History & Culture
From the Archives:
The Lunacy of It All: Lunar Phases and Human Behavior
In the 1980s, two experts re-examined studies regarding whether there is a relationship between aberrant human behavior and phases of the Moon.
By James Rotton and Ivan W. Kelly
NPS Homestead
Editor's note: This article is a reprint, and originally appeared in Mercury's second decade.
"Dear Ann Landers," the letter began, "I have been dating a lovely woman for a year ... But one thing worries me. When the Moon is full she becomes hyperactive, vile-tempered — in short, a little wacky. I've heard that the position and shape of the Moon have a very real influence on the behavior of certain people. Is it true or false?"
"It's true," Ann Landers told her readers, on August 27, 1984: "The word 'lunatic' is derived from the Latin 'luna,' which means Moon. Some people get loonier than others when the Moon is full."
Ann Landers is not alone in believing that a full Moon brings out the worst in people. In a study conducted at Florida International University,1 we found that nearly one-half of 157 students indicated that they believed that "some people behave strangely when the Moon is full." Likewise, Gordon Russell and Manjula Dua reported2 that about 44 percent of students at a Canadian university believed that a full Moon affects behavior. Similar levels of belief have been recorded among students in Singapore by Laura Otis and E. C. Y. Kuo.3
In our research we have found that students who score high on our measure of lunar beliefs also endorse items indicative of belief in paranormal phenomena, such as telepathy, precognition, and witchcraft. Men and women receive almost identical scores on our scale, and we have yet to find a personality variable that predicts belief in lunar effects. But one of our students, Pat Elortequi, recently found that police officers were more likely to subscribe to lunar beliefs than either professionals in a mental health setting or a haphazard sample of pedestrians (the proverbial "man and woman on the street").
Blame It on the Moon
At one time or another, the Moon has been worshipped as a goddess, feared as a devil, welcomed as a blessing, loved as a mother, and used as an excuse for human foibles. In more recent times, investigators have attempted to link phases of the Moon to everything from arson and homicide to several types of abnormal behavior ("moon madness"), including paranoia, epilepsy, suicide, sleepwalking, and, of course, lycanthropy ("the werewolf syndrome").
There are several reasons why one might think that the Moon would have an effect on human behavior. The one that has received the greatest amount of attention is Lieber's "biological tide theory." In The Lunar Effect (a book published in 1978 by Doubleday), Miami psychiatrist Arnold L. Lieber likened the Moon's pull on the human body to ocean tide. At first glance, Lieber's analogy appears reasonable. After all, the Moon is responsible for tides in the ocean, and the human body is 80 percent water. Maybe there are some people whose "tidal balance" is upset by a full Moon.
Unfortunately for Lieber's theory, gravitational effects depend as much upon the mass of the object being "pulled" (one's body) as the object doing the pulling (the Moon). If you look out on the Atlantic Ocean from the shore in South Florida, you cannot help but be impressed by the Moon's effects on the tides; however, if you drive a few miles north and gaze out on Lake Okeechobee, the second largest body of water in the United States, you have to look hard to discern a ripple on its surface. Not trusting our less than systematic impressions, we asked Roger Culver, an astronomer at Colorado State, for a more precise estimate of the Moon's pull on the human body. He passed along the results shown in Figure 1, which indicate that the Moon's tidal tug on a newborn baby is 12 million times less than the pull of its mother. Obviously, those who worry about the Moon's gravitational pull should be more concerned about the pull of nearby buildings. Even more devastating to the "biological tide" hypothesis is that there can be no significant tide unless the tide-producing force differs significantly from one part of the person to another — which can only happen if the person is as big as a planet!4
The Full Moon's glow lights a dark night.
[Photo by Aron Visuals on Unsplash]
Moonshine
Although moonlight is nothing more than the Sun's reflection, some people have suggested that there might be enough of it to upset our circadian (or daily) rhythms. However, as astronomers Roger Culver and Phillip lanna noted in their book, The Gemini Syndrome (Prometheus Books, 1984), the Moon's light is minuscule when compared with the Sun.
