‘Life is mostly good, despite the greater power of bad things’

You use the phrase 'Bad is stronger than good' to summarise the claim that there is 'a universal tendency for negative events and emotions to affect us more strongly than positive ones'. Echoing Jennifer Corns' 2018 article 'Rethinking the negativity bias', what makes an event or emotion good (positive) or bad (negative)? For example, you note that receiving criticism can be painful but that it need not be and even when it is, it can lead to all sorts of positive consequences. How do we know when something 'is' bad, good, neutral, or multifaceted? 

This is a profound question, and a surprisingly difficult one. As one of my professors (Hans-Georg Gadamer, the great German philosopher) pointed out, it is impossible to speak only with terms that have been defined. Good and bad are some of the most fundamental concepts in language, and so it is hard to define them in terms of yet more basic concepts. 'Good' and 'bad' are among the first concepts that children learn (even dogs learn them!). Moreover, brain researchers say that the brain classifies something as good or bad almost immediately after figuring out what it is, in less than half a second. 

To be sure, there are complicated phenomena that mix good and bad, and there may be some things that are completely neutral. For most things, however, everyone gets a clear emotional/affective signal about whether it is good or bad. Probably many things link back to survival and reproduction. Things that help sustain life in these ways are experienced as good, and things that are detrimental to survival and reproduction are bad. I have long suspected that the first 'conscious' experience in evolution was pain, because it signals the brain that the body is being damaged – and preventing such damage is useful for sustaining life. 

Again, I could attempt to give you a complicated philosophical answer, but it might be beside the point. In practice, people determine quite rapidly and automatically whether something is good or bad. 

But there are clearly degrees of both good and bad, so justifying or assessing a claim that 'Bad is stronger than good' seems to require being able to compare the effects of 'equivalent amounts' of bad and good. If it is true that couples need 'five fucks for every fight' to achieve relationship satisfaction (p.23), how do we know that this is evidence of a negativity bias rather than it being the case that 'fights' are much bigger bads than 'fucks' are goods within romantic relationships? Presumably, your Rule of Four ("It takes four good things to overcome one bad thing", p.27) will not usually hold if the good things are making your partner cups of tea and the bad thing is admitting that you're having an affair with their twin.

Yes, correct, the assumption that bad is stronger than good is based on the assumption that the objective magnitudes of the two can be equated. I think we do have some examples and caveats in the book to emphasise your point, that a huge negative cannot be offset by a few minor good things. Indeed, most people say that no quantity of good deeds will offset a murder. 

Fortunately, much of the laboratory research was able to find clever ways to equate the good and bad. In the loss aversion studies, for example, the same amount of money is gained or lost (and people react more strongly to the loss than the gain). A study on learning compared having children either get a marble for each correct answer or lose a marble for each wrong answer, so that the total number of problems and marbles was the same. (The study on teacher influences on children's learning made the same salary bonus contingent on the same level of performance improvement by the students, and the experiment simply varied whether the teacher received the bonus up front and would have to pay it back in case students fell short, vs. the teacher would only get the money after the test if the students met the criterion.) Impression formation researchers have likewise measured the evaluative power of each trait and so can come up with a strict mathematical formulation of how the different traits should combine to form an average impression – only to show that the most negative trait has a disproportionate effect on the final impression. 

In short, there is an abundance of evidence from studies that carefully and precisely equated the quantities of good and bad. They provide the most rigorous evidence for the greater power of bad. 

Throughout your book you seem to give examples of exceptions to or moderators of the negativity bias, e.g. depending on people's past experiences, self-esteem, insecurity, sexuality, attributional style, gender, age, and age x gender interactions. You also note that people typically exaggerate their virtues, overestimate their abilities and their power to control their destinies, and experience all sorts of other 'positive' biases. You note that the vast majority of people experiencing terror do not develop post-traumatic stress disorders and 'typically emerge more capable [and] stronger'. Finally, you note wide individual differences in the extent to which people suffer from negativity bias (exceptionally illustrated by your description of 'Fearless Felix' Baumgarten's exploits in Chapter 3). In what sense, then, is the negativity bias 'a universal tendency'? 

