This View of Life Anything and everything from an evolutionary perspective.
FIND tvol:
God Is Watching You: How the Fear of God Makes Us Human
Culture Religion
Image credit: Oxford University Press
Dominic Johnson
Dominic Johnson
is Alistair Buchan Professor of International Relations at the University of Oxford

In Dominic Johnson’s new book, ‘God is Watching You: How the Fear of God Makes Us Human‘, he presents a “new theory of the origins and evolution of not only religion, but also human cooperation and society.  He also explores how fear of supernatural punishment exists within and outside of religious contexts, and uses an interdisciplinary approach that draws on new research from anthropology, evolutionary biology, experimental psychology, and neuroscience” (source).  Below is an excerpt from ‘God is Watching You’. 

In the court of fourth-century B.C. King Dionysius II of Syracuse, Sicily, there was a man called Damocles who was in awe of his king. The king lived in great luxury, reigned with power and authority, and was surrounded by great men. Damocles counted the king a lucky man. Intrigued, Dionysius suggested they switch places so that Damocles might see what being king was really like. Once seated in the throne, Damocles surveyed his position of greatness and was distraught to find a heavy sword hanging directly above him, dangling by a single horse’s hair. The sword threatened to drop and kill Damocles at any moment. Horrified, he pleaded to switch back again. He no longer wanted to be so lucky as the king. The Sword of Damocles, as Dionysius’s lesson told, was the ever-present threat of lethal danger that a king must endure when all eyes are upon him. No matter what his luxuries and power, or indeed precisely because of them, he feared every moment for his life.

We do not need to be powerful kings to live under the burden of scrutiny and threat. Supernatural punishment is like the Sword of Damocles. If there is a God, then all of us are under watchful surveillance and the possibility of supernatural punishment. We go through life with freedom and tempting opportunities, but if ever we make a misstep, there is a sword waiting to strike us down. In fact the situation is worse than for Dionysius, because there are two swords raised above us. One is the relatively blunt instrument of real-world consequences—the retaliation or injury to our reputation we may receive from our fellow man. This sword is dangerous, and damaging to Darwinian fitness, but it is erratic and not usually lethal. The other is the weighty and sharp instrument of divine punishment, in this life or the afterlife, that we may receive from the gods. This sword is truly terrifying, with a blade that is infallible and can deal a fate even worse than death. For a believer, life is doubly dangerous, and the sharp, heavy sword of God is the one that elicits the greatest fear and is the greatest deterrent. But this is exactly where natural selection may have extracted real-world benefits from supernatural beliefs. The blunt sword of our fellow man is real, and the sharp sword of God may not be, yet a fear of the sharp sword can save us from the blows of the real one. A belief in the wrath of God steers us away from risking the wrath of man.

Sign up for our newsletters

I wish to receive updates from:

Because egoistic behavior is deeply ingrained and hard to control, we may need encouragement to stay on the rails. Rational restraint is often too weak, too slow, or simply unavailable. This is where God can help us, whether he is real or not. People who fear supernatural punishment—those living with a supernatural Sword of Damocles above their heads—are likely to be more cautious in their behavior, and as a result may incur lower real-world costs.

While writing this book I asked a colleague trained in theology and religious studies whether there was a standard work in the field on the role of supernatural reward and punishment that I might have missed. He paused and thought for a minute, then said, “no I don’t think there is anything like that. It sounds more like economics.” This was a telling observation because it is precisely the economics of costs and benefits that underlies an evolutionary perspective on religion. What evolutionary biology does is scrutinize an organism’s traits from the perspective of the cold economic accounting machine of natural selection, and explores how investments in these traits contribute to dividends or deficits in reproductive success. Costly investments that fail to bring returns cannot survive in a competitive marketplace. As Richard Dawkins notes, evolution quashes any tiny extravagance because “natural selection abhors waste.” It is economic gains and losses, therefore, that fundamentally underlie the evolution of our physiology, brains, and behavior. If a trait doesn’t pay, it usually falls by the evolutionary wayside. Indeed, paleontologist Geerat Vermeij called his treatise on the 3.5 billion year history of the evolution of life on Earth “Nature: An Economic History.” The simplicity of natural selection is what unnerves many people about it. But that is also its power. It offers a fresh insight on all human affairs because it forces us to think hard about how any given behavior would have had positive or negative consequences for fitness in our evolutionary past. Without thinking that through, we cannot develop a clear picture of whether something we observe today is an adaptive trait favored and shaped by evolution, or a mere byproduct of other traits or a recent innovation of culture. We all know that supernatural reward and punishment plays some role in the everyday lives of billions of believers around the world today. The bigger question is what effects these beliefs may have had in our evolutionary history. Obtaining and interpreting data on our earliest ancestors, and on contemporary hunter-gatherer societies as analogues of what life was like for our Pleistocene ancestors, is not easy. Nevertheless, I suggest that a belief in supernatural punishment helped, not hurt, the economic bottom line of Darwinian fitness, and in so doing pushed human beings along a path to lower costs of selfishness for themselves and greater benefits of mutual cooperation with each other.


