He found the seal discrepancy on what felt like a Thursday, fourteen days in.
He had spent the morning with a set of the Archive's administrative seals: the physical stamps and their corresponding records, the documentation of when each seal had been in use, which officials had held each seal, and which documents had been stamped with each one. This was the kind of material that was excruciating to review and that almost no one ever reviewed, which was precisely why it was worth reviewing. The most carefully hidden things were usually hidden in the most boring places.
The seal on the chapter of the Verdant Prophecy that dealt with the Black Dragon was dated to a period of use that was three hundred and eight years ago at the earliest.
The chapter it was stamped on was described in the Prophecy's own preamble as part of the original document, nine centuries old.
He had known this in outline for several days. He had noted it as a discrepancy early in his review. What he had not done until now was trace the seal registry to establish the precise dates with documentary certainty.
He found the registry. He found the seal. He found the entry for its period of validity. He compared it to the date on the document.
The seal could not have been on the original document. It had not existed when the original document was supposedly written. It had been created, according to the registry, three centuries after the document was supposedly composed. And yet there it was, on a chapter the Prophecy claimed as original, stamped with an authority that had not existed at the time of the original's claimed composition.
He wrote this down in his most precise handwriting, with all the registry references, with the page numbers for the chapter in question, with the date documentation laid out in parallel columns so that the discrepancy was visually apparent as well as logically established.
Then he went to find Lysera.
She was in the cataloguing room. He set the notes in front of her. She read them. She asked for the registry, which he had brought. She looked at the registry entries. She looked at the document dates. She looked at the parallel columns.
She was quiet for a very long time.
He had learned, over the two weeks they had been working together, the difference between her silences. There was the silence of processing, which she used when she was reading or considering. There was the silence of choosing, which she used when she had processed something and was deciding what to say about it. There was a third silence he had encountered only a few times, which was different from both: the silence of someone who has been presented with confirmation of something they already knew, and the confirmation is worse than the suspicion because confirmation makes it real.
This was the third silence.
He waited.
She said, finally: The seal cannot have been on the original document.
He said: No.
She said: The chapter was added after the seal was created. The seal was created three hundred and eight years ago. The chapter is presented in the Prophecy as part of the original nine-hundred-year-old text.
He said: Yes.
She said: This is the clearest evidence yet.
He said: It is the most documentable evidence. The earlier discrepancies required interpretation. This requires only the ability to read the dates and compare them.
She said: Anyone who looked at both documents could see it.
He said: Yes.
She was quiet for another moment. Then: No one has ever looked at both documents.
He said: Not as far as the records indicate.
She said: Because no one thought to look. Because the seal registry is an administrative record and the Prophecy is a sacred text and they are filed in different sections under different classification systems, and no one who cared about the Prophecy spent time in the seal registry, and no one who spent time in the seal registry thought to apply the dates to the Prophecy.
He said: Administrative segregation as concealment strategy.
She said: Whether intentional or not, yes.
He said: I think intentional.
She looked at him.
He said: The segregation is too clean. The seal registry and the Prophecy copies are filed under incompatible classification systems, which means they do not appear in cross-reference searches. That is not the default state of a well-organized archive. That is the result of a decision about how to organize the archive.
She said: The classification system was last comprehensively revised three hundred and twelve years ago.
He said: Four years after the seal was created.
She said: Yes.
He said: I need the complete list of documents that were reclassified in the comprehensive revision. The full scope of what moved and where it moved.
She said: That will take several days to compile.
He said: I will help.
She looked at him. She said: It will be tedious.
He said: I have been doing tedious for nine years.
She looked at him for a moment with the quality of attention she reserved for things she was filing carefully. Then she pulled out a fresh sheet of paper, and he pulled his chair around to her side of the table, and they began to compile the list together.
They worked until well past the point when the Archive's light had shifted to its nighttime quality. When they finally stopped, the list was three pages long and neither of them had said much for the last two hours except for the specific, precise communications of people doing parallel work on a shared project. The room was quiet in the way it was quiet when it contained two people who had stopped needing to fill the silence because the silence had become a shared thing rather than a gap between them.
He wrote, before he slept: the seal dates the chapter insertion. The reclassification dates the cover strategy. Four years between the insertion and the strategy to prevent the insertion from being discovered. Someone very organized did this. Someone who thought not only about making the change but about making the change undiscoverable. That level of planning does not happen in a moment of crisis. It was a deliberate program. I am looking at the work of a very patient and very organized person who has been managing this secret for three centuries.
