The Extinction Puzzle

With six billion humans and counting – and our individual consumption growing – we face enormous challenges in using our natural resources sustainably. And there is one clear measure of how we perform at that task: extinction rates. For this reason, some colleagues and I have calculated the rate at which bird species became extinct in the recent past and the rate at which they will likely become extinct in the future.

Bird extinctions are our best window onto humanity’s massive and irreversible environmental impact. For every one of the world’s 10,000 bird species, there may be 10 or perhaps even 100 other unknown species of animals, plants, or fungi. Birds’ popularity means that they provide an unrivaled source of information about which species live where and how well those species are doing.

Estimates of the number of extinctions have tended to vary wildly, owing to differences in the assumed total number of species, which range from the one million species that have scientific names to an implausibly high estimate of 100 million species. To avoid such uncertainties, my colleagues and I introduced the extinction rate – the number of extinctions per year per species or, to make the numbers more reasonable, extinctions per million species-years — E/MSY. With the exception of the past five mass extinction events, estimates from the fossil record suggest that an approximate background rate is one extinction per million species-years, (1 E/MSY). For the 10,000 bird species, there should only be one extinction in each century.

Before European exploration, the Polynesian expansion across the Pacific probably exterminated species at the rate of one every year or two, which is 50 to 100 times more extinctions than should occur naturally. For example, parrots, rails, and doves once occured across the Pacific on sufficiently large islands. Those islands that have been well explored archaeologically yield bones of species the Polynesians exterminated, while the islands now lack them.

We can predict which islands were likely large enough to have once supported unique species and so estimate how many species are missing. The total is nearly 1,000. We have yet to do comparable analyses on the fate of Caribbean and Indian Ocean islands following first human contact, but it is likely that large numbers of extinctions occurred there, too.

Early European exploration likely continued this rapid pace of extinction. The number of species known to have expired between 1500 and 1800 is increasing as taxonomists describe new species from skeletal remains they find in caves, for example, where bones survive. These searches are far from complete.

Subscribe to Project Syndicate
Project Syndicate Digital Premium Image

Subscribe to Project Syndicate

Enjoy unlimited access to the ideas and opinions of the world’s leading thinkers, including long reads, book reviews, topical collections, short-form analysis and predictions, and exclusive interviews; every new issue of the PS Quarterly magazine (print and digital); the complete PS archive; and more. Subscribe now to PS Premium.


Linneaus invented the modern process of naming species in the mid-1700’s, although naming did not get under way until the peace following Napoleon Bonaparte’s defeat in 1815. Whereas bird taxonomists had described only 2,000 bird species by 1815, in the next 50 years they described about 5,000 more. This means that most of what we know about birds and their extinctions dates from the 1800’s. To get an accurate picture of extinction, we must adjust the total number of extinctions for the length of time science has known bird species.

A second adjustment is necessary because conservationists follow the principle that a species survives even if it is not recently recorded in its native habitat. This assumption prevents terminating conservation efforts prematurely, even as it underestimates the total number of extinctions. We added species that are probably extinct, too. With these two adjustments, the extinction rate has been about a 100 times higher than natural over the last two centuries.

Our predictions of twenty-first century extinctions jump an order of magnitude to 1,000 E/MSY. Not only are 12% of bird species already in danger of extinction, but more than twice that proportion have small geographical ranges (the key predictor of threat) and mostly live in habitats that are seriously threatened. Tropical forests, for example, hold the majority of bird species – and likely most other kinds of species too. If present trends of deforestation continue, we project bird extinction rates to climb to 1,500 E/MSY by the last decades of this century.

Moreover, our calculations are conservative, as they do not include extinctions caused by invasive species, the expansion of human technologies (such as the long-line fishing that harms many seabirds), or global warming. All would increase our estimates.

Likewise, birds are poor models for other kinds of species, because public concern over birds generally leads to rapid actions to save them. Indeed, we showed that aggressive conservation actions have reduced bird extinction rates over the last few decades to about 25 E/MSY – still very high, but down sharply from the 100 E/MSY projected were conservation actions not in place.

Certainly, some other species also benefit from the protections that we afford birds, but many other species do not. Moreover, human actions threaten substantially higher proportions of other well-known groups – such as flowering plants and amphibians.

Our results show that conservation efforts can reduce the irreversible loss of biodiversity. Nevertheless, future rates of extinction constitute a problem that will be very much harder to solve.