In just over two weeks, Rafael Nadal will be making his final appearance at the Summer Olympics.
Widely acclaimed as one of the best tennis players in history, Nadal is also famous for his on-court rituals. Adjusting his underwear before each serve. Crossing the lines with his right foot first. Arranging his drink bottles in a diagonal pattern, each label facing his side of the court.
Nadal claims that "routines" like these help keep him focused. He denies being superstitious, but is he right about that? What is superstition anyway?
A brief definition
In his 2019 book Superstition: A Very Short Introduction, Stuart Vyse defines superstitions as beliefs that are inconsistent with science, practically useful, and culturally rare. ("Culturally rare" meaning rare within a culture, or rarely observed across the entire range of human cultures.)
To illustrate, consider the person who's afraid to stay on the 13th floor of a hotel. There's no scientific basis for their fear. They choose another floor anyway, believing this to be useful for preventing bad luck or disaster. According to a USA Today/Gallup poll, only 13% of Americans share this belief. (13% fear the 13th floor...a coincidence that may have some deeper meaning if you're superstitious.)
As for Rafael Nadal, his on-court rituals are also rare and useful. However, they'd only count as superstitions if he had said something unscientific about them (e.g., they give me power). It's not unscientific to say that doing such-and-such keeps me focused. All the same, Nadal has made some puzzling remarks, noting that his routines are a way of "ordering my surroundings to match the order I seek in my head." What does that mean exactly? Is there something unscientific here after all?
In contrast, Naomi Osaka, who has avoided stepping on lines and as well as the Melbourne logo, freely admits to being superstitious. During the Australian Open in 2022, she told reporters "I would say that most athletes have very strong superstitions. It could be one thing, it could be two things. For me it's definitely the lines and the logo." But is it superstitious if she doesn't explicitly say that stepping on lines and logos would undermine her play? Would she have to claim that the lines emanate some sort of negative energy? Is she really superstitious?
Deciding that any one person is more or less "superstitious" can be tricky. As you'll see, Nadal might be superstitious after all, while Osaka might not. It's clear at least that the gambler who thinks that today's a good day to bet on the number 8, or the person who assumes tarot cards can predict the future, are exhibiting what we would call superstitious beliefs.
What prompted this newsletter is a new meta-analysis on the causes of such beliefs. This meta-analysis tests a prominent theory of superstition and other kinds of unfounded, "irrational" thinking. It's also a great illustration of the strengths and weakness of statistics when we rely on them for guidance on how the mind works.
Why is this important?
Some superstitions are benign. Finding a four-leaf clover may not bring you good luck, but it doesn't hurt to think it will. Likewise, nobody's harmed by the absence of a 13th floor in some western hotels, or by the buildings in China missing a 4th or 14th floor. (In most dialects of Chinese, the word for "4" sounds like the word for "death" – same phoneme, different tone – and one way to pronounce "14" sounds like the phrase "will die" or "want to die".)
It's the divergence from scientific understanding that can make superstitions dangerous. The pandemic gave rise to a seemingly endless parade of disturbing examples. Scarecrow-like figures moving from house to house in Cambodia to ward off the coronavirus. Americans claiming that COVID-19 infections can be healed by tea and essential oils. A former student of mine from the U.K. who believes she's still warding off the coronavirus via positive energy.
We need to find ways to dislodge superstitions like these, or at least encourage people to reflect on them and remain open to other perspectives (e.g., open enough to get vaccinated). Perhaps understanding what causes superstition can help protect us against the most harmful ones.
Theories of superstition
The anthropologist Bronislaw Malinowski, living among the Trobriand Islanders of New Guinea, noticed that they engaged in superstitious rituals before fishing in the open ocean, but not when they fished a closed-off lagoon. In his 1954 book Magic, Science, and Religion, Malinowski commented that
"in the lagoon fishing, where man can rely completely upon his knowledge and skill, magic does not exist, while in the open-sea fishing, full of danger and uncertainty, there is extensive magical ritual to ensure safety and good results."
Though Malinowski was far from the first scholar to theorize about superstition, this anecdote captures many of the theoretical explanations that social scientists continue to explore:
–superstitions represent irrational, magical beliefs
–superstitions are learned from other members of one's culture
–superstitions are motivated by fear
–superstitions arise from a need for control or certainty
There's evidence supporting each of these explanations, and yet each one, by itself, doesn't tell the whole story. Studies suggest that highly rational people can be superstitious. Idiosyncratic superstitions, like wearing your lucky belt to a job interview, don't seem to be learned from others. And, if you believe that four-leaf clovers bring good luck, but you can't say exactly what good they do, it's not clear how this belief would make the world less scary or give someone a greater sense of control over what happens.
