Letter from W. Rowat, DFO DM April 7, 1997, to provincial counterparts. Links added, to notes below. This letter was mentioned in the Press, but I was never (officially) copied on it nor do I have the exact list of recipients.

KNIB annotations:

Firstly note what is NOT in Rowat's letter: any denial of discrete populations (the word 'populations' is there, but conveys little as placed and it's probably the residue of an incomplete edit). The 'population' question was probably the big item that worried DFO strategically. They had stuck their own foot in it by demanding a by-population (using management areas as a proxy) treatment, and this was indeed consistent with the developing science and understanding at the time. Actually the science was very strong. But DFO tactically wanted to get away from a by-areas designation because [1] you lost the averaging that could make the worst areas seem not so bad, and [2] you were much less likely to get the E-word on a single-unit basis (one chance that was less likely through the averaging) than you were on a by-areas basis (ten chances, many more likely to get "Endangered" because there was no averaging with neighbouring populations that might be in better shape. Through the months surrounding this, DFO refused to discuss the population question with the press.

"postal ballot" was a process by which members would, prior to the meeting, record a straw vote on status. They could change their vote at the meeting, but having the postal ballot on record could reduce their wiggle room in two respects. [1] If they found the Report convincing (which Cosewic's own chosen reviewers did) and voted some populations Endangered, that would be a position DFO would have to negotiate them back from; holding off an 'endangered' designation would be much easier to do so if the postal ballot did not occur. [2] If they voted at all, in terms of the by-areas designations recommended in the Report (the same by-area basis Doubleday had demanded), it would solidify the by-areas treatment and that had the potential if not certainty of the E word appearing for at least a few populations.

the 'criteria' complaint was patently thin; the criteria Rowat objected to were substantially those already shortlisted by Environment Canada, and the same (roughly equivalent to IUCN criteria) were later, following a process that did include DFO personnel, adopted by Cosewic. Cosewic at the time did not have explicit quantitative criteria -- the 1997/8 Cod Status Report seems to have been the first application of quantitative criteria at Cosewic. Similar criteria were subsequently adopted by Cosewic, but by then it was too late, DFO had won the first skirmish.

"decline of Atlantic cod stocks has been arrested": not true. No evidence presented, similar claim made often without evidence, hindsight (2004) shows it didn't happen. It was a characteristic of the upper levels of DFO during this process to be very free with conclusions and also accusations (ad hominem attacks), very sparse with substantiation -- many instances are documented (rebutted) in the Appendix to the 1998 Report.

"full debate" to DFO meant (1997) controlled. Real debate in COSEWIC doesn't really happen. There is no forum for assessing the assertions of a Jurisdiction. This leaves the Jurisdiction free to bandy about all sorts of assertions and accusations, and even to change their mind on them (180°!) without Cosewic requiring an explanation.
   What was DFO's idea of "full debate" in 1997?   Sending someone who would cite a recent workshop proceedings to the effect that all the Report's invokation of populations was out of date (he didn't exactly say 'wrong', but that was the natural conclusion listeners would hear -- you should see the transcript), but oh my golly he didn't have a copy of the proceedings with him; it took repeated requests to get a copy, and when it came it did not gainsay, but in fact strengthened, the Report's acknowledgment of populations. That speaker is clearly the kind of guy who'll tell you the time, but if it matters you'd best ask him what time zone on the planet he's talking about. Those misrepresentations of its own documents helped Cosewic to vote for a deferral; they cost Cod one more year of possible protection and restoration efforts, that would have been required by an 'Endangered' designation.
      What did "full debate" mean 1998?   Having the actual Chair, Erich Haber, mentioned in Rowat's letter, step aside for the Cod discussions, so, obviously, they could be chaired by someone more acceptable to DFO (described (surprisingly) in Cosewic official minutes of AGM 1998). The replacement chair told the committee to ignore satellite DNA evidence, other evidence of populations, and also ignore DFO's own documentation that supported multiple populations, AND all that despite COSEWIC's own precautionary rule that said that if there was even a reasonable possibility of populations, even if not fully proven, designation should proceed on that precautionary basis because the error in that direction resulted in only over-protection, but the error the other way could result in serious under-protection (see Cosewic's published procedures 1992-1997, and 1998) that could lead to biodiversity loss.

----- extra details on KNIB comments -----

"demanding": Signed by W. Doubleday, DFO's comments that were due in the fall of 1996 were not sent until officially late. They included an insistence that designation should be on the basis of management areas. As well and presumably strategically, they were date-stamped Dec. 23, the day after the recipient at Cosewic departed for an extended break, until the end of January. According to the rules, late comments have no force. But the cooperation of Cosewic in bending this rule as a favour to the Jurisidiction allowed these late comments to commence a train of events to engineer a delay that would furnish an excuse for Cosewic to defer the matter for another entire year. Canadian Geographic (July/August 1997, p.23) reported " 'I think there was a plan to make it late,' says a COSEWIC member who asked not to be identified."

I remarked: '180°!' change of mind -- DFO (W. Doubleday) argued forcefully in 1996 for treatment of Cod as multiple populations, and even pointed out that this was certainly within Cosewic's rules. The same issue had concerned me, andI had queried Cosewic and their reply had been to treat Cod as one unit; but the emerging evidence from the mid-1990s was becoming more and more persuasive, so in fact I agreed with DFO on this point, and therefore I made that change. But DFO abruptly (1997 meeting, see above) changed its mind. Cosewic accepted that change without a blink, and never required DFO to explain.