Thursday, June 21, 2007

Those who live by empiricism will die by empiricism...

:-) Note smiley, and add smilies throughout. I'm not having a go at anyone here.


Bill came up with sensible comments below in the 'painting the ceiling' post. He ends...

I do wish Cornell would come back up to the plate on this. After all, it's their video, not mine. I've never even claimed to have seen one of these freekin' birds yet I've still been awarded an honorary "crow."

Seems a bit unjust to be given a crow-tag. Maybe someone at Cornell will come up to the plate (is that the same as assuming responsibility for a penalty kick in the shoot-0ut?). But let's face it... it's futile because it is GAME OVER for Ivorybills. Sure it is possible to pick holes in analysis of video artifacts, wingbeats 4-8, deinterlacing, but it does nothing more than delay the obvious which is that Luneau bird was not an Ivorybill. If anyone hangs their hopes for Ivorybills on things that weren't explained in previous papers then those hopes will only last until someone publishes evidence that addresses wingbeat 4-8, deinterlacing and video artifacts. At this point, with the 'double knocks' taking a hammering from gadwalls, the 'kents' being made by anything and not even matching IBWOs very well, all the robobirders picking up extant species of woodpecker, nobody being able to see an IBWO properly, IBWO searchers having to doublethink their way round IBWOs being impossible to see because they are nomadic AND impossible to see because they have very restricted ranges in vast habitat AND have become more difficult to see since they were on the verge of extinction 70 years ago. FACT - IBWOs were never that difficult to see. FACT - their calls were incessant and carried over half a mile. FACT - their calls do NOT carry over 60 years echoing round the woods since the last ones died! No amount of technical rebuttal of any paper will change those things. There is nothing left except a series of sight records that, in isolation, are starting to look like a series of very understandable misidentifications. It was a nice dream while it lasted, but it's time to wake up now. There, I said it. ;-)

13 comments:

Ilya Maclean said...

So that fence was finally hurting your arse:-)

wstrnstar said...
This comment has been removed by the author.
Bill Pulliam said...

Up to the plate: baseball batter comes up to home plate to attempt to put the ball in play. So yes, similar to the football analogy.

Hey, as soon as someone produces any video, super 8 movie, series of still shots, what not that shows any bird that simultaneously matches CLOSELY (not just approximately, sort of, I guess so, if you squint and disregard and wave your arms a lot) the moderate amount of flight dymanics and small amount of plumage pattern that is discernable in the vid, accounting for imaging and processing artifacts in a way that can be shown to be reasonable not just by yada yada yada, I'll be happy (relieved) to accept the ID whether it's a Pileated, a hornbill, or a crumple-horned snorkack. Until then, the bird remains unidentified, with tangible, discrete, as-yet-incompletely explained-by-alternatve-theories indications of being a large B/W woodpecker that do not match a PIWO.

You realize you're making the same falacious "body of evidence" argument that is criticized in the other direction: All these other bits of evidence are called into question, so the video is obviously bogus too. What do double knocks in Florida have to do with a video from Arkansas? Not a damn thing. The weakness of the body of evidence indicates that overall the global population is damn close to zero. If there was one valid encounter buried in there, it means that that global population is damn close to but still slightly greater than zero. Or at least it was three years ago. That difference between equal to zero and slightly greater than zero is huge. It's the difference between a patient being dead versus almost but not quite dead.

Martin said...

Hi bill

Hey, as soon as someone produces any video...

I can accept that. Hold that thought.

I accept paragraph two as well (paragraph 3 if you include marching out to the wicket). What do double knocks in Florida have to do with a video from Arkansas? Not a damn thing. True. Especially if the double knocks are ducks in Florida and the video is a Pileated in Arkansas ;-) and millions of dollars of conservation funding are being wasted on them :-O

Bill Pulliam said...

It's true that 95% of the claimed IBWO field marks in that vid wither under the spotlight of closer examination, but 100% of the supposed PIWO field marks evaporate entirely under that same bright light. Folks who still cling hopelessly to their black trailing edges and "we may have never documented it, but a pileated COULD fly that way if it wanted to!" arguments are just as sad to watch as people who still believe that double knocks and odd squeeks are definitive evidence. Nearly everyone involved here is doing the same thing: massaging dodgy data and discrediting or ignoring inconvenient bits to support whatever opinion they held as of March, 2006, while projecting all this purely on to those who disagree with them. It's the human way. Meanwhile the great masses have lost interest and moved on to juicier topics long ago.

I think there's only one more published paper to be squeezed from the Luneau vid. That's the formal presentation of the sort of image artifact and flight mechanics issues I've been babbling about for the last year or two (plus the inevitable rebuttal). If Cornell never get around to writing such a paper, us "amateurs" will just have to. There really hasn't been a fundamentally new idea about this vid advanced in public recently, and those are all the old ideas that are left that haven't yet made it in detail into a proper peer-reviewed journal (where Sibley et al will no longer be able to ignore them). Then it just become history for future readers with more hindsight and less emotional involvement to sort out.

Ilya Maclean said...

