This installment will focus the media response to and online discourse about our paper (both the preprint and the peer-reviewed version) and how the 2005 Arkansas controversy and its aftermath helped shape our paper’s reception. It will also examine the way the discourse has privileged visual material, especially video, at the expense of the totality. As with the previous installment, I have chosen not to provide links to specific stories, blog posts, or social media. I don’t want to drive traffic or identify individuals or specific media outlets. Even some of the stories I have liked and linked to display the same tendencies. Earlier installments are here and here.

I’ll begin with an anecdote. After the publication of our paper, we did an interview with a major newspaper. Very early in the conversation, the reporter asked a seemingly rhetorical question suggesting the Arkansas reports had been “debunked”. I corrected the journalist, explaining that Arkansas remained controversial, that neither the authors nor the journal had retracted, and that the Fish and Wildlife Service had not changed its position on the evidence, delisting proposal notwithstanding. Things went downhill from there. I’ve had some bad experiences with media – having dealt with shock jocks and doing interviews on countless drive-time shows – but this was one of the worst. The journalist was rude and condescending. Personal feelings aside, the incident highlights the enduring impact of the 2005 debate, and the effectiveness of the Arkansas critics’ messaging.

The shadows of Arkansas extend to the non-issue of federal dollars discussed in Part 2. At least one prominent ornithologist and 2005 extinctionist (not “skeptic”, another Arkansas shadow that feeds a false binary narrative – skeptics v. “true believers”, science v. faith) raised this red-herring in response to the preprint. It re-emerged in articles about the peer-reviewed publication. As previously noted in Part 2, complaints about government spending – likely exaggerated in 2005 – have no merit in 2023. It’s unfortunate that, in the drive for the illusion of “balance”, major media outlets obtain critical quotes from “experts” (though it’s unclear how an academic ornithologist qualifies as an expert on budgetary matters) but don’t provide an opportunity to respond or rebut, let alone fact check the statements.

The Arkansas data included video, multiple sightings by qualified observers, and recordings of putative kent calls and double knocks. But the public discourse about the evidence tended to reduce it to the video – often characterized as “blurry” – and little or nothing else. These mischaracterizations infected the scientific literature and, in at least one iteration, further reduced the physical evidence to a single image:

For example in 2005, based on a brief sighting and a pixelated image, the ivory-billed woodpecker was declared to have been rediscovered (Fitzpatrick et al., 2005), resulting in the mobilisation of resources for management strategies and recovery plans (Gotelli et al., 2012). However, based on the evidence its rediscovery was brought into question (Sibley et al., 2006), and subsequent extensive searches have failed to result in further sightings (Gotelli et al., 2012).[14] (Roberts DL, Jarić I. (2016) Inferring extinction in North American and Hawaiian birds in the presence of sighting uncertainty. PeerJ 4:e2426 https://doi.org/10.7717/peerj.2426 (Accessed January 17, 2024).

It’s remarkable that the authors were able to squeeze so many inaccurate assertions into a single, brief paragraph. The rediscovery claim was not based on “a brief sighting and a pixelated image”. It was based on a contested video, ARU recordings, and at least fifteen sightings by competent observers. The Fitzpatrick et al. 2005 paper included seven of the fifteen. 

Subsequent, extensive searches did not fail “to result in further sightings”. They failed to produce indisputable proof. There were unconfirmed sightings by members of the public in 2005-2006 (per the final report for that season). In addition, there were seven visual and eight acoustic encounters in a nearby Arkansas WMA in 2006-2007. One of the sightings was by a Nature Conservancy ornithologist. And that’s just in Arkansas. Suggestive audio was also recorded during this period and afterwards.

What’s most relevant here is the reduction of a contested video to a single “pixelated image”. Many other descriptions of the Arkansas data reduced it to a single “blurry” video, more truthful but not by much. Media treatments of our paper and some of the online discussions have often been almost identically reductionist and inaccurate. They have neglected the significant quantitative and qualitative differences between our data and Cornell’s (or anyone else’s). These differences include the number of birds involved, the almost 3-hour recording of kent calls accompanied by a few double knocks, the multiple trail cam images and videos obtained over a two-plus year period, and the history of encounters spanning more than a decade.