He paused. He wrote: I respect the craft. I am still going to undo it.
He spent the next three days tracing the complete scope of the classification revision.
The work was methodical and tedious in the specific way that the most important work often was: not intellectually complex, requiring only the ability to maintain attention on the relevant details without introducing interpretation, but demanding in terms of sustained concentration and the avoidance of the shortcuts that extended periods of attention-demanding work made tempting.
The classification revision of three hundred and twelve years ago had moved, or reclassified, or removed from accessible record, a total of three hundred and forty-seven documents. He established this by cross-referencing the revision log against the catalog entries before and after the revision, a comparison that required Lysera's assistance because the catalog systems before and after the revision used different organizational schemes and the cross-reference required someone who could navigate both fluently.
She could. She navigated them with the ease of someone who had spent sixty years in a building and knew every room.
Of the three hundred and forty-seven documents, two hundred and nineteen had been moved from the public section to the restricted archive. Seventy-three had been reclassified within the restricted archive, moving to higher restriction tiers. Fifty-five had no documented destination: they had been removed from the catalog entirely, without a notation of where they had gone.
He looked at the fifty-five.
He said: Documents don't disappear from an institutional record unless someone decided they should.
She said: No.
He said: The fifty-five are the most sensitive materials. The ones that couldn't be managed by reclassification because their existence anywhere in the institutional record was a risk.
She said: That is consistent with what I have been able to reconstruct about the content of the removed materials, based on cross-references in other documents.
He said: What did they contain?
She said: Based on the cross-references, I believe they included the original diplomatic correspondence between Elyndor and the Drakonir prior to the modification period. The records of the joint ceremonies that had maintained the Veylwood binding before the modification. And, she paused, documents describing the Drakonir as they actually were: as a people, a culture, a political entity with their own history and their own understanding of the covenant they had signed.
He said: All of that was removed.
She said: All of it.
He said: So that there would be no comparison point.
She said: Yes. If the history of the relationship prior to the modification was gone, the only history available was the modified version. And the modified version described the Drakonir as they were presented in the modified Prophecy: as a threat, a creature, an enemy to be opposed.
He wrote: the removal of the pre-modification documentation about the Drakonir was the most important element of the concealment strategy. More important than modifying the Prophecy itself. The modification could be discovered, as we are discovering it, through internal textual analysis. The removal of the comparison materials was what was supposed to make the internal textual analysis insufficient to complete the picture.
It was not sufficient, because of Miron. Miron had found the comparison in the linguistic record, which was not part of the administrative archive and was therefore not subject to the same control.
He wrote, separately, in smaller letters: I should make sure Miron is careful. If Caerion understands what we are assembling, he will recognize that the linguistic evidence is the gap in his concealment strategy. And Miron is the most exposed of the three of us. He has no institutional position, no formal role in the verification project, no standing that would make removing him complicated.
He closed the notebook. He went to find Miron. He found him in the courtyard with his fruit wine and his linguistic notes, and he sat down across from him and said: I need you to be careful about what you tell people and who you tell it to.
Miron looked at him over his cup. He said: I have been careful for eleven years.
Renji said: I know. I am asking you to continue being careful for a few more weeks.
Miron said: Until we have the document.
Renji said: Until we have the document and we have presented it in a context where removing us is more complicated than acknowledging what it says.
Miron said: And when will that be?
Renji said: I am working on it.
Miron said, in the tone of a man who has been working toward something for eleven years and has learned to be patient: Good. I will wait.
He thought, that evening, about what three hundred and twelve years of a cover strategy meant for the person who had been maintaining it.
Three hundred and twelve years was the length of Caerion's involvement in this specific concealment. He had been High Oracle for two hundred and forty years before the modification, and had overseen the modification sixty years after taking the position. He had then managed the concealment for three hundred and twelve years after that.
Three hundred and twelve years of ensuring that certain documents were not cross-referenced. Of ensuring that the classification system remained structured in a way that made the cross-reference impossible without someone who knew what to look for. Of monitoring who was in the Archive, what they were looking at, what conclusions they were drawing, and whether those conclusions were approaching the dangerous ones.
He had removed Aelindra Vey when her conclusions approached the dangerous ones. He had managed Lysera for sixty years because her conclusions had approached but not reached them, and she had never quite provided a sufficient justification for removal. He had managed the eleven years of Miron's research by ignoring it: the council records office had filed Miron's papers and not distributed them, which was a different kind of management but equally effective.