To illustrate what's missing from these explanations, I want to share a personal anecdote that's slightly embarrassing and yet typical, in the sense that a lot of people do things more or less like this.
I keep five of my daughter's stuffed animals from the time when she was a tiny person. (A bear, a giraffe, an orca, a pumpkin, and a brachiosaurus.) Keeping them is not superstitious per se, nor is the fact that I treat them gently so that they don't get damaged. But here's something I just realized: I always arrange them heads-up. After my oldest granddaughter tosses them around, I put them back in an upright position. (It's more comfortable that way, right?)
I've always done this, even though I'm sure that stuffed animals don't have feelings, much less care whether they're upright or not.
Now let's put my beliefs to the test. Can I walk over to the cabinet where I keep the animals, grab each one by the neck (except for the pumpkin, which doesn't have one), and turn it upside down?
No. I mean, I could do that, but it would feel uncomfortable.
What theories of superstitious behavior need to account for is the complexity of the mental processes that underly them, and the discrepancies that people like me exhibit between beliefs and behaviors.
One could speculate that I'm afraid of what might happen to the animals (or to me, or to my daughter) if they're not arranged comfortably. The problem with this explanation is that (a) I haven't noticed any fear, and (b) it doesn't explain why my behavior doesn't align with my beliefs.
Maybe I can't help the behavior? Some observers have suggested that Rafael Nadal suffers from OCD, because his on-court rituals are numerous and extremely fussy. The problem with this explanation is that for many people – including me – you can see superstitious behavior but no OCD-like tendencies. I'm not anxious about arranging the animals in any particular way. I wouldn't feel uncomfortable for long if they got turned over. My discomfort would be a passing feeling, brief and yet genuine.
Half-belief
Some researchers use the term half-belief to describe what's happening when people engage in superstitious behaviors they don't actually believe in. A 2023 study conducted by an Israeli team showed that superstitious behaviors and beliefs are somewhat independent of each other. Some people cross their fingers because they truly believe it brings them luck. Some people cross their fingers even though they know that it won't change anything. In between those extremes are folks who cross their fingers but more or less doubt the usefulness of it. You might say these half-believers are doing it "just it case." When you choose seats on a flight, you may avoid the 13th row while saying to yourself: I know this is superstitious nonsense, but why take a chance?
The concept of half-belief offers a more nuanced way of understanding superstitious behavior and beliefs, but it's more descriptive than explanatory. In other words, it doesn't tell us why some people are superstitious. Why treat a stuffed animal kindly if you know it's just an object?
Core knowledge confusions
This brings me to the work of University of Helsinki emeritus Marjaana Lindeman, whom Stuart Vyse described in an email to me as "arguably the foremost modern researcher on superstition and paranormal belief." Lindeman is indeed widely cited, and it's her theory that was put to the test in the meta-analysis I'll be discussing.
Lindeman's theory can be called a "dual-process theory". This is a social science term for any theory which assumes that people have two ways of thinking: A quick, intuitive way, and a slower, more rational way. For instance, intuitive thinking might lead you to immediately like or dislike a person you've just met, whereas a more reflective process might result in a more complex impression.
Children are mostly intuitive thinkers. As they grow up, rationality increases, but rational thinking never completely replaces the intuitive variety. For instance, an adult who's in a hurry, or stressed out, or distracted, might make a quick, intuitive decision about what to eat for lunch. In calmer circumstances, their decision might be slower and more "rational".
Lindeman assumes, based on strong evidence, that even though everyone engages in both intuitive and rational thought, some people think more intuitively than others. These are the superstitious folks. They're also the folks who exhibit paranormal beliefs, belief in the supernatural, and magical thinking. In Lindeman's theory, these kinds of "irrational" thinking are all, in essence, the same. In other words, believing in ghosts or ESP is fundamentally the same kind of thing as believing in lucky charms or the magical power of numbers.
Lindeman argues that superstitions arise from "core knowledge confusions" that take place when we're thinking intuitively. These CKCs illustrate what's known as category errors.
Category errors are not just mistakes; they're logically insensible. For instance, it's factually wrong to say that a grain of sand weighs 10,000 pounds, but it's not entirely senseless, because each grain of sand does weigh something. One could imagine a planet where the gravitational field is strong enough that a grain of sand would be quite heavy. However, if you say that an idea weighs 10,000 pounds, the statement is both wrong and senseless. It's a "category error" because mental phenomena such as ideas don't have weight, texture, color, or any of the other properties we find among physical things. (You can speak of having "heavy" thoughts, but that's just a metaphor.)
