In the September/October, 2003, issue of the premium magazine for bird watchers, the Bird Watcher's Digest, there appeared a posthumous article by the renowned Eirik A. T. Blom on the subject of the Ivory Billed Woodpecker. This article remains available, free, at
I urge you to read this article, because it is exactly what and how a reasonable person writes. The article bristles with common sense, a profound understanding of causality, and decades of experience with birds. Blom told us that the ivory billed woodpecker is extinct.
On April 28, 2005, a team searching the Big Woods region of eastern Arkansas reported that they had seven sightings of an Ivory Billed Woodpecker collected throughout 2004. John W. Fitzpatrick, director of the Cornell Lab of Ornithology in Ithaca, New York, perhaps the premier ornithologist in the USA, reported the findings and the identification. He noted:
a) The bird's size matched the species's estimated 19.5-inch (50-centimeter) length, from beak to tail tip. The length of the tail was particularly revealing.
b) The bird's wing patterns, both at rest and in flight, had the black-and-white markings characteristic of an ivory-billed woodpecker.
c) The bird's back had a conspicuous area of white plumage.
There was also video, though fuzzy, of the bird in flight.
I saw this news on the internet, watched the fuzzy video, and sent out an email to dozens of friends, fellow birdwatchers and coyote mentoring graduates about this spectacular, dream-come-true event.
A few weeks later I wanted to watch the fuzzy video again and searched for it on the internet. What came up were many matching sites, one of them for a skeptic. Out of idle curiosity, I went to the skeptic's site to check it out, figuring that this must be a kook, conceitedly arguing with a premier PhD ornithologist about a bird several people had seen and which had been videotaped.
The skeptic's name was Tom Nelson. He has a master's degree in electrical engineering and works in Minnesota, where is also an avid birdwatcher. He had some doubts about the existence of these sightings in eastern Arkansas. He also had a link to some pictures. As many of you know, the new digital SLR cameras can be “locked on” to take dozens of pictures. This had been done with a pileated woodpecker about to take flight from a tree limb. The camera was nearly under the tree, looking sharply upward at the under wing feather pattern as the bird took flight. There were about 50 pictures of the first two seconds of the flight of this pileated woodpecker.
I left this internet site open and began a separate internet search for the Arkansas ivory billed woodpecker fuzzy video. I spent about 25 minutes comparing the fuzzy video with the dozens of pictures of a pileated woodpecker taking off into flight.
At the end of this examination, I could not say that the fuzzy videotape was certainly a bird other than a pileated woodpecker. This is a simple statement, but in science and in logical argumentation, it is dynamite. What I had observed was that the null hypothesis had not been dis-proven. The video evidence could be a known existing bird instead of a rediscovery of a spectacularly rare species.
Among my mentoring graduates are some particularly talented birdwatchers. One of them, perhaps the most skilled of all, was working at a wetland wildlife refuge on the shore of the Chesapeake Bay. I explained my examination and consternation. Did the pros have anything to say about this skepticism?
I got back an email saying that he would check.
Later I got some cut-and-paste emails between ornithologists. They were arguing about the Arkansas sitings. They weren't of a common mind. They weren't pleased by the discovery or the publicity. Some were sardonic or cynical. One mentioned being deep in the woods of eastern Texas and seeing a bird that he couldn't identify because it was so very different from anything he'd seen. He didn't want to risk his own credibility over such an iffy sighting. So he stayed quiet.
I also emailed Tom Nelson in Minnesota and reported what I'd found in examining and comparing the fuzzy video with the known stills of a pileated woodpecker takeoff. What have we got, I wanted to know, on the field marks of the Arkansas supposed discovery?
Tom was ruthless and thorough on his blog about the science behind the ivory billed rediscovery. At that time he was looking into the audio reports of calls that had been reported from deep within the swamp. A lab was going to compare them with actual 1935 recordings of ivory billed woodpecker vocalizations. There would also be expert examination of the fuzzy video.
Here's a summary of what I discovered.
The sightings were in good pileated territory but only marginal ivory billed territory due to the specialization of the ivory bill woodpecker's diet.