He had been very good at this for three hundred and twelve years.
He was continuing to be good at it, in the case of Renji. The meeting. The redirection toward the ceremony. The reminder to Lysera about her regular duties. All of it was skilled management.
He wrote: the skill of the management is not evidence of the rightness of what is being managed. Skilled management of a concealment is still management of a concealment. Caerion is skilled. He is also wrong. Both things are true.
He wrote: in nine years I have found that the most effective concealments are usually run by the most competent people. Incompetent people make mistakes. Competent people make almost no mistakes. But they cannot make zero mistakes. They always leave something. I have always found the something. I will find it here.
He returned to the classification revision records one more time, on the afternoon of the second day of working through them, because something in the pattern had been bothering him and he had not yet identified what.
He read through the full list again. He looked at the dates. He looked at the document categories. He looked at the destination notations.
Then he saw it.
Twenty-three of the three hundred and forty-seven documents had been reclassified not to the restricted archive but to a third location: the Oracle's personal archive, a collection held not in the Archive building at all but in the Oracle's residence. This was a classification tier that did not appear in the public catalog at all, because the public catalog only covered the Archive building. He had only found it because the revision record listed destinations, and the revision record was one of the administrative documents he had reviewed in his first week.
Twenty-three documents, held in a private archive that was not subject to any of the oversight procedures that governed the main Archive. No catalog. No access log. No external review.
He wrote: the twenty-three documents in the Oracle's personal archive are not accessible through any institutional mechanism I currently have. They are not accessible to Lysera in her role as Archive Keeper. They are not accessible to Miron. They are not accessible to me.
They are accessible to Caerion.
He wrote: the most sensitive materials were not placed in the restricted archive but in a collection entirely outside the institutional review structure. This is the most complete concealment of the three categories: the restricted archive can at least be requested through official channels, even if the requests are denied. The Oracle's personal archive cannot be requested. It simply does not exist in any accessible documentation.
He wrote: I need to determine what those twenty-three documents contain. I cannot determine it through the institutional channels I have. I need to think about what other options exist.
He spent the afternoon of the third day in the classification revision records working through the content descriptions for the reclassified materials.
Most of the descriptions were terse: administrative record, third tier; correspondence file, sealed; ceremonial procedure document, restricted. The descriptions were designed to indicate the category without describing the content in enough detail to allow a reader to know what specifically was being restricted.
But occasionally a description was slightly less terse than the others. A description that included a date range, or a reference to a specific person, or a notation about the document's relationship to another document. He collected these. Twelve such descriptions, out of the three hundred and forty-seven.
He cross-referenced the twelve with the dates and categories he had already established.
Four of the twelve were from the original covenant period: documents dated to within thirty years of the covenant's initial establishment, before the modification, dealing with administrative topics that should have been routine.
Four were from the modification period itself: correspondence from within the three years during which the modification appeared to have been carried out.
Four were from the post-modification concealment period: administrative records dealing with the reclassification process itself, including two that appeared to be the instructions for how the reclassification was to be organized.
The instructions for the reclassification. He wrote this carefully. The person who had organized the three-hundred-and-twelve-year concealment had left a record of how they organized it. That record was in the restricted archive. He could not access it through the channels available to him.
But it existed. And it would, if he could access it, establish not just what had been done but the specific thought process behind the doing. The concealment plan itself, written down, by the person who had designed it.
He needed that document.
He wrote, at the end of the third day in the classification records: the concealment was designed by someone who understood both the institutional structure they were working within and the nature of the evidence they needed to hide. The design is sophisticated. It is also now fully mapped. Every element of the concealment program has been documented. The three hundred and forty-seven reclassified documents, their destinations, the dates of the reclassification, the instructions for the reclassification process. I do not have access to all of it. I have documented all of it that is accessible and I know what is not accessible and where it is. When the time comes to present this, I will be presenting a complete picture of a deliberate program. Not a collection of anomalies. A program. That is different. That is what turns strong circumstantial evidence into something that cannot be explained away. He was nearly done. He had one more category to cross-reference and then the classification revision picture would be complete. He would finish it in the morning. He closed the file. He turned off the lamp over his section of the reading room. The Archive settled around him into its nighttime quiet, patient and old and full of things it had been waiting to say.
You've reached the end of this chapter
He found the seal discrepancy on what felt like a Thursday, fourteen days in.
