Title Author Date

James Downard Berlinski, David Jul 28, 2003

Some mammal-like characteristics reverse in the cladistic record and others jump over gaps. Such are Woodmorappe’s chief claims. Downard does not deny they are true so much as suggest they are unexceptional. In what follows I ignore small errors in detail in Woodmorappe’s paper, noting only that Downard’s critique contains errors of its own. (Thus Woodmorappe’s “widely contradictory,” (Woodmorappe, p. 10) becomes Downard’s “wildly conflicting,” (Downard, p. 8), a rewriting changing both the tone and the meaning of the original.) In what follows I simply note a few points of substance where Downard has failed to establish that Woodmorappe’s paper is preposterous.
Downard argues that Woodmorappe has not applied his own scheme of classification consistently (Downard, p. 12 0f 19), most notably by not distinguishing carefully characters whose polarity is unknown from characters that fail to be consistently progressing. I think Downward is right, but what he has failed to demonstrate is that Woodmorrappe’s conclusions are sensitive to such inconsistencies, his own discussion (paragraphs 1-4 on p 12 of 19) lapsing inconclusively. Without further argument, this must remain a methodological issue, interesting but incomplete.
Then again Downard argues that Woodmorappe’s discontinuities may be explained as artifacts, the thirty million years separating pre-therapsids from therapsids quite sufficient to explain what appears to be a mammal-burst in the record. But Downard has begged the very point at issue, and that is whether it is reasonable to suppose that in those thirty million years organisms missing from the fossil record filled in intermediate positions smoothly.
In still other sections of his paper, Downard’s zeal exceeds his good sense. Woodmorappe’s mammalness index may well be perfect nonsense (Downard, p 2 of 12), but not because numbers might have been replaced by alphabetical tags. It is the numbers that are added, not their names. A better argument is needed. And a better argument is needed again if Downard is to justify his indignation (p, 1 of 12) over Woodmorappe’s cool refusal to make ‘anatomical judgments,’ when, in fact, Woodmorappe has explicitly accepted the anatomical judgments made by his sources (Woodmorappe, p. 2).
Finally, the question whether Woodmorappe has offered a cladistic analysis or an analysis of cladistics seems to me to show only that Downard is insensitive to the nuances of the English attributive adjective. Although literary criticism is criticism of literature, it is not, after all, very often literary. What hinges, may I ask, on the distinction?
I would be glad in these exchanges to defend views that I hold, but there is something lunatic in asking me to defend a paper that I have never endorsed.

Related Articles: A Tale of Two Citations

Title Author Date

James Downard Berlinski, David Jul 30, 2003

Paris
To the editor
Re: Nilsson & Pelger’s derivation of the number of one percent steps needed to form an eye

In my Commentary essay of December 2002, I observed that Nilsson and Pelger provided no justification for their claim that precisely 1829 one percent steps were sufficient to transform an initial light-sensitive patch into an eye whose geometry was comparable to that of an aquatic organism.
There are two points of importance that must be stressed. In the first place, Nilsson and Pelger have made an historically important claim, most notably in view of Darwin’s own concern that his theory somehow accommodate the development of organs of extreme perfection. And in the second place, the claim that 1829 one percent steps are required to complete their proposed transformation represents the very heart of their paper and so its argument.
With this in mind, consider the following. Although Nilsson and Pelger do say that computations justifying their conclusion were made, those computations are a) not included in the body of their paper; b) not presented in their footnotes; c) not mentioned in any of their sources; and d) not developed in a technical appendix. Furthermore there is no indication in their paper that the calculations had been given to the Proceedings of the Royal Society of London, nor in the ten years since the publication of their paper have Nilsson and Pelger ever made their calculations available in any public format whatsoever. With this in mind I entertained the hypothesis – I did not nothing more – that Nilsson and Pelger had not performed those calculations for the full range of their sequence, but had instead undertaken calculations for a handful of points and completed their sequence by extrapolation. Whether this is what they did, I wrote, was something that I did not know. Nor do I know it now.
Mr. Downard writes that he has in fact been shown examples of Nilsson and Pelger’s calculations, but a) he has not chosen to publish these examples either, and b) examples, no matter how interesting, fall short of the complete sequence. That it would take many pages to show all of the relevant calculations is, no doubt, true, but it is also entirely irrelevant. Had John Woodmorappe used space as an excuse for not showing his calculations, Mr. Downard would properly have responded with contemptuous indignation.
I invite readers of Talk Reason to consult the editorial requirements posted by both Science and Nature. They are available on-line. Both journals demand explicitly that all relevant calculations be included in the body of the paper or made available in a public forum at the time of publication.
Under these circumstances, I see no reason to change my conclusions. Nilsson and Pelger have provided no justification for their most important claim.