Beliefs about the Moon's evil influence arose long before the widespread use of electricity. As David Campbell and John Beets observed in an early review of the literature5 in this area, it makes some sense to assume that criminals and unsavory characters might have followed the lead of farmers who took advantage of moonlit nights to plow their fields. However, it makes just as much sense to conclude that criminals prefer to take advantage of the dark provided by a new Moon. Since electric lighting is more prevalent in urban than rural communities, it occurred to us that this matter could be settled by comparing studies done in different locales, which included a few on people's behavior in less developed nations. As we report in a more recent review of the literature6, none of our comparisons attained statistical significance. One is no more likely to meet a "lunatic" on moonlit than any other nights.
Other writers have advanced more esoteric theories to account for the Moon's effect upon behavior. Writing in Environment and Behavior (vol. 14, p. 408), for example, Salvatore Garzino attributed lunar effects to positive ions:
"During the full-moon phase, positive ions come down to earth in great abundance. But positively charged ions are now suspected by some scientists to create depression and irritability by increasing levels of serotonin in the nervous system. Serotonin is a mood-modifying chemical, a "downer" (italics in original.)"
It is true that there are more positive ions in the air when the Moon is full, but the effects of positive ions are a "sometimes thing." They show up only when individuals are bombarded by thousands of positive ions in a controlled laboratory environment. The crucial word here is "controlled," because solar radiation, air conditioners, and air pollution also generate ions. Commenting upon Garzino's article in the same issue of Environment and Behavior (p. 418), David Campbell pointed out that one is much more likely to feel the effects of positive ions in an air-conditioned building.
There is also some reason to believe that serotonin is responsible for some (but certainly not all) of the effects attributed to positive and negative ions.
There are circadian and even seasonal variations in levels of serotonin, melatonin, and several other hormones in our bodies. Of the thousands of studies done on circadian rhythms, however, not one has reported anything resembling a 14-day or lunar (29.54-day) cycle in hormone production.
The Moon's surface holds a complex record of past impacts and collisions in the inner solar system.
[Photo by ThisIsEngineering at Pexels]
Moon Over Miami
Theories about mechanisms linking the Moon to behavior are interesting, but they beg the more fundamental question: is there really a relationship to be explained? Is there any evidence that a full Moon brings out the worst in people? To answer this question, we began by doing a computer search of the literature. We were able to locate 41 empirical reports that dealt with phases of the Moon and abnormal or criminal behavior. We also asked authors for copies of their data. One of the things that surprised us was the number of errors we found in often cited reports, including Lieber's now infamous study of homicides in Miami and Dade County, Florida.
In a study done with Carolyn Sherin, which was published in the American Journal of Psychiatry (vol. 129, p. 69), Lieber claimed that murders were more likely to be committed when the Moon was full than at other times of the lunar (29.54-day) month. This claim is based upon three statistical tests of significance, which are reported in their article. However, after we obtained a copy of their data analyses, we found that they had actually performed 48 nonindependent tests of significance. If their tests had been independent, 2.4 (i.e., .05 x 48) of them would have attained statistical significance by chance alone. What Lieber and Sherin did is akin to telling somebody that five out of five coins came up heads, which is a rare event, without mentioning the fact that this occurred after several minutes of coin tossing.
Taking a closer look at the data in Lieber and Sherin's study, we found that their statistical tests were far from independent. For example, in one set of analyses, Lieber and Sherin not only tested for excess numbers of homicides during the 24-hour interval around full moons, they did so for excesses during 48- and 72-hour intervals. To make matters worse, they also attempted to assess delayed or "lagged" relationships by expanding their intervals to include 24-, 48-, and 72-hour periods after full- and new-moons. In other words, their conclusions were based upon overlapping tests of significance. This is what gamblers do when they try to "cover all bets.'' Playing roulette, for example, one can record a large number of "wins" by betting on "red" as well as "odds" and a fair share of favorite numbers (e.g., 2, 7, 11, 23, and 35).