The issues with both the trail cam imagery and the drone videos relate to the limits of the available technology. Sensor size, focal length and distance from the camera, ambient lighting conditions, even early morning condensation on trail cam lenses all impact image quality. The imagery suffers from poor resolution, no doubt, but not so poor as to defy analysis.

The birds in the drone videos are obviously large woodpeckers; they could not be just ‘anything’, as one critic suggested in print. They are not Red-headed Woodpeckers, as one prominent birder and online critic initially argued. This was immediately obvious on my initial review of the captures, as the birds involved were clearly much too large.

Below are a couple of the trail cam images I find most compelling. They illustrate that graininess and blur are not the real issues.

Grainy and blurry are epithets, and their use reinforces the expectation that images of ivorybills must equal the quality of the famous Singer Tract pictures. It is self-evident that some will reject anything less than that, and given the eternally moving evidentiary goalposts for the ivorybill, I suspect that some would dismiss even images of that quality. But “why can’t you get higher quality imagery?” should be the beginning of a discussion, not the end of it.

Other evidence gets lost in the grainy-blurry sauce. And there are additional Catch-22s to contend with. We made a choice not to include sightings in the preprint. We did so in response to the post-2005 climate in which sightings have been dismissed, discounted, or attacked in various and sometimes contradictory ways. I’m a (retired) attorney. I know that eyewitness testimony is notoriously unreliable, but eyewitness testimony from trained and knowledgeable observers is something else again. We included the best individual sightings in the paper in response to peer review. The sightings make for good story-telling, and many media outlets handled the story-telling component fairly and well.

A few of the criticisms of the sightings echoed the claims of cognitive bias that were part of the Arkansas discourse. One extinctionist offered a watered-down version of this critique, suggesting that the only people who are seeing ivorybills are those who want to see them. This word choice may reveal more about the mental state and biases of the critic than anything else. But the newspaper accounts didn’t emphasize the number of sightings over years, their quality, or the competence of the observers. This enabled one critic to claim the mantle of science and play the old skeptic-true believer card, prefacing a comment with, “as a scientist . . .”, as if ours were an amateur effort.

In fact, our group includes several professional ornithologists. But in yet another catch-22, some critics have suggested that only ornithologists are competent and trustworthy in their identifications. Others have said that some ornithologists may lack sufficient bird identification skills. The team also includes high-level birders, and long-time Ivory-billed Woodpecker searchers. All have extensive field experience and are very familiar with the confusion species. The paper included only sightings by active team members (Frank Wiley’s excepted) where the observer had high confidence in the observation. (And we only mentioned in passing the many ambient single and double knocks and responses to ADKs that team members have heard regularly over many years.)

In one case, a poor quality cell phone photo accompanies and supports the sighting. The contemporaneous accounts address the confirmation bias issue. In my own sighting, I initially assumed that the bird was a duck, until I saw it tuck its wings. In that moment, I knew what I was seeing was an ivorybill. It’s the only time I’ve been sure.

As with Arkansas, the quality and duration of our observations equalled or surpassed the universally accepted sightings from the 1986 Short study in Cuba. That expedition obtained no physical evidence at all. The last photo from Cuba was taken by George Lamb in 1956. It’s only marginally better than some of our images. Granted, there are no other large woodpeckers in Cuba, so confusion with Pileated Woodpeckers is not an issue. Still, glare could result in confusion with the Cuban Crow. Issues of confirmation bias and group-think are at least equally salient in the context of the then extraordinary US-Cuban collaborative expedition.

At nearly 40 years’ distance, it’s hard to imagine how charged the experience must have been for the Reagan-era participants. The treatment of the Short expedition’s evidence contrasts starkly with the treatment of North American reports since Eckelberry, even when accompanied by poor photos. This is yet another example of the slippery evidentiary standards when it comes to the North American ivorybill.

Although the shadows of Arkansas have lingered and the old framing has shaped some responses to our paper, we have opened minds, made a strong case for persistence, and may have delayed or prevented the delisting. We have overcome some major obstacles, especially given the tainted discourse. Still, one wonders what it will take to convince some of the die-hards.

I’m planning at least one more post discussing the paper and the acoustic evidence. I may do a fifth installment on some of the trail cam imagery.

Here’s the news for 2024.