Related Articles: A Tale of Two Citations

Title Author Date

James Downard Downard, James Aug 12, 2003

Spokane

Berlinski opted to ignore those “small errors in detail” littering Woodmorappe’s charts due to my “errors” (plural). While I usually quote extended passages verbatim, extracting text strings for consistency, I relied on my memory for the “wildly-conflicting” version of Woodmorappe’s “widely contradictory.” Guilty of dyslexic haste in quotes, as charged.

Then again, I became “Downward” in Berlinski’s next paragraph, and it is downward from there.

Though Berlinski’s grammatical charity was more elastic when berating my insensitivity for not reading “cladistic analysis” as “an analysis of cladistics,” how had my goof changed “both the tone and the meaning” of Woodmorappe’s claim that the cladograms differed unacceptably? As David neither mentioned nor challenged the details, his jujitsu view that I did “not deny they are true so much as suggest they are unexceptional” begs whether he (or anyone else) has any grounds to deem the paleontological condition exceptional.

Insisting “A better argument is needed” to nix the mammalness index, Berlinski defended his quicksand: “It’s the numbers that are added, not their names.” We differ here big time: those “numbers” aren’t numbers. A trait of “2” is a relationship, something that isn’t “1”—not one more of whatever “1” was. The numerals may be treated as numbers and pureed as Woodmorappe did, but to what gain in understanding?

The nub: rigorous discourse is a duet between general theory and specific examples. But Berlinski prefers to warble solo, “ardently” embracing the reptile-mammal transition without revealing the score on which his own tune rests. If anatomical, shouldn’t he have applied that scale to Woodmorappe’s paper and considered how the “gaps” (such as that sub-millimeter wiggling of a “reversing” bone) related to some range of natural variation he presumably theoretically accepts? Alas, Woodmorappe “explicitly accepted the anatomical judgments made by his sources” only as an evasion. Just as Berlinski paid no attention to them when dangling the piece—or Phillip Johnson before that, foraging at Answers in Genesis.

While it may be “lunatic” to ask David to “defend a paper that I have never endorsed,” does that mean his conduit Johnson isn’t accountable either? This would remind me of Thomas Nast’s famous Tweed Ring cartoon, where everybody tries to blame the next in line, were not Mad magazine’s Alfred E. Newman more apt: “What, Me Worry?”

Related Articles: A Tale of Two Citations

Title Author Date

James Downard Matzke, Nicholas Aug 12, 2003

Dr. Berlinski writes,

========
In my Commentary essay of December 2002, I observed that Nilsson and Pelger provided no justification for their claim that precisely 1829 one percent steps were sufficient to transform an initial light-sensitive patch into an eye whose geometry was comparable to that of an aquatic organism.

There are two points of importance that must be stressed. In the first place, Nilsson and Pelger have made an historically important claim, most notably in view of Darwin’s own concern that his theory somehow accommodate the development of organs of extreme perfection. And in the second place, the claim that 1829 one percent steps are required to complete their proposed transformation represents the very heart of their paper and so its argument.
========

Berlinski then go on to criticize Nilsson and Pelger for not providing the excruciating detail of each calculation for how they arrived at the 1829 steps.

But the procedure really is trivial. Nilsson and Pelger needed to calculate the number of 1% steps so that they had some approximate quantification of how much morphological change was required. They made no claim to this being "precise" as Berlinksi claims above.

