An interview with Paul S. Martin
It takes time to overturn decades of accepted scientific thought. For example, Paul Martin has spent 40 years accumulating evidence that humans triggered the mass extinction of dozens of species of large animals about 10,000 years ago in North and South America. His theory—that small bands of human hunters moved south from Alaska across the New World and rapidly hunted much of the large game to extinction—faced strong opposition when first introduced in the 1960s. Some researchers still argue that climate, not humans, caused the extinctions, but Martin now has much of the scientific community on his side of the debate.
Martin, an emeritus professor of geosciences at the University of Arizona at Tucson, recently published a book on the subject, Twilight of the Mammoths: Ice Age Extinctions and the Rewilding of America (University of California Press, 2005). In it, he discusses his many years of research, the reaction to his theory and the implications of his work for future wildlife conservation efforts. "Without knowing it," he writes, "Americans live in a land of ghosts." He proposes introducing animals such as elephants to parts of North America where their extinct relatives once roamed.
American Scientist assistant book review editor Amos Esty recently interviewed Martin by e-mail and asked him to talk more about his work and his new book.
In Twilight of the Mammoths, you present a lot of evidence in support of the overkill hypothesis, but you also point out that there are a number of skeptics. What would you say is the strongest piece of evidence linking the arrival of human beings to the extinctions? What really convinced you that you're right?
The global extinction pattern, which Ross MacPhee of the American Museum of Natural History terms "a deadly syncopation," convinced me. Near-time extinctions [those that occurred within the past 50,000 years] of large animals swept Australia over 40,000 years ago, peaked in America 10,000 to 13,000 years ago and ended historically with the settlement of remote Pacific islands, which saw within the last two millennia the elimination of many rails [a family of birds] and native land snails.
As you make clear in your book, the cause of the extinctions is a hotly debated issue. What makes it so contentious? What's at stake?
Of the many revolutions over billions of years in the history of life, humans appear to be agents of extinction in just the last 50,000 years. Many vertebrate paleontologists consider climatic change a satisfactory explanation in most of the last 65 million years. The cause of extinction may be less well understood than the origin of species.
Is the dispute primarily the result of a lack of data?
I would say no, and I suspect that the reason for the dispute would trigger as much disagreement as the dispute itself. [Historian and activist] Vine Deloria's trashing of overkill reflects (in my view) his lifelong war against anything that he believes might reflect badly on Native Americans and/or upset those who do not believe that Native Americans were always here (in the New World). I doubt he was ever seriously interested in the extinctions during near time or in the problem of what caused them.
How do the other continents fit into this debate? Is the matter more settled for Africa and Eurasia?
In Eurasia the extinctions struck early in the Pleistocene in the southern part of Europe and advanced slowly toward the pole. The very last woolly mammoths on Wrangel Island in the Arctic Ocean only 4,000 years ago. The last woolly mammoths in Beringia were still alive on St. Paul's island in the Pribilofs just over 5,000 years ago. Both Wrangel Island and St. Paul's Island were apparently unknown to prehistoric hunters until thousands of years after the spread of Clovis hunters into North America.
In Africa, extinctions of large animals—especially of taxa of proboscideans, large primates and suids (pigs)—may span a million years or more. I know of nothing like this in South America, where large mammal extinction in the Pleistocene appears to have been concentrated only 11,000 radiocarbon years ago or later.
Did you realize when you first proposed the overkill hypothesis in the 1960s that you would be stirring up so much trouble? What has it been like to be at the center of such a heated debate?
I did not expect to find so much resistance among some anthropologists and vertebrate paleontologists. The resulting debate has been stimulating—what would new research on prehistoric extinctions reveal next?
You also discuss the controversy about whether Clovis people were the first human beings to arrive in North America, about 13,000 years ago. What would the implications be for the overkill hypothesis if harder evidence of pre-Clovis settlements were found? Would it force you to rethink the possibility of overkill?
If many indisputable pre-Clovis archaeological sites were to be discovered, I would have to rethink the "overkill model." Claims of pre-Clovis invasion familiar to me make entertaining TV shows but do not appear to be robust.
Why is a short overlap between the arrival of humans and the extinctions so important to overkill? How much of an overlap would make it untenable?
If mammoths, mastodons and humans coexisted in America for 10,000 years before the mammoths and mastodons vanished, the overkill model of a super predator entering a virgin land mass and triggering mass extinction of other large mammals clearly would not work.
As Ross MacPhee said, the pattern indicates a deadly syncopation on different land masses. As humans first arrive ("ka!") they rapidly increase in numbers and trigger megafaunal extinction ("tunk!") in a few thousand years at most: 50,000 years ago in Australia (ka-tunk!), 11,000 years ago in America (ka-tunk!), 5,000 years ago in Cuba and Haiti (ka-tunk!), and less than 1,000 years ago in New Zealand (ka-tunk!).
The debate about the extinctions, you say, "points to some fundamental differences in methodology, outlook, attitude, and training that could be as illuminating as the final answer to the extinction controversy itself." Does that mean that differences of opinion about the extinctions are more about how people conduct their research and draw conclusions than it is about the actual evidence?
This is a hard question. The attitudes, no less than the training, of those investigating evolution and extinction are crucial. Conservation biologists have long sought to restore wild places to what America allegedly was like in 1491, with no awareness that natural baselines may be at least 10,000 years older.
What makes 10,000 years ago a more natural baseline than, say, 10 million years ago?
Ten million years ago would put us into the Miocene with primates but no hominids as yet. Many Miocene extinctions did (of course) occur, but not because of hominids.
Have you had much luck convincing conservation biologists that they should take into account the long-term presence in North America of animals such as wild horses that only became extinct about 10,000 years ago? How would it change their approach on an everyday level?
Aware of the ravages of extinctions in the late Holocene, zoologists in New Zealand have begun to restock lost species with living relatives. The idea is gaining traction in the United States (see, for example, the article "Re-wilding North America," by Josh Donlan and others in the August 18, 2005, issue of Nature).
If you were suddenly put in charge of an effort to repopulate North America with species lost during the late Quaternary, where would you begin? What animals would you most like to see roaming the continent again?
I would like to see free-ranging elephants in secondary tropical forests of the Americas. Until the end of the Pleistocene, forest and savanna in the New World tropics supported three families of elephants. Observations of free-ranging Indian or African elephants in a reserve of secondary forest should reveal much about fruit dispersal and forest tree ecology under conditions more like those that prevailed before the megafaunal extinctions.
So would the benefits of rewilding the Americas be primarily scientific? What do humans stand to gain by reintroducing animals such as elephants?
To thrive outside Africa and Asia, elephants need sizable reserves like the Elephant Sanctuary in Hohenwald, Tennessee. Ecologists would like to learn more about elephant behavior in dispersing fruits in tropical and temperate forests. Conservation biologists are beginning to consider the importance of sizeable tropical reserves in the New World, which once harbored mammoths, mastodons, and gomphotheres, for the only surviving species, Asian and African elephants. For Asian elephants we need tropical American equivalents of Tsavo or Amboseli National Park, popular African elephant reserves in Kenya and Tanzania.