Yes, Celebrities Really Do Die in Threes

And with a dash of linguistics and psychology, it’s simple to see why.

I’ve scoured the web to find the history of the Celebrity Death “Rule of Threes,” and by golly, this one particular myth doesn’t seem to have an origin story. You know, that pop-culture curse that dictates how celebrities are doomed to pass away in groups of three, no more and no fewer. A smattering of headlines going back into the 1930s and 1940s tell tales of tragedies and catastrophes happening in trios, but no links to the celebrity folklore that (somehow) we all know as common vernacular today.

Left: Arizona Independent Republic on Friday, December 2, 1938. Right: The Lowell Sun on Friday, July 12, 1940. Completely perplexing that I looked through news archives and could only find these two data points. Hmph.

The “Rule” most likely transitioned into the realm of celeb culture when, according a Washington Post story on the subject, “Buddy Holly, Ritchie Valens and the Big Bopper crashed and died more or less simultaneously in an Iowa cornfield on Feb. 3, 1959.” Still, even if this seems to be the earliest modern occurrence of the Rule in action, only in hindsight does it seem to align with the type of foreshadowing omen that we’ve come to expect today, rather than a freak tragedy of the time.

Yet despite having such an inexplicable appeal to the popular culture fandom, and still with no clear point of origin, it’s common for news outlets to take to the airwaves and claim from the rooftops: “Celebrities Don’t Really Die in Threes…”

Yes, they actually do, and it’s not because your brain is tricking you. It’s because the Rule makes no sense in the first place.

The Psychology of it all…

First and foremost, yes we’ll acknowledge the mental gymnastics in the room. Our brains do really like to search for patterns in the world — it’s called apophenia and it’s used to explain a whole slew of phenomena, from conspiracy theories to optical illusions, and even why clouds look like bunnies. There’s even a more specific term, “triaphilia,” when looking for patterns in threes.

See, this article does have science in it.

Amazingly the term was coined as recently as the 1980s, just around the time when the “Rule of Three” terminology started to become popular-enough in text for Google to recognize my search query. Thanks, Google.

NGram, you’re the greatest.

So the terminology beneath the modern-day omen all began to coalesce around the 80s, even if the lore and superstition had persisted almost invisibly throughout our spoken culture for centuries prior. Celebrities are eye catching, especially as headlines, so it was just a matter of time for the pieces to fit together.

But all this evidence seems to be pointing towards the common ammunition used to “debunk” the rule as a figment — simply our brain seeking patterns in a chaotic world — which is the main cornerstone of articles like this from the New York Times claiming to finally get to the bottom of the mystery. It’s all junk, and to show the proof in the pudding, we need to parse out the essence of the rule itself. Just briefly, at least.

The So-Called “Proof”

For starters, let’s look at the methodology that the Times uses to prove the adage incorrect:

We defined “celebrity” as anyone whose obit ran at least 2,000 words, roughly two-thirds of a printed page when photos are added… Since 1990, 449 such people have died. In 75 cases, two of them died within three days of each other. But in only seven cases did three of them die within a five-day period. According to my colleague Boris Chen, a statistician, this is about what you’d expect by random chance.

A 2,000 word obituary defines celebrity status? That’s a strange assumption on which to base an analysis as significant as the Rule of Threes. Life and death hang in the balance, and yet their definition gets drawn at a seemingly arbitrary point in the sand.

Amy Winehouse was omitted from the New York Times survey because, according to Curiosity.com, her obituary happened to fall short of the cut-off.

It’s one of the most glaring issues with attempting to assign any legitimacy to rules around celebrity death — or celebrity anything for that matter. Neither dictionaries nor wikipedia can offer a universal definition of “celebrity status,” which is understandable. I’m glad our bandwidth for linguistics research is being used on for more pressing matters.

Each and every one of us defines celebrity differently — it could be an icon in your own country, or niche interest group, or someone who impacted you more that they did the person writing an obituary. Take this (random) actual 2010 thread from Escapist Magazine as a good example:

User 1: As you may (or may not) heard, Gary Coleman and Dennis Hopper died recently… The purpose of this thread is this: who will be next?
User 2: the writer of thundercats died yesterday too.
User 1: There you go. The circle of three is complete.

Honestly? Sarcasm, humor, and Internet threads aside, this point is spot on. One cannot define celebrity in a universal capacity — over any given week, there could be one, three, ten celebrity deaths depending on who you ask and how you define the word.

A World Without Parameters

Ah, but there’s another rub. Why just a week?

That Times article casts their net at five days; other sources online wait seven, and others like this article from ABC News just go for broke with the idea that “Jimi Hendrix, Janis Joplin and Jim Morrison all died with weeks of each other in 1970.” Hendrix died on September 18 of that year and Morrison on July 3, 1971.

That’s 288 days, ABC News. C’mon.

Even in more reasonable cases, an interesting trend begins to emerge. Two celebrities die, and we find our pop culture enthusiasts asking “When will the third one drop?” Gross…but also, that defeats the whole point! Without a set duration of time for the Rule to take effect, of course we’re going to confirm our preconceived assumptions about omens and folklore. There’s no ticking clock and no definition for celebrity. Nothing matters anymore.

So yes, the Rule is real — decades (maybe centuries) of undiagnosed apophenia turned folklore turned adage of the Internet-age that can be neither proven nor disproven, caught in limbo until the end of time.

Pointless Conclusion: Do Celebrities Die in Threes?

Any argument that makes a claim of either proof or disbelief can define an arbitrary span of time and cherrypick celebrities to show, using the same data, that Rules both exist and do not exist all at once. It’s madness! I could claim a “Rule of Eighteen” theory and showcase that celebrities always die in 18s…but over the course of six months. While totally asinine, it’s also impossible to prove wrong.

All these articles I’ve referenced above are cut from the same cloth, all doomed to resurface on the web every now and again, often amidst tragedy. But given the apophenic nature of our squishy human brains, I understand why.

As I referenced above, though, the trend worsens when articles make claims like: “These two celebrities have just died, so who’s next?!” Looking at you, New York Times in 2014:

NOT MY GRAPHIC!! This comes from the New York Times. I’m not so depraved.

That’s not how this works! Either pick a fixed timeline and lower your expectations for celebrities, acknowledge that you’re expanding your window to “wait” for the next high-profile tragedy, or accept the reality that it’s all clickbait and you can’t claim to realize some parameters while tossing others to the wayside.

The “Rule of Threes” is just as real as it is fake. It has no beginning, no end, no truth, and no falsehood. I’m okay with that, but so long as our minds will seek patterns in the world, we just need to acknowledge the absurdity of it all. You can’t disprove something that doesn’t exist.

Except bigfoot, I disproved that already. You’re welcome.