We mistakenly reason in terms of the whole and the average, without realizing that what really matters lies in the extremes and their frequency. Yet today, “standard deviations”—which measure the dispersion of data around the mean—are increasing in nearly every field, just as the frequency of extreme values is rising.
The average is also reassuring because it obscures the outliers and their frequency, even though their consequences are significant. This is easiest to understand when it comes to temperature. No one would really have a reason to worry if the local temperature rose by just one or two degrees.
But a global average is meaningless: a global average increase of 2 °C masks considerable variations across regions, seasons, and extreme events. Land areas are warming more than the oceans, and northern regions more than tropical regions; certain mid-latitude or semi-arid regions will experience warming at twice the average rate; Arctic warming will be at least 2.4 times the global average and even higher in some parts of the Arctic; in winter, Arctic warming could even be four times greater than summer warming. Finally, summer warming in the Mediterranean will be well above the global average, due to soil drying and atmospheric feedbacks. More broadly, land areas will warm more than the global average, while the oceans (particularly the Southern Ocean, the eastern Pacific, and the subpolar North Atlantic) and small islands will warm less than the global average.
In other words, if the average increase is limited to 2 degrees and the standard deviation is 10, temperatures in extreme locations will be 7 degrees higher than current levels and, in some regions and at certain times, may even be significantly higher.
Furthermore, beyond the standard deviations, the probability of extreme events occurring is also important—a factor that neither the average nor the standard deviations reveal. However, it turns out that the hottest temperatures are becoming more frequent, while the coldest temperatures are becoming less frequent. We will therefore see an increase in average temperatures, a widening of the range, and a concentration of occurrences toward the extremes.
Extremely hot days will be hotter, more frequent, and concentrated over inhabited areas; whereas extremely cold days will be colder, rarer, and concentrated in uninhabited or uninhabitable regions such as the oceans. The climate problem is therefore much more serious than a comparison of local average temperatures would suggest.
This is not true only for temperature. While many poor countries have closed part of the gap in terms of life expectancy, this is not the case in most other areas, where catastrophic situations are becoming both more catastrophic and more frequent. Biodiversity is declining very unevenly across regions, habitats, and biological groups. Some species are going extinct; many others (particularly plants, including cycads, certain trees, certain conifers, and birds) are seeing their populations collapse, while thousands of still-unknown species are discovered each year. Moreover, life is first and foremost a network of interactions between species, and the extinction of one has an immense impact on all the others. The average means nothing. While the number of people with access to safely managed drinking water has increased globally, the number of people excluded from such access has risen and will increase significantly in low-income countries; and the range of conditions will become wider; shortages will be more extreme; and extremes of scarcity will be much more frequent (and much more severe) than extremes of abundance.
This is also true of educational attainment. Average achievement is declining across the OECD, but the variation in outcomes among students, social groups, regions, and socioeconomic levels is also worsening. While 57% of children in low- and middle-income countries were unable to master basic reading by the end of elementary school before the pandemic, that figure has now risen to 70%: children who read poorly are reading even worse than before, and there are more of them. The same is true for math. Here again, not only are standards falling, but the gaps are widening, and the frequency of the most extreme failures is rising very rapidly.
The same is true for homicides, where the global average is falling while disparities between countries are increasing and, where violence is on the rise, it is more extreme and more frequent than ever. This is also true in the media, where the average tone of news coverage is shifting toward negative stories, with increasingly wide gaps between news items—whether true or false—that are becoming more and more anxiety-inducing and more frequent, while positive news, though growing in number, is becoming increasingly rare or, at any rate, less widely reported. A crime, an accident, or a murder has more impact than ten police successes or ten major scientific discoveries.
This explains what is happening in politics. Citizens remember (and are more affected by) extreme events—even rare ones—more than they do ordinary events, even frequent ones. If the average of public opinion shifts slightly to the right, the gaps between the far right and far left grow wider, and extreme positions occur with increasing frequency. Moderate opinions—that is, those of the centrist parties—no longer reflect the true state of the world.
Until they understand this and break free from their gray mediocrity, the leaders of the so-called “governing” parties will fail. They must propose bold ideas and rise above the mean—not by veering to the sides, like the extremists, but by reaching higher, like statesmen.