Fair enough. I think in psychology a 'general rule' assumes that all else is equal, and in that sense the negativity bias appears pretty universal. (In fact, when we reviewed the scientific literature, we searched hard for genuine exceptions, because they would have really helped to flesh out the theory, and probably would have made it more interesting/complicated. But we were unable to find any genuine exceptions, in which good is stronger than bad if all else is equal. Likewise, since that review was published in 2001, I have heard multiple researchers saying they wanted to find an exception, but somehow these never seem to materialise.) 

Some of the counterexamples you mention simply show that some people are more sensitive to negativity than others. This is to be expected – people differ, after all – and it does not really constitute an exception. Your other points indicate that the negativity bias can be overcome. So it still exists, but we can deal with it. Of course we cope with bad things by trying to minimise or forget them, while we feel no such urge to downplay good things. That is a separate fact from the power of bad. 

Indeed, one of the book's themes is that life is mostly good despite the greater power of bad things. The mind evolved to overreact to negative things, presumably for good reason – but despite that, many people have happy, successful lives. The power of bad is a tool and/or a challenge, not a sentence of doom.

You cite a huge number of diverse empirical studies to support and illustrate your claims. I am not alone in worrying about the trustworthiness of large parts of the existing empirical literature. How did you avoid this particular negativity bias? Do you feel confident in the validity of the empirical results that you refer to for support – especially as so much of it seems to be drawn from the areas of psychology that are perhaps most persistently accused of being questionably created and particularly low in replicability? 

Hmm, much could be said. Let me start by saying that I for one think the replication brouhaha is widely overstated (like almost everything else in psychology). 

If one embraces the dismal outlook suggested in your comment, then one must discard almost all the work done by researchers over the decades, so we know almost nothing about psychology. There is then not much to talk about.

My own view is much more positive. Yes, there are some problems, and some published findings are undoubtedly wrong. But I think most social scientists are sincere, honest, hard-working people whose work does contain genuine truth. I think the strongest distorting factor that leads away from the truth is political bias, which I think is a much bigger problem for the field and the scientific literature than the replication issue. But political bias is not at work in the negativity bias, for the most part. The liberal-progressive political outlook that predominates among social scientists does not get any benefit from hyping the greater power of bad than good. Therefore, this literature is not contaminated by the biggest source of untruth.

As a literature reviewer, I rely on and trust the work of countless other scientists. I have, however, developed some methodological clues to know when to be extra suspicious and when to trust. When researchers have a motive to prefer one conclusion (e.g. with political bias), one is suspicious, but as I said, that does not apply. When they might be fabricating findings to fit preconceived notions, one again is suspicious – but that also does not apply to negativity. Many of the studies we reviewed were not even seeking to compare good and bad, but the findings emerged anyway, often as a problem for the researchers. (For example, in my work on social belongingness, we consistently found that rejection had stronger effects than social acceptance, as compared to the neutral control – we would have preferred that they be completely symmetrical.) 

Meanwhile, I have greatest trust in conclusions that are supported by converging evidence across multiple domains with different methods. And that precisely describes the negativity bias. The range of methods and areas in which we found bad to be stronger than good is overwhelming. 

Some critics of the field's body of knowledge say that many findings are the result of capitalising on chance. I think this argument is weak, but let's allow it to stand for a moment. It is true that one of twenty studies testing a false hypothesis will yield a significant finding, just by random chance. However, random chance works equally in both directions. There should therefore be just as many chance findings indicating that good is stronger than bad as the opposite, if that's what produced these effects. But there aren't. 

If you look back at our 2001 review article, you will see many different domains and phenomena, and over and over the greater power of bad was consistent. And remember, we looked hard for contrary findings. To me, this is about as persuasive a body of evidence as one can reasonably expect.

You say that you want to spread optimism, and many of the suggestions you make for dealing with the problems you identify involve accentuating the positive and downplaying the negative. This is so true that many of those leaving comments to your piece in The Wall Street Journal seem to think that you are largely advocating 'The power of positive thinking'. Is this not close to suggesting that people should ignore or downplay bad rather than assessing it rationally and proportionately? Should we not strive for realism rather than either optimism or pessimism? 

