The exchange of letters between Ofqual and exam board, Edexcel, revealed by TES today, are a vital missing link for those trying to understand the GCSE English grading controversy.
Ofqual's initial report into the affair, published ten days ago, set out in detail its reasons for concluding that exam boards and their examiners had "acted properly" and that lower than expected June grades "were right".
But as Sir Mike Tomlinson, who led an independent inquiry into a strikingly similar affair a decade ago, pointed out, the report left a major question unanswered: "What advice and guidance did the regulator give the exam boards?"
"I can't believe they gave none," the former Ofsted chief inspector told TES last week. "And to what extent was that advice influential in the grade boundary decisions?"
We now know exactly what Ofqual told one of the exam boards and it was a lot more than mere "advice and guidance".
The regulator may not have actually used its statutory power to formally "direct" a board to move grade boundaries. But its letter to Edexcel on August 7 is quite clear that it would have done so if the board did not toe the line.
The language used in the letter from Ofqual's head of standards, Dennis Opposs, is also enlightening. It will shatter the delusions of anyone who still believes that exam results are based on simple, straightforward, objective decisions about the quality of a pupil's work.
Instead the grading process is presented as something more akin to climbing Everest - a long, hard slog with a pre-determined goal in mind.
"We know that the awards in the new English suite have proved to be challenging," Mr Opposs begins diplomatically. "I am grateful to you and your colleagues for all your efforts, which have resulted in GCSE English language outcomes which are close to predictions."
But it is Edexcel's GCSE in English, not English language, that concerns him and the discussion that follows provides suggests a lack of certainty and, seemingly, even guesswork in Ofqual's regulation.
Mr Opposs was searching for explanations for the gap between the board's proposed results and statistical predictions of where they should be - according to Ofqual's policy of keeping grading in line with previous years.
He suggests that "one possibility" is that Edexcel's previous English qualifications - which the predictions were partly based on - "could have been too lenient". He adds that data for two English qualifications has been reviewed and "it does appear that this may be part of the explanation".
But it takes a reply from Edexcel to point out that the most generous of these qualifications is "not relevant" as it was a pilot qualification, since withdrawn, that was never used for the board's statistical predictions.
Edexcel's insistence, elsewhere in its letter, that its original planned grades were a "fair award", despite the fact that they would have pushed C grades up 8 percentage points, poses a direct challenge to a central Ofqual finding.
The regulator has been quite clear that work submitted in June was "properly graded".
But we now know that at least some of those grades were only set after Ofqual had intervened and ordered a toughening up.
The case put by Edxecel, that much more generous grades were appropriate, is in line with the views and expectations of many schools who argue their staff have years of experience in knowing what a C looks like.
If it turns out that the final June grades were in fact wrong, then Ofqual's finding that January's more lenient grades were "too generous" must also be thrown into doubt.
But if Ofqual's original position on the grades stands, it still has to answer uncomfortable questions about why it failed to intervene on January's grades at the time. Its point that there was not enough data available maybe factually correct, but it does not really explain why the regulator failed to anticipate, spot, and act on an entirely foreseeable problem.
Edexcel is only one of five GCSE boards. It is another - AQA - whose English GCSEs have caused most concern, because they are the qualifications taken by the largest number of schools.
Asked on GCSE results day, last month, whether Ofqual pressure had led to tougher grades, AQA chief executive, Andrew Hall said: "Have I felt leant on by anybody? No, AQA has carried on exactly as it always does."
But there are now likely to be calls for more correspondence between the regulator and exam boards to be released - if it exists - to reveal exactly what was said.
On August 28, Glenys Stacey, Ofqual chief regulator told TES: "We are quite happy to put in the public domain all of the exchanges that we have with exam boards, with government, with anybody. We have nothing to hide here and we wish to be absolutely transparent."
But when this offer was followed up Ofqual only emailed links to two general letters, already published, about this summer's awarding procedures. A press officer later told TES that no further correspondence was going to be released.
The letters being published in full today, by TES, were not made public by Ofqual. Their contents may well explain Ms Stacey's equivocation when asked, in the same interview, whether Ofqual was aware of the specific changes made to GCSE English grade boundaries this summer.
"I won't say we were aware of the specific details of all grade boundary changes," she said. "We were aware that awarding bodies were following processes to make sure that their grade boundaries were right. And we were aware that was particularly challenging for them on the controlled assessment units."
Asked whether that meant she was unaware of the specific English changes, Ms Stacey replied: "I think that is probably not the right question to ask at this stage."
She said her report, then still to be published, would set out "exactly what our role is in all of this".
It was another pledge that, in hindsight and with knowledge of the Edexcel letters, was not met.
Finally, asked if there had been communication between Ofqual and the exam boards about the English GCSE grade boundary changes, Ms Stacey said: "I am not in a position to answer questions of that level of detail."
That communication, that we now know did take place, tells us about more than the detail of the English GCSE. It raises doubts about the whole approach on which Ofqual has based its system of "comparable outcomes" - designed to counter grade inflation and keep standards constant over time.
Edexcel's letter asks Ofqual whether it was "confident" that evidence from key stage 2 SATs - taken by the GCSE cohort when they were 11 - was really "sufficient" for the statistical predictions it was demanding such tight adherence to.
Mr Opposs ignored the question in his reply to the exam board. But it can be argued that under its own rules Ofqual should not have been applying the controversial measures to GCSE English in the first place.
TES revealed on Friday that the new qualification does not fulfil four of the five conditions that Ofqual says must all be met for "comparable outcomes" to be used.
So why did the watchdog insist on exam boards employing it for GCSE English, in such vigorous terms? Ms Stacey has fiercely and consistently denied any suggested of acting under political pressure when it comes to grading issues.
But questions are bound to be raised again about just how much education secretary, Michael Gove, and his department, knew about what was going on.