One might still wonder why a supernatural belief would really be needed to deter selfish behavior. In their review of the role of supernatural punishment in the evolution of religion, biologist Jeff Schloss and philosopher Michael Murray end by asking why, if selfish behavior was so costly, evolution did not favor a simpler solution for suppressing selfishness than the “excessive deterrence of belief in all-knowing, all-powerful agents.” I suggest that there are several good reasons why God—or other supernatural agents—actually represent the best possible deterrent against selfishness.

God may work precisely because he is excessive. A belief in supernatural punishment may be a good method of deterring people from the real-world costs of selfish actions because of an asymmetry in the possible errors one may make. Beliefs can be wrong in two ways: believing X is true when it is false (a false positive error), and believing X is not true when it is (a false negative error). “Error management theory” (EMT) suggests that wherever the costs of false positive and false negative errors are asymmetric, we should expect natural selection to favor some method of avoiding the more costly error. For example, we often think a stick is a snake (which is harmless), but we do not tend to think snakes are sticks (which is potentially deadly). This is obviously a highly adaptive bias. Similarly, when we set the sensitivity of a smoke alarm, it is better to err on the side of caution because the costs of a false alarm are negligible (however annoying it may be at the time), whereas the costs of being burned to death in a fire are great indeed. One should thus expect the smoke alarm to go off a bit too often. Whenever the true probability of some event—such as the detection of snakes, fires, or selfish behavior—is uncertain, a biased decision rule can be better than an unbiased one if we want to avoid the more dangerous error. To be clear, the reason why a bias is best is because an unbiased decision rule centered on the true probability of the event will make both false positive and false negative errors—it will generate some false alarms, and it will fail to go off in some real fires. By contrast, a biased decision rule will err on the side of caution—it will have a lot of false alarms, but more importantly it will rarely fail to go off in a real fire. This logic of managing errors is thought to underlie a range of human cognitive biases, including overconfidence, assuming the worst in others, and differing perceptions of sexual interest among men and women.

Plugging this logic back into the context of religion, if we aim to assess the true probability of detection for cheating, then half the time we will overestimate the probability and get away with it, and half the time we will underestimate it and get caught. That might seem a reasonable, fifty-fifty strategy. But the kicker is if these two mistakes entail different costs. If false negative errors (assuming stealth and getting caught) are more costly than false positive errors (assuming detection and missing a reward), then only exaggerated estimates of the probability of detection—such as a belief that supernatural agents are observing your behavior all the time—will help you to avoid the worst of the two errors. As long as there is some uncertainty about the true probability of detection, simply improving accuracy will not help. The best solution to avoiding detection, therefore, is a mechanism that overestimates the true probability that detection will occur—exaggerated estimates outperform accurate estimates, because the latter will engender more mistakes.

This means that it is not just an asymmetry of costs and benefits that may be decisive (God-fearers doing better than Machiavellians), but a bias to overestimate what those costs and benefits are (God-fearing itself is favored by natural selection). Given the dangers of our social minefield, an exaggerated belief that one is constantly being watched and judged by supernatural agents—the fabled hyperactive agency detector device (HADD)—may be an especially effective guard against careless selfishness. Schloss and Murray’s concern that God is a “seemingly excessive” deterrent is prescient—the threat of punishment may need to be excessive, such as a belief in an omniscient and omnipotent God, because this is the most effective way to avoid dangerous mistakes.

God is also a good deterrent because supernatural agent concepts are easily accessible and fast. The cognitive science of religion literature suggests that beliefs in supernatural agency and supernatural consequences are the cognitive default, and are more easily accessed than many other concepts. In particular, it appears that negative, incriminating information is particularly salient and rapidly available to people in tests that are designed to measure people’s subconscious, implicit associations. Other forms of deterrent, if they are (or once were) rivals, might therefore have been trumped simply by the ease and speed of supernatural cognition.