Here is how N-P did their calculations:

In order to quantify the amount of morphological change, N-P constructed graphical models of various stages in the process (Figure 2) and decided to calculate the number of 1%-change steps in-between each stage. As an example, it takes 70 1% steps in order for a structure to double in length (due to the compounding of change -- think compound interest -- it takes only 70 steps rather than 100 in order for doubling to occur). They admit that there is some subjectivity in deciding *how* to measure morphological change, but they decide on the following as simple measures:

- length of straight structures
- "arc length of curved structures"
- "height and width of voluminous structures"
- changes in radius of curvature use the arc length of the inside and outside of the curved structure
- changes in lens refractive index above the starting point of 1.34

With this method they came up with 1829 1% morphological steps for the evolutionary sequence. They note that in actual evolution, some of the changes could happen simultaneously (e.g., lense development and aperture narrowing could occur together), but because they are being pessimistic, they restrict the steps to happen in series.

Note that only *after* this measurement of morphological change has been made, do they move on to calculating how long it might take for a population to undergo this amount of change. This can be discussed elsewhere

What, may I ask Berlinski, is so mysterious and dubious about calculating the change in length or width of something?

Related Articles: A Tale of Two Citations

Title Author Date

James Downard Wein, Richard Sep 08, 2003

On the subject of the reptile-mammal transition, David Berlinski is engaging in a form of sophistry which I call "argument by innuendo". It is based on the principle of "plausible deniability". He first introduces Woodmorappe's paper in support of his claim that "neither Darwinians nor design theorists can look to the fossil record with perfect equanimity", but, when that paper is challenged, he objects that "there is something lunatic in asking me to defend a paper that I have never endorsed." Far from being lunatic, a reader is perfectly entitled to expect Berlinski to defend a paper (in terms of either its content or its authority) which he has cited in support of his own claim. If he is not prepared to so, then his claim remains unsupported, and the citation of the paper has served no function other than to plant unjustified seeds of doubt in the minds of some readers. The planting of such seeds is, I believe, the purpose of Berlinski's tactic.

Now, Berlinski may respond that he did not explicitly state that he considers Woodmorappe's paper to support his claim. But his recall of that claim at the conclusion of a paragraph which discussed Woodmorappe's paper implies as much, and will certainly create such a belief in the minds of readers. What other purpose could this juxtaposition have had?

Berlinski may also respond that his claim is unexceptional. Indeed, no-one would deny that some aspects of the fossil record are puzzling. But, if Berlinski is making so unexceptional a claim, why make such a song and dance about it? Why bother to cite Woodmorappe in support of it? Clearly, the aim is to create in the mind of the reader a belief that something is seriously amiss with the mainstream interpretation of the fossil record.

So this is the argument by innuendo. Berlinski deliberately creates the impression that he believes Woodmorappe's paper poses a serious threat to the mainstream interpretation of the fossil record. But, when we examine his words carefully, we see that he has not explicitly committed himself to such a position, and so can refuse to defend it without appearing to have backed away from anything.

I should add that this is not the first time I have observed this device in Berlinski's writing. It is a favourite of his, and he is a master of the art.

Related Articles: A Tale of Two Citations

Title Author Date

James Downard Wein, Richard Sep 08, 2003

On a related point, I note that Berlinski casts himself as "a forthright defender of the reptile to mammal sequence". I would like to know if he can cite a single article, letter or internet post of his whose net effect is to defend the sequence. I exclude any articles taking the line "I support the reptile-mammal sequence but here are some arguments against it", whose net effect is to undermine the position rather than defend it. Indeed, I believe that is their purpose. I doubt that Berlinski has ever been a forthright defender, and suggest that he presents himself as such only to strengthen the effect of his contrary innuendo on the impressionable reader: if even a forthright defender of the sequence is starting to have doubts about it, then the mainstream interpretation must be in trouble! In other words, the claim to be a defender of the sequence is just another rhetorical device.

Related Articles: A Tale of Two Citations

Title Author Date

James Downard Curtis, Tom Sep 08, 2003

David Berlinski writes "... there is no indication in their paper that the calculations had been given to the Proceedings of the Royal Society of London, nor in the ten years since the publication of their paper have Nilsson and Pelger ever made their calculations available in any public format whatsoever. With this in mind I entertained the hypothesis – I did not nothing more – that Nilsson and Pelger had not performed those calculations for the full range of their sequence, but had instead undertaken calculations for a handful of points and completed their sequence by extrapolation. Whether this is what they did, I wrote, was something that I did not know."