The audio evidence was very brief, very vague, and not a good match with the 1935 ivory billed audio. It could have been other birds, especially blue jays. My own contribution is that it could have been squirrels imitating (to quarrel over territory) with crows or blue jays. I have personally heard squirrels do this, and is sounds quite similar to the audio evidence.
The video was fuzzy. This gets us into the specialized area of color “bleeding” that occurs with fuzzy evidence. Some colors, like black, don't bleed. White bleeds and red bleeds (but differently from the way white bleeds). To make a very long story very short, an ivory billed woodpecker, filmed in flight from below with fuzzy focusing, would show the brightest white patch under the wings. A pileated woodpecker (which also has a lot of under wing white feathers), would nonetheless show the brightest white patch along the body (or “fuselage” in aviation terms). The fuzzy video showed the brightest white along the body of the bird.
Things were starting to lean toward a conclusion that this was a pileated woodpecker misidentified as an ivory billed woodpecker. And then the rafters fell in on the ivory billed woodpecker.
Unlike the pileated woodpecker, the ivory billed woodpecker has very stiff wing feathers. The feathers clatter in a distinct, noisy racket when the bird is in flight, at least as noisily as a grouse. I've seen pileated woodpeckers in flight, and they make no such clatter. None of the seven sightings in eastern Arkansas mention clattering wing noise. The null hypothesis, that existing birds can fully account for the evidence, is thus fully demonstrated. Dr. Fitzpatrick's team asked for and got several more years of field research in the Big Woods. Much of the observations were achieved through robot cameras and recordings. Nothing conclusive was found. There were no nests located. There were no ivory billed woodpecker feathers or eggshells or scat.
The ivory billed woodpecker is still extinct.
The US Fish and Wildlife Service has blown $14 million looking for evidence to back up this amateurish, erroneous sighting. Dubious and hysterical additional sightings have not helped conclude this matter efficiently. Ron Rohrbaugh, from Cornell, who continues to lead the original research team, has not announced that the sightings were wrong nor that the ivory bill is extinct.
Not only is he wrong, but he won't and can't admit it, which is exactly what happens when cognitive bias takes over.
Once Fitzpatrick sided with the ivory bill identification, authority bias took hold. This also caused the herd instinct to kick in among the researchers. So did the clustering illusion, the tendency to see patterns where actually none exist.
Visually, the availability heuristic, observer-expectancy effect, Pareidolia and ingroup bias all convinced the researchers that they were seeing a rediscovered bird of great rarity. This puts us at the edge of apophenia.
And these are scientists.
And they unapologetically blew $14 million of public money.
You might also add (though it doesn't relate exactly to your specific argument) that virtually all of these sightings occurred when the observer was alone, just like virtually every Bigfoot sighting. Thus we have about as much evidence for the current existence of the ivory-billed as we do for Bigfoot. There are a lot of dubious eyewitness accounts and one fuzzy video (corresponding i suppose to the Roger Patterson film of Sasquatch).
ReplyDeleteThe late Eirik A. T. Blom, in his September/October 2003 Birdwatcher's Digest article linked in the blog post, says the Bigfoot sightings actually outnumber IBW sightings.
ReplyDeleteThere's an old, reliable, unrepudiated 19th century rule for an indisputable rare bird sighting -- definitive field marks simultaneously by three observers (or shoot the bird dead and bring it in as evidence). Even after the press conference in 2005, the further field work did not put interns together in teams of three for live sightings. "That's not good science."
see also http://news.mongabay.com/2007/0314-ibwo.html
ReplyDeleteok, i finally had time to read the Blom piece. It's an excellent bit of critical thinking.
ReplyDeleteAs a life sciences student at a major American University in the last half of the 20th century, i can tell you that the field is loaded with cognitive bias both in the professors and students.
Most life science majors (zoology, ecology, environmental sciences, biology, and so forth) are there to confirm their world view, to seek evidence of foregone conclusions, or to gain ammunition for political activism, NOT to learn about the life sciences per se.
The exceptions seem to be the people who study and teach about very small things like microbiology or genetics.
We have produced at least two generations of non-critical scientists and political activists. Even Carl Sagan got into the act with his promotion of the "nuclear winter" which is based on a formula with variables we do not know the values of.
It's really pathetic what's going on in universities today. And i think the same can be said of other sciences and journalism.