There are more appropriate statistical procedures for comparing events during different phases of the Moon.7 Applying them, we found that homicides were no more likely to occur during one phase of the Moon than another. It might be noted, in passing, that Lieber based his "biological tide theory" upon homicide rates in Cleveland, Ohio, as well as Miami, Florida. Lieber and Sherin's data covered a period between 1958 and 1970. Nick Sanduleak, an astronomer at Case Western Reserve University, has recently published a follow-up study on Cleveland homicides.8 The data in this follow-up study spanned the ten-year period between 1971 and 1981. Sanduleak's results are aptly described by the title of his article: ''The Moon is acquitted of murder in Cleveland."
Looking at data from other studies, we found that apparent relationships between phases of the Moon and behavior were hopelessly confounded with weekend and seasonal trends. For example, Donald Templer and colleagues at the California School of Professional Psychology, in Fresno, claimed that they had uncovered a reliable correlation between phases of the Moon and traffic accidents in three states. This claim appeared in volume 55 of Perceptual and Motor Skills (p. 280). However, in volume 57 of the same journal (p. 856), they reported that the correlation disappeared when they took weekends, holidays, and seasonal differences into account.
While humans perceive the Moon's visible changes during its 29.5-day cycle, scientific evidence does not support there is any relationship between the Moon and human behavior.
[Photo by Sotiris Savvides on Unsplash]
Meta-Analysis: Recycling Lunar Cycles
We will concede that a few investigators have uncovered what appear to be "statistically significant" relationships between phases of the Moon and abnormal behavior. But all statistical significance means is that a result is unlikely to occur by chance alone. For example, you'd get what appears to be a statistically significant result if ''heads'' came up five times in a row when you tossed a coin. The probability of getting five heads in a row is 1/2 raised to the fifth power or .03125. That is, we would expect such an outcome about three percent of the time. That is unlikely enough to justify publication in many journals. But what would you do if somebody else got five tails in a row? This was the problem facing us. In our review of lunar lunacy research5, we located three reports whose authors could show that more "lunacy" (i.e., criminal and deviant activities) occurred when the Moon was full. But these were matched by four reports of significant deficits, which meant that fewer abnormal and criminal behaviors occurred when the Moon was full than when it was not. Still other investigators recorded excesses when the Moon was new or in its first and third quarters.
To deal with this problem, we resorted to meta-analysis, which is a statistical procedure for combining results from different studies.9 In our meta-analysis, we found no evidence for commonly held beliefs about the effects of a full Moon. About 25.7 percent of all unusual episodes occurred when the Moon is in its full phase, but the Moon is full 25 percent of the time. A difference of less than one percent could easily occur by chance alone. Further, even if one is reluctant to accept chance as an explanation, it should be emphasized that our estimate is based on results obtained in studies whose authors did not include controls for meteorological conditions, weekend differences in human activities, and seasonal trends in behavior.
Of course, it could be argued that a full Moon affects some people more than others. To check out that possibility, we looked at differences in psychiatric admissions, disturbances in mental hospitals, calls to crisis intervention centers, suicides, criminal offenses, including homicides, and several other types of behavior. In no case did we uncover a reliable difference. We also looked at sex (no difference); the Earth's distance from the Moon in terms of the latter's apogee-perigee cycle (no relationship); differences between urban and rural communities, where people would be more likely to notice the Moon (nothing); and even the distance between each study's site and the equator (nada).
In sum, scientific evidence does not support commonly held beliefs about the Moon and human behavior. We cannot, of course, prove that a full Moon doesn't affect behavior — any more than we can prove that werewolves, unicorns, and other interesting creatures don't exist. Fortunately, we don't have to do so. The burden of proof lies with those who claim that people's behavior is affected by the Moon. They will have to obtain a great deal more — and better — evidence before we will feel obliged to regard the Moon as anything more than a handy symbol in monster movies. ✰
(Originally published in the May/June 1986 issue)
JAMES ROTTON was a member of the Psychology Department of Florida International University. IVAN W. KELLY is a member of the Department of Educational Psychology at the University of Saskatchewan.
Mercury is distributed by the Astronomical Society of the Pacific, a nonprofit organization dedicated to astronomy literacy, education, and outreach.
Mercury is an advertisement-free publication. If you are interested in supporting Mercury, please email us.
