MYTHS OF THE ALIENS - The New Wild: Why Invasive Species will be Natures Salvation - page 69

“The high ranking of non-native species as an extinction threat was due almost entirely to the inclusion
of Hawaii,” which “clearly has a dramatically different invasion history” from most of the rest of the
country. 16 The extent of this skewing is remarkable. Hawaii makes up 0.3 percent of the area of the
United States yet was responsible for nearly two-fifths of the “imperiled” birds and plants in Wilcove's
data. 17
So, in effect, Wilcove presented data from Hawaii as statistics for the United States as a whole, while
subsequent users, among them UN agencies, made out that they represented the entire world. Hawaii
does not equal the world. The conclusions are nonsense. This analysis, incidentally, is also the usually
cited source of the oft-repeated claim that invasive species are the world's second biggest threat to glob-
al biodiversity, after habitat loss. That claim may or may not be true, but Wilcove's paper certainly does
not show it.
Wilcove is not primarily to blame for this. His paper was scrupulous about underlining how little
data he had to go on, saying there were “few quantitative studies of threats to species.” And he made
clear that “the attribution of a specific threat to a species is usually based on the judgment of an expert
source.” When I asked him if he agreed with me that that the conclusions drawn by others could not be
justified by his paper, he replied: “You are correct; my 1998 paper focused strictly on US endangered
species (not extinct species). I don't know the source of the 40 percent figure for animal extinctions.”
All he knew was that it wasn't him. So one of the most widely quoted claims about the damage done by
alien species apparently has no basis in published science.
I wanted to track down another “fact”—since deleted—on the UK invasive species secretariat's web-
site. Apparently contradicting the previous one, it said that “invasive non-native species are a known
factor in 54 per cent of animal extinctions and the only factor in 20 per cent.” 18 Again the trail led back
to a single source, this time a paper published in 2005 in the journal Trends in Ecology and Evolution by
Miguel Clavero and Emili García-Berthou of the University of Girona, in Spain. Titled “Invasive Spe-
cies Are a Leading Cause of Animal Extinctions,” it had been cited 444 times in other papers, according
to Google Scholar. Not bad for a paper that is just four paragraphs long. 19
This paper did, at least, deal with extinctions rather than threats. The authors got their percentages
by accessing the IUCN's database on known species extinctions and checking out the attached notes on
what may have caused them. But the authors admit that only a quarter of the 680 listed extinct species
had any such notes. The rest were silent on the matter. So it is not true that nonnatives were a “known
factor” in 54 percent of animal extinctions. Nor is it true of those extinctions that got into the database.
It is similarly untrue that scientists know that aliens are the “only factor” in a fifth of animal extinctions.
And, given what we know about the biases in how invasion biologists choose their research topics, it
seems to me highly likely that the 170 cases where there is a note in the database about likely causes of
extinction will be biased toward cases involving aliens.
I thought I might try to check this potential bias by finding out which species the authors had in-
cluded in their list. The four-paragraph paper offered no clues, and, when I contacted them, the authors
said they could not provide a list. They did not have one. Nor could they say in detail how they had
done their analysis. So I could not establish how individual extinction cases reached the threshold of a
“known factor” or the “only factor.” Nobody that I could find had replicated this analysis. And since the
database had changed since their study, doing so now was probably impossible. That seems, on the face
of it, to break two of the cardinal rules of science: that other researchers should be able to try to replicate
the findings and that nothing should be accepted as fact until it has been replicated.
Clavero and García-Berthou presented their four-paragraph paper as a riposte to a previous—and
much more substantial—paper in the same journal by Jessica Gurevitch and Dianna Padilla of Stony