He seems to have underestimated the force of his prose. In fact he wrote:
"Moreover, Nilsson and Pelger do not calculate the visual acuity of any structure, and certainly not over the full 1,829 steps of their sequence. They suggest that various calculations have been made, but they do not show how they were made or tell us where they might be found. At the very best, they have made such calculations for a handful of data points, and then joined those points by a continuous curve."

and:
"As I have already observed, the number 363,992 is derived from the number 80,129,540, which is derived from the number 1,829-which in turn is derived from nothing at all."

In fact, Nilsson and Pelger are very clear where the 1,819 steps come from, and very clear about the source of their calculations. Figure 1 (page 54) of their article consists of three graphs. The first shows increase of spatial resolution as the light sensitive patch invaginates (model sequence steps 1 to 3). The text commenting on this graph refers us to Nilsson 1990 for further details.

The second graph shows improvement of detectable resolution as the appature narrows when pit depth equals diameter (model sequence steps 3 to 6). The text cites two papers and the equation used to calculate this (and, I believe, the previous) graph.

The third graph plots increase of optical resolution as the refractive index of the lense increases. Again the text cites the source of the equations used (Fletcher et al. 1954).

So though figure three only plots the optical performance of the first seven model eyes, the performance of the continuos range of intermediate eyes has been previously plotted in figure 1.

Finaly, had Berlinski been in doubt as to the treatment of the subject, or the exact source of the various figures, he need only have written one of the authors and asked. As Berlinski has written to Nilsson on a related matter, such an enquiry would only be polite before making accusations of fraud.

Related Articles: A Tale of Two Citations

Title Author Date

James Downard Curtis, Tom Sep 08, 2003

Further to my previous message, here is the complete list of the 1829 steps as sent to me by Dan -E. Nilsson. I apologise for the loss of formating.

type of change start end 1% steps sum 1% steps
Stage 1 to 2 176
corneal width (curve) 46.5 46.5 0
corneal thickness 3.35 10.1 110
upper retinal surface width 46.6 47.2 1
lower retinal surface width 46.6 51.2 9
upper pigment surface width 46.6 53.9 14
lower pigment surface width 46.6 58.0 21

Stage 2 to 3 362
corneal width (curve) 46.5 46.5 0
corneal thickness 10.1 25.8 94
upper retinal surface width 47.2 75.6 47
lower retinal surface width 51.2 85.9 52
upper pigment surface width 53.9 97.7 59
lower pigment surface width 58.0 108.9 63

Stage 3 to 4 270
corneal width (curve) 46.5 40.2 15
corneal thickness 25.8 37.0 36
upper retinal surface width 75.6 130.6 54
lower retinal surface width 85.9 143.9 51
upper pigment surface width 97.7 145.6 40
lower pigment surface width 108.9 158.7 37

Stage 4 to 5 225
corneal width (curve) 40.2 29.5 31
corneal thickness 37.0 43.7 16
retinal surface width 137.2 202.3 39
pigment surface width 152.1 208.6 31

Stage 5 to 6 192
corneal width (curve) 29.5 37.2 23
corneal thickness 43.7 46.6 6
central refractive index increase from 1.35 0.010 0.060 180

Stage 6 to 7 308
corneal width (curve) 37.2 55.4 40
corneal thickness 46.6 42.3 9
iris width 4.76 11.99 92
retinal surface width 202.3 132.1 42
pigment surface width 208.6 143.7 37
central refractive index increase from 1.35 0.060 0.090 41
lens width 28.4 21.9 26
lens height 24.0 20.2 17

Stage 7 to 8 296
corneal width (curve) 55.4 60 8
corneal thickness 42.3 30.7 32
iris width 11.99 15.88 28
retinal surface width 132.1 72.6 60
pigment surface width 143.7 82.6 55
central refractive index increase from 1.35 0.090 0.180 70
lens width 21.9 16.7 27
lens height 20.2 16.7 19

Grand total 1829

Related Articles: A Tale of Two Citations