Consider triskaidekaphobia, or fear of the number 13. If you feel uncomfortable about staying on the 13th floor of a hotel, it's because you think something bad may happen later on. Perhaps you have something specific in mind (e.g., falling and breaking your ankle), perhaps you don't. Either way, you're assuming that a number can influence the physical world. Because numbers are abstract concepts, it's a category error to believe they can have a physical influence (e.g., by causing you to trip and fall). For this reason, numerology can be viewed as a form of superstition.
Likewise, the belief that stuffed animals and other inanimate objects have feelings would be a category error, because only living things feel. (This is one of the reasons people dismiss the possibility that AI bots have consciousness. Because AI is not a living thing, treating it as thoughtful and possessing self-awareness feels to some like a category error.) So, Lindeman might say that the way I arrange my daughter's stuffed animals reflects a category error that can be traced back to intuitive rather than rational thinking. That makes me feel less embarrassed (sort of).
A new meta-analysis
Two weeks ago, the journal Personality and Individual Differences published a meta-analysis conducted by Albina Gallyamova, Elizaveta Komyaginskaya, and Dmitry Grigoryev at HSE University in Moscow. The purpose of the meta-analysis was to review studies on the relationship between core knowledge confusions and paranormal beliefs. In essence, Gallyamova and colleagues were putting Dr. Lindeman's theory to the test.
(Recall that Dr. Lindeman treats superstitions, paranormal beliefs, supernatural beliefs, and magical thinking as fundamentally the same. Many of the studies Gallyamova and colleagues reviewed focused on one or more of these kinds of beliefs, or failed to make a clear distinction. To keep things simple, I will continue to use the term "superstition", with the understanding that something more inclusive is intended.)
Gallyamova and colleagues reviewed 22 studies published between 2010 and 2024 which reported data on a total of 16,129 participants. In these studies, core knowledge confusions were measured by asking people whether certain statements are best understood literally or metaphorically. For instance, consider the statements "rocks live a long time" and "a house knows its history". A person who indicates that these statements are best understood literally would be showing signs of core knowledge confusion, since inanimate objects, by definition, aren't alive and can't know things.
Gallyamova and colleagues showed that although the results were quite heterogeneous from study to study, core knowledge confusion was consistently associated with superstitious beliefs. The greater the confusion, the more superstitious people turned out to be.
Dr. Dmitry Grigoryev, senior author on the meta-analysis, told me via email that he attributes this finding to an "evolutionary tendency" to think intuitively. Thinking quickly and unreflectively gives rise to superstitions and other kinds of unfounded beliefs, in his view. As with other evolutionarily-driven tendencies, such as aggressiveness, this form of thinking can be adaptive, and some people exhibit more of it than others.
The average effect size in the meta-analysis was a correlation of .40, which is high enough to be meaningful, but still leaves much variance unaccounted for. In other words, core knowledge confusions contribute to superstition, but there are many other influences as well. What else is going on?
I reached out to Marjaana Lindeman for her impressions. It was gracious of Dr. Lindeman to reply, given that she retired four years ago and, as she put it, had to "do some archeological excavations in my mind" to answer my questions. (She was speaking metaphorically, of course.)
Dr. Lindeman noted that one wouldn't expect any one study to account for all sorts of irrational beliefs. As she put it,
"there are a huge, if not an infinite number of paranormal / magical / superstitious / supernatural beliefs, and they vary in different times (trendiness) and cultures. All potential beliefs can therefore never be captured in scientific studies."
She also commented on the difficulty of measuring core knowledge confusions, because "they are assumed to be implicit and unconscious". For instance, until I started this newsletter, I never noticed that I was keeping my daughter's stuffed animals upright, or that I was therefore unconsciously giving some credence to the idea that inanimate object have feelings. Meanwhile, asking people how they interpret metaphors gets tricky when you start to ponder what they're actually thinking. Rocks do live a long time, literally speaking, if one takes "live" to mean "exist in current form".
Finally, Dr. Lindeman stressed that some people may "just adopt the opinions of their environment uncritically, without any special “cognitive determinants”. Which is to say that superstitions may be learned beliefs that we don't reflect on. This would explain why behaviors like knocking on wood have endured for centuries.