I agree completely. I am in favour of realism. John Tierney and I both detest the simplistic 'power of positive thinking' sort of advice. At best it is useful as a temporary antidote to the negativity bias. But we favour seeing the world as it is. (To be sure, a case can be made for putting a slightly optimistic spin on things, as in Shelley Taylor's point that normal, healthy people do make pervasive albeit small distortions in a positive direction.) With learning, for example, we note that it is indisputable that getting both positive and negative feedback is the most helpful, because it gives you the most information needed for learning – learners learn fastest when they know both what they are doing right and what they are doing wrong. True, if only one kind of feedback is available, then it seems that one learns faster via punishment and criticism than via reward and praise. But getting both is the most effective. 

Your book contains many attention-grabbing evaluations and claims. You lambast educators for moving away from 'harsh punishments' and religious leaders for ceaselessly 'preaching benevolence'. You extoll 'competition, not collaboration'. You reassure readers that you are 'not suggesting a full restoration of Victorian practices' because 'it's a very good thing that beatings are no longer routine', but you nevertheless 'advocate less carrot and more stick' [emphases added]. You declare that 'we've escaped the Four Horsemen' of [premature] Death, Famine, War, and Pestilence and only 'imagine that terrorism or nuclear weapons or climate change pose an existential threat to civilization or life on earth' [emphasis added]. How would you respond to claims (e.g. by Stuart Jeffries in reviewing your book for The Guardian) that many of your views and recommendations are 'reactionary' rather than balanced conclusions from the available evidence? 

By 'reactionary', do you mean a political label referring (pejoratively) to ultra-conservative viewpoints? Neither John nor I hold that political outlook. We do have various political views that span the full left-right political spectrum as normally understood. And I suppose that strong ideologues of either far left or far right would find something to object to in our analysis, prompting them to accuse us of belonging to their opponents. 

Nevertheless, I would like to pass this question on to John.

JT: If by 'reactionary' you mean someone who resists change and progress, we hardly fit that label either. In the book, we discuss how people have been deluded for thousands of years by the 'Golden Age fallacy' – the belief that the past was better than the present, and that the future will be still worse. Focusing on the past three centuries, we show how measures of human well-being have been consistently improving while intellectuals since Malthus have been consistently making mistaken predictions of doom. So I don't understand how we can be accused of ignoring the 'available evidence'.

We don't deny that there are problems in the world, among them terrorism, climate change and nuclear weapons. But we also recognise that humans have historically overcome problems, and that today we have more knowledge and better technology than ever to deal with new challenges. One can always argue, of course, that history is no guide to the future, and that the latest doomsday prediction will come true. But we prefer to be guided by long-term historical trends, and we're willing to back our confidence by following one of the strategies we recommend in the book: betting on the future. 

In The Power of Bad, we urge doomsayers to put their money where their doom is. If someone believes that any of the long-term positive trends in human well-being – such as measures of life expectancy, health, nutrition, literacy, education, violence – will reverse, we suggest you go to the Long Bets website and propose a wager. You can choose any measure and any future date. We're willing to bet that the measure will show continued improvement. 

Looking to the future, now that you have detailed how 'we can flourish despite the power of the bad', and you have previously answered such questions as 'Why is there evil?', 'What is the self?', 'What shapes human nature?' and 'What is the meaning of life?', which question do you anticipate turning your attention to next? 

Currently I am about 80 per cent finished with a book on the self, so the question 'what is the self?' is a central challenge in my current work. The meaning of life issue I addressed in a book published almost 30 years ago, but I do have in mind a project to revisit that question with all the accumulated scientific evidence from the intervening years (not to mention my own presumably greater intellectual maturity as compared to when I first grappled with this question!). 

Down the road, I have long thought that the problem of cultural change is one of the ultimate interdisciplinary social science challenges, and I hope to write a book on that topic, assuming I live long enough and can gain enough knowledge to formulate something novel. In the mean time, I am also occupied with the question of political bias and hostility, hoping to formulate some integrative understanding of what the left and right have to contribute and why they so dismally fail to appreciate each other. Last, for some years I have been working on the psychology of time, seeking to understand how the mind constructs the future and uses those thoughts to guide behaviour in the present. 

The Power of Bad by John Tierney and Roy Baumeister is published by Allen Lane.

See our archive for more from Dr Baumeister, on self-control and rejection.