Bill / Martin. Would it not be fair to say that no matter how hard you look at or analyse the Luneau vid, the fact remains that it consists of a few blurry pixels in motion? Fundamentally, the bird is unidentifiable and no matter how you look at it, nor any subsequent events will change that fact. I don’t really see that there is anything to be gained by arguing the details of the video further. The whole arguing of the details things smacks a bit too much of academic masturbation.

I would add this though. The burdon of proof lies with Cornell, as they published the paper and it should be up to them to back their claims-up their claims with credible evidence, not up to others to eliminate the uncertainty. Consequently I disagree with the statement that “Folks who still cling hopelessly to their black trailing edges and we may have never documented it, but a pileated COULD fly that way if it wanted to! arguments are just as sad to watch as people who still believe that double knocks and odd squeeks are definitive evidence” as it fundamentally ignores where the burdon of proof should lie.

Bill Pulliam said...

This is not a legal case, though you can't tell it from the way most people treat it. This IS an academic, scientific discussion. It's not about "burden of proof." It's about evaluating evidence and attempting to extract as much real information from it as possible. There's no judge and jury here. Refering to efforts to determine in depth exactly what can and cannot be determined from that video as "academic masturbation" smacks of being antiscientific.

So, Ilya, I'd have to say, I disagree with your entire comment, and the black/white adversarial assumptions underlying it. This should be a collective attempt by the community as a whole to evaluate the evidence that has been acquired. It should never have become a dispute between warring camps. There will always be disagreement, but there do not have to be battle lines.

Ilya Maclean said...

Bill - it has nothing to do with legal norms. I was referring to academic norms (I am an academic). When one publishes a paper in a scientific journal it is up to those publishing to present sufficient evidence to support their arguments and not up to others to disprove them. In the case of Cornell's paper: Ivory-billed woodpecker persists in continental North America (the title of their paper). It is this way round for very good reason. If it wasn't, one could make entirely ridiculous claims such as “aliens persist in continental Europe”, supported by weak evidence.

I don’t really see that there are any black/white assumptions underlying this statement, it’s just that its scientific norm not to publish “grey” ideas in “black” literature.

Bill Pulliam said...

But at this point you are speaking of a paper published a couple of years ago. Their data (sightings, audio, video) have long since permeated far beyond the original ms and its rebuttal. There's nothing at all unusual about other authors reanalyzing previously published data for many years after its original presentation. Your statement (paraphrased) that there's nothing useful to be gotten from the video is simply an opinion, not an objective fact. I think I have highlighted several important points recently that had not been presented in explicit detail in either the original ms or the rebuttal, and that have substantial bearing on their conclusions. You say that the CLO should have initially presented better support for their original conclusions. Fine. But, they didn't. That's the way it is. Done deal, two years ago. Are we supposed to rewind history so they can do it right? Or just totally ignore what information they did present? That's just silly. The important matter here is not if, how, when, and why CLO did what, but how much evidence is there, in fact, that this species was not extinct as of spring, 2004. I'm not going to drop that question on a procedural technicality of whether or not CLO should have presented their info in the way they did.

Ilya Maclean said...

At the risk of being hypocritical (about academic masturbation:-)

I think your blending my two points Bill. My first was that since the bird is unidentifiable and shall forever remain so, I don’t really see that there is anything to be gained by arguing the details of the video further. I stand by that, but concede that it may have a bearing on the probability of it being one species or the other (or another altogether), but not by a particularly useful amount. My reasons for stating this, is that the video is only useful if it can demonstrate with a very high degree of certainty that the bird is an IBWO. Unless positive ID as an IBWO is, or is very close to, 100%, the probability of the bird in the video being an IBWO is actually more influenced by the probability of IBWOs persisting in the area (itself influenced by the video). The whole argument becomes circular and the probability of it being an IBWO is very low. Say for the sake of argument that PIWOs outnumber IBWOs by 100 to 1, even if we can be 95% certain that the bird in the video is an IBWO (all other things being equal), then there is still a less than 1% chance that the bird in the video is an IBWO (95^100/100^100). I don’t think we’ll ever reach that kind of certainty, making it highly probably that PIWOs outnumber IBWOs by more than 100 to 1, making the probability even smaller, etc etc.

My second point had nothing to do with the strength of Cornell evidence, but merely with where the burden of proof should lie and why disputing claims is not the same as making them.

Bill Pulliam said...

It needs to be kept in mind that this video is not from just any random time and place. It is only even being examined because it comes from the same place and the same time period as a cluster of reported IBWO sightings by what seem to be competent observers.

It is also not ultimately the judgement of you, me, or Martin that matters. It is the judgement of people with decision-making positions in agencies, organizations, review committees, etc. that matter. My primary goal is to disect and analyze the video thoroughly and accurately as an aid to them and the community as a whole. It is less important to me that people read what I write and say "oh yes, of course, it's obviously an Ivorybill" than to read what I say and understand the points I made and the interpretation issues I describe, then form their own well-informed judgement based on the collection of analyses from everyone. Much of what has been published about this video is, in my opinion, superficial and poorly-researched misinterpretations, and that must be addressed.

Bill Pulliam said...
This comment has been removed by the author.
Bill Pulliam said...

%^$& blogger keeps submitting my comments twice.. grrr.