What I find especially persuasive about Dr. Lindeman's theory is that it makes room for many of the other contributors to superstition that scholars have discussed. For instance, she told me
"Fear and lack of control can easily promote intuitive processes, and thereby also core knowledge confusions, at least in those individuals who have these confusions. However, I would not regard fear or lack of control as causes or determinants of superstitions as such, because they can arouse a variety of outcomes, even opposites of superstitions, for example the search for scientific knowledge. Irrationality, in turn, may take several forms and I think that ontological confusions are one manifestation of irrationality."
In other words, fear and a need for control sometimes, but not always, spur those intuitive processes wherein CKCs are born. Same goes for irrationality. Fear and the rest are simply triggers for a quick and less-than-optimally rational way of thinking that gives rise to superstitions.
The less important question
At the outset I noted that Rafael Nadal claims that he's not superstitious, while Naomi Osaka claims that she is. Are they right?
The answer would depend on a much deeper conversation with them.
Stuart Vyse and others argue that professional athletes tend to be superstitious because they're engaged in high-stakes, low-certainty activities. Even the best of the best often lose, miss shots, strike out, etc., and so they have good reason to fear failure and to seek more control over their performance. But the comments Nadal and Osaka have made to reporters illustrate that it may not be easy to interpret their particular superstition-like behavior.
If an athlete believes that their rituals affect their luck, or otherwise magically influence how they perform, we might call them superstitious. As Lindeman would put it, it's a category error to assume that stepping on a painted line in between points would undermine one's game a few moments later.
If the athlete believes that their rituals merely help them concentrate, or keep them calm, we might say they're not superstitious. Still, we couldn't be sure. How does the ritual have those effects? Does the athlete recognize the ritual as a meaningless act that provides comfort and focus simply because it's familiar, or do they think it radiates magical, calming energy? If they think it's magical, are they fully superstitious, or is this a sort of half-belief?
In short, I don't think we can say for sure that Nadal isn't superstitious, or that Osaka is. We couldn't know unless we talked with them at length.
This illustrates a limitation on how well statistics can help reveal the mind's inner workings. We can demonstrate correlations between core knowledge confusions and superstitious beliefs. We can identify other variables that affect the strength of those correlations. But deciding whether or not any particular person is superstitious, and understanding what gives rise to their particular beliefs, requires a great deal of knowledge about that person's thinking. Quantitative research struggles to achieve this level of granularity.
The more important question
We'll be fine not knowing what motivates the rituals of great tennis players. It's more important to figure out how to dislodge the most harmful superstitions, whether it's the belief that particular ethnic groups bring bad luck, the expectation that faith healers cure COVID-19, or one particular gambling addict's confidence that their "system" will break the bank at Caesars Palace next weekend.
In her message to me, Dr. Lindeman didn't sound very optimistic about changing a superstitious belief that's already established:
"I'm pretty skeptical that an outsider could change an adult's beliefs. Attempts to change another person's views are rarely successful. The change must start either from the person, i.e. from the fact that s/he starts doubting the foundations of own beliefs for one reason or another. For these reasons, I think the best way is to teach children critical thinking (reading, listening) at school from the first grade."
Dr. Grigoryev, senior author on the meta-analysis, also stressed the role of education, while expressing a bit more optimism about what educators call direct instruction:
"I really believe education plays a crucial role. Specifically, addressing ontological confusions by more explicitly defining ontological categories and their attributes could prove effective. Incorporating regular exercises that enhance accurate attribute assignment within educational curricula may help reduce these errors in further reasoning."
I agree with both scholars regarding education. Incorporating more critical thinking into the K-16 curriculum would be helpful, if teachers can find time for it, and there are lots of opportunities to uproot category errors in classroom settings.
As for changing other peoples' minds, I agree with Dr. Lindeman that this a difficult task.
However, shifting people from intuitive to rational thinking is possible, to an extent, and there's a ton of research on countering disinformation, some of which I've discussed in prior newsletters. Dr. Grigoryev is one of many scholars who explore commonalities between superstitious thinking and phenomena such as conspiracy theorizing. Identifying common contributors to unfounded beliefs, as Lindeman and Grigoryev have done, at least simplifies the problem that needs to be addressed.
One of the silver linings of the disinformation age is our increasing concern about all sorts of unfounded beliefs. Although fear can make us more superstitious, it can also motivate more careful, critical thinking as well. May the Force (of reason) be with us!
Thanks for reading!
Dr Springer, have you ever looked at the prevalence of real world adverse events and the Covid vaccine? I believe there is not an issue here but I have not seen a review of data similar to what you do. Thanks