He had spent the morning with a set of the Archive's administrative seals: the physical stamps and their corresponding records, the documentation of when each seal had been in use, which officials had held each seal, and which documents had been stamped with each one. This was the kind of material that was excruciating to review and that almost no one ever reviewed, which was precisely why it was worth reviewing. The most carefully hidden things were usually hidden in the most boring places.
The seal on the chapter of the Verdant Prophecy that dealt with the Black Dragon was dated to a period of use that was three hundred and eight years ago at the earliest.
The chapter it was stamped on was described in the Prophecy's own preamble as part of the original document, nine centuries old.
He had known this in outline for several days. He had noted it as a discrepancy early in his review. What he had not done until now was trace the seal registry to establish the precise dates with documentary certainty.
He found the registry. He found the seal. He found the entry for its period of validity. He compared it to the date on the document.
The seal could not have been on the original document. It had not existed when the original document was supposedly written. It had been created, according to the registry, three centuries after the document was supposedly composed. And yet there it was, on a chapter the Prophecy claimed as original, stamped with an authority that had not existed at the time of the original's claimed composition.
He wrote this down in his most precise handwriting, with all the registry references, with the page numbers for the chapter in question, with the date documentation laid out in parallel columns so that the discrepancy was visually apparent as well as logically established.
Then he went to find Lysera.
She was in the cataloguing room. He set the notes in front of her. She read them. She asked for the registry, which he had brought. She looked at the registry entries. She looked at the document dates. She looked at the parallel columns.
She was quiet for a very long time.
He had learned, over the two weeks they had been working together, the difference between her silences. There was the silence of processing, which she used when she was reading or considering. There was the silence of choosing, which she used when she had processed something and was deciding what to say about it. There was a third silence he had encountered only a few times, which was different from both: the silence of someone who has been presented with confirmation of something they already knew, and the confirmation is worse than the suspicion because confirmation makes it real.
This was the third silence.
He waited.
She said, finally: The seal cannot have been on the original document.
He said: No.
She said: The chapter was added after the seal was created. The seal was created three hundred and eight years ago. The chapter is presented in the Prophecy as part of the original nine-hundred-year-old text.
He said: Yes.
She said: This is the clearest evidence yet.
He said: It is the most documentable evidence. The earlier discrepancies required interpretation. This requires only the ability to read the dates and compare them.
She said: Anyone who looked at both documents could see it.
He said: Yes.
She was quiet for another moment. Then: No one has ever looked at both documents.
He said: Not as far as the records indicate.
She said: Because no one thought to look. Because the seal registry is an administrative record and the Prophecy is a sacred text and they are filed in different sections under different classification systems, and no one who cared about the Prophecy spent time in the seal registry, and no one who spent time in the seal registry thought to apply the dates to the Prophecy.
He said: Administrative segregation as concealment strategy.
She said: Whether intentional or not, yes.
He said: I think intentional.
She looked at him.
He said: The segregation is too clean. The seal registry and the Prophecy copies are filed under incompatible classification systems, which means they do not appear in cross-reference searches. That is not the default state of a well-organized archive. That is the result of a decision about how to organize the archive.
She said: The classification system was last comprehensively revised three hundred and twelve years ago.
He said: Four years after the seal was created.
She said: Yes.
He said: I need the complete list of documents that were reclassified in the comprehensive revision. The full scope of what moved and where it moved.
She said: That will take several days to compile.
He said: I will help.
She looked at him. She said: It will be tedious.
He said: I have been doing tedious for nine years.
She looked at him for a moment with the quality of attention she reserved for things she was filing carefully. Then she pulled out a fresh sheet of paper, and he pulled his chair around to her side of the table, and they began to compile the list together.
They worked until well past the point when the Archive's light had shifted to its nighttime quality. When they finally stopped, the list was three pages long and neither of them had said much for the last two hours except for the specific, precise communications of people doing parallel work on a shared project. The room was quiet in the way it was quiet when it contained two people who had stopped needing to fill the silence because the silence had become a shared thing rather than a gap between them.
He wrote, before he slept: the seal dates the chapter insertion. The reclassification dates the cover strategy. Four years between the insertion and the strategy to prevent the insertion from being discovered. Someone very organized did this. Someone who thought not only about making the change but about making the change undiscoverable. That level of planning does not happen in a moment of crisis. It was a deliberate program. I am looking at the work of a very patient and very organized person who has been managing this secret for three centuries.
He paused. He wrote: I respect the craft. I am still going to undo it.
He spent the next three days tracing the complete scope of the classification revision.
