referenced in our discussion about this article.
How slang, and low fluency spread from person to person can create divergent dialects.
The apocryphal story of cutting the pot roast to fit the pan. How things start true but get transformed through transcription errors. The main theme and the message may be retained, but the specifics get jumbled up.
Instead of just tearing down the current usage, you should have at least proposed an alternative that captures the metaphorical qualities that are being sought with the current terminology.
well written and interesting. although i really do love the technical deep dives.
i am fond of using the cargo-cult analogy, and invariably many people have not heard of it so the story is told and retold. i'm fairly happy that my usual descriptions of the phenomenon were much less inaccurate or exaggerated than they could have been; generally closer to the John Frum reality than the "pop-culture" one. not at all like mondo cane (which i was unaware of). for example, i've said something like "to this day, there is a cult in which members paint themselves USA 'uniforms' and march in military style with 'guns' made of sticks'" (which appears accurate). i completely missed, however, the pre-ww-ii "cargo cult" beliefs which add quite a different perspective.
unfortunately, i don't know if i quite agree with abandoning the metaphor. the literal Feynman quote is about science. we in engineering have co-opted the term and use it (when imho done correctly) in a Feynman sense. i describe it as an Feynman anecdote. but it is one with significant grains of historical truth.
i find the curated list of HN examples illuminating because it appears that 1/2 or more of them are using the analogy poorly, missing the point, or simply as a kind of slur. meta-cargo-cult if you will. it is as said: "is simply a lazy, meaningless attack". i agree that it is heavily misused.
but in the conclusion, this leads to an argument that i see as a bit of a false dichotomy. i don't agree that Feynman's central point was either "doing something that has no chance of working" or we (mis)use it as "works but isn't understood". when Feynman said "but it doesn't work" i think meant within the analogy it didn't work: the planes did not show up. i don't think that when applied to science or engineering it only applies to something that "doesn't work". i think it's very much more about the central fallacy at play: misunderstanding processes that are built to support the science as being the science itself. misunderstanding effects for causes. misunderstanding and generalizing specific observations where they don't apply.
i think Feynman's anecdote is close enough to the anthropological one and not really detailed enough to be considered wrong. it's factually true that john frum cultists do what they do. the reasons they do it aren't quite right in our stories, but clarifying all the anthropological history doesn't kill the analogy, it might even strengthen it.
to me, used correctly the analogy is describing a religious or cultish adherence to principles that are not understood, in the hopes of some desired affect happening. it's similar to affirming the consequent. the fact that real cargo cults developed prior to ww-ii in places affects the story telling, not that its a cult. the fact that it's dangerous and harmful to the adherents is a good point for the analogy. the fact that the cults developed partially as a result of decades colonial oppression and mistreatment is a better framing than "look at the dumb thing those ignorant savages did". the fact that the cult members are expending energy which harms them for reasons they do not understand is still the truth. i've certainly never been as glib as "US soldiers show up with their cargo and planes, the indigenous residents amusingly misunderstand the situation, and everyone carries on."
the points about it being insensitive are well taken, however. no doubt.
- certainly there's a large amount of misuse of the analogy. and these uses are misused whether it be relative to pedantically accurate anthropology, Feynman, or pop-culture variations. but people using an analogy wrong does not make the analogy wrong.
- i think it's fine to use an anecdote and an analogy to communicate an idea about a harmful phenomenon. the anecdote does not even need to be true at all. but in this case it isn't too far off, depending on the story telling. Feynman's short description doesn't seem as extreme as what is described as the "pop-culture" definition.
- it can certainly be told in a way that is very culturally insensitive. i think this could also be done in a more neutral manner, but it's something to be careful of for sure. certainly, sticking closer to the history would probably improve things, however this may be the achilles heel. (in other words, out of all the reasons for abandonment given, i'm most convinced by this one)
- the biggest issue, for me, left is this: what do you recommend replacing this with if we avoid it altogether? the imagery of religious behavior is a big part of what that analogy covers. and the ideas of observing something and then copying those behaviors to achieve a result without any real understanding.
anyhow, thank you for a very thought provoking article. i'm clearly not as good of a communicator as you are (or Feynman).
But the point is that the islanders were not mistaking effect for cause, but simply believing some wrong things.
For example, they were clearing the airstrips, not because they believed that doing so caused the cargo to appear, but simply to facilitate the delivery of the cargo if and when it came.
in some cases, but in others they may believe in a kind of sympathetic magic [1][2]. the John Frum's don't march around with stick guns and uniforms "simply to facilitate the delivery of cargo"
in either cases it's still highly apropos to the engineering analogy.
did you (hypothetically) add an unnecessary statement because you observed "good" programs doing it and believed if you did the same your program would be better? or because you thought you were facilitating something that was needed to be done to support something that isn't needed or won't happen? to me these are two sides of the same coin.
i've seen these kind of things (these are quick examples of the top of my head, not the worst things by far):
const char *str = "hello\0"; // make sure it's null terminated
if (ptr != NULL) free(ptr); // don't free if not allocated