The idea that a cognitive bias or cognitive illusion may be a good way to achieve adaptive behavior is not so bizarre as it may sound. In fact it is turning out to be a commonly recognized phenomenon in psychology. There is a huge list of human cognitive and motivational biases, and many of them are thought not to be malfunctions or constraints of the human brain, but rather adaptive heuristics that help us solve important problems. In a complex world with a mesmerizing barrage of information and the need to navigate through it efficiently, the brain has developed numerous shortcuts that serve to quickly and effectively direct our behavior. In the previous chapter we saw several examples of apparently irrational biases that underlie superstitious behavior in all of us. But even superstitious behavior can have advantages if it somehow reduces anxiety, helps us attempt otherwise daunting tasks, or improves performance. As has been demonstrated in laboratory experiments, superstitious beliefs can actually increase success at solving problems. In the real world, too, there is evidence that superstitious beliefs can help us out. One study, for example, found that higher performing basketball teams, and higher performing players within them, showed more superstitious behaviors. We also saw that evolutionary biologists Foster and Kokko suggest superstitious behaviors are common among animals, and this is no accident: they were favored by natural selection because they increase fitness. As David Sloan Wilson has argued, “the unpredictability and unknown nature of our environment may mean that factual knowledge isn’t as useful as the behaviors we have evolved to deal with this world.” For this reason, to understand adaptive behavior we should not look for accuracy or rationality, but rather for what works. “Adaptation is the gold standard against which reality must be judged,” Wilson notes. By this standard, God is a great solution.

Published by Oxford University Press. Copyright 2015 by Oxford University Press. Reprinted with permission of the publisher. All rights reserved.

Join over 3.000 visitors who are receiving our newsletter and learn how to optimize your blog for search engines, find free traffic, and monetize your website.
We hate spam. Your email address will not be sold or shared with anyone else.
1 Comment

Join the discussion

One Comment

  1. Bryan Atkins says:

    Hear thee regarding the adaptive efficacy of the god app; and hear Dr. Wilson re adaptive behaviors generated by evolution over time may be more adaptive than factual knowledge.
    But think exponentially accelerating complexity, which includes exponentially accruing knowledge, blows up both arguments, at least, should these complexity gains continue. If we suffer a collapse of civilization, seems likely that both god and archaic, non-factual adaptive behaviors will retain their survival merit.

    Think survival is largely a function of processing complex network relationship information with sufficient speed, accuracy and power. Per our knowledge explosion, our reach in and across geo, eco, bio, cultural & tech networks, and across time is, as you know, unprecedented.

    Because our species power / reach has increased exponentially, think the margins for the non-factual have been greatly reduced. If you’re a primitive hunter-gatherer from 15K or even 1,000 years ago, you can roll with myths. Get your myth wrong, not a lot of significant consequences, e.g., you can’t pound the world with nuclear weapons because your god/religious story told you its judgment day. But if large groups of people think praying is going to reverse climate change, or non-factual behaviors are going aid us, long term, in navigating / processing these levels of complexity, think that’s fallacious, or Dead Wrong. Both those factors impede the accuracy, speed, and power of information processing, i.e., survival.

    “Knowledge is a constructor.” Physicist David Deutsch – Constructor Theory of Information. To handle / process / navigate these levels of complexity and our new reach, think we absolutely need the constructor power of knowledge, just to survive.

    In addition, think tech, the internet, ubiquitous cameras, etc. are moving us along the eusocial spectrum, to more integration, with the potential to become a world culture. At one point we had roughly 200,000 political entities, tribes, chiefdoms, etc. Now we’re somewhere around 200 nation states. Instead of reverting to the archaic omniscient, omnipresent, omnipotent culture app: god, think we need to create a technological genome, a neocortex of sorts for culture – a genome better matches the incredible reach of culture and technology. We do this: Because code is information-processing tech / relationship infrastructure in bio, cultural & tech networks: genetic, epigenetic; religious, math, language, legal, monetary, etiquette, moral; software . . . and because increasing amounts of our bio and cultural genomes are being overrun by complexity, again, reducing their efficacy for reality interface. We’re converting the sky into a lethal gas chamber (largely a gas chamber of commerce), exponential extinction rates, etc. We’re bleeding out, at an accelerating rate.
    The tech genome will be a god of sorts, as it already is in a limited, still primitive manner: The internet destroyed the dentist who shot the lion; cameras catch cops shooting unarmed people in the back as they’re running away, people speeding, etc. The NSA has already done a lot of the tech genome work with their increasingly ubiquitous monitoring.