The work was methodical and tedious in the specific way that the most important work often was: not intellectually complex, requiring only the ability to maintain attention on the relevant details without introducing interpretation, but demanding in terms of sustained concentration and the avoidance of the shortcuts that extended periods of attention-demanding work made tempting.
The classification revision of three hundred and twelve years ago had moved, or reclassified, or removed from accessible record, a total of three hundred and forty-seven documents. He established this by cross-referencing the revision log against the catalog entries before and after the revision, a comparison that required Lysera's assistance because the catalog systems before and after the revision used different organizational schemes and the cross-reference required someone who could navigate both fluently.
She could. She navigated them with the ease of someone who had spent sixty years in a building and knew every room.
Of the three hundred and forty-seven documents, two hundred and nineteen had been moved from the public section to the restricted archive. Seventy-three had been reclassified within the restricted archive, moving to higher restriction tiers. Fifty-five had no documented destination: they had been removed from the catalog entirely, without a notation of where they had gone.
He looked at the fifty-five.
He said: Documents don't disappear from an institutional record unless someone decided they should.
She said: No.
He said: The fifty-five are the most sensitive materials. The ones that couldn't be managed by reclassification because their existence anywhere in the institutional record was a risk.
She said: That is consistent with what I have been able to reconstruct about the content of the removed materials, based on cross-references in other documents.
He said: What did they contain?
She said: Based on the cross-references, I believe they included the original diplomatic correspondence between Elyndor and the Drakonir prior to the modification period. The records of the joint ceremonies that had maintained the Veylwood binding before the modification. And, she paused, documents describing the Drakonir as they actually were: as a people, a culture, a political entity with their own history and their own understanding of the covenant they had signed.
He said: All of that was removed.
She said: All of it.
He said: So that there would be no comparison point.
She said: Yes. If the history of the relationship prior to the modification was gone, the only history available was the modified version. And the modified version described the Drakonir as they were presented in the modified Prophecy: as a threat, a creature, an enemy to be opposed.
He wrote: the removal of the pre-modification documentation about the Drakonir was the most important element of the concealment strategy. More important than modifying the Prophecy itself. The modification could be discovered, as we are discovering it, through internal textual analysis. The removal of the comparison materials was what was supposed to make the internal textual analysis insufficient to complete the picture.
It was not sufficient, because of Miron. Miron had found the comparison in the linguistic record, which was not part of the administrative archive and was therefore not subject to the same control.
He wrote, separately, in smaller letters: I should make sure Miron is careful. If Caerion understands what we are assembling, he will recognize that the linguistic evidence is the gap in his concealment strategy. And Miron is the most exposed of the three of us. He has no institutional position, no formal role in the verification project, no standing that would make removing him complicated.
He closed the notebook. He went to find Miron. He found him in the courtyard with his fruit wine and his linguistic notes, and he sat down across from him and said: I need you to be careful about what you tell people and who you tell it to.
Miron looked at him over his cup. He said: I have been careful for eleven years.
Renji said: I know. I am asking you to continue being careful for a few more weeks.
Miron said: Until we have the document.
Renji said: Until we have the document and we have presented it in a context where removing us is more complicated than acknowledging what it says.
Miron said: And when will that be?
Renji said: I am working on it.
Miron said, in the tone of a man who has been working toward something for eleven years and has learned to be patient: Good. I will wait.
He thought, that evening, about what three hundred and twelve years of a cover strategy meant for the person who had been maintaining it.
Three hundred and twelve years was the length of Caerion's involvement in this specific concealment. He had been High Oracle for two hundred and forty years before the modification, and had overseen the modification sixty years after taking the position. He had then managed the concealment for three hundred and twelve years after that.
Three hundred and twelve years of ensuring that certain documents were not cross-referenced. Of ensuring that the classification system remained structured in a way that made the cross-reference impossible without someone who knew what to look for. Of monitoring who was in the Archive, what they were looking at, what conclusions they were drawing, and whether those conclusions were approaching the dangerous ones.
He had removed Aelindra Vey when her conclusions approached the dangerous ones. He had managed Lysera for sixty years because her conclusions had approached but not reached them, and she had never quite provided a sufficient justification for removal. He had managed the eleven years of Miron's research by ignoring it: the council records office had filed Miron's papers and not distributed them, which was a different kind of management but equally effective.
He had been very good at this for three hundred and twelve years.
He was continuing to be good at it, in the case of Renji. The meeting. The redirection toward the ceremony. The reminder to Lysera about her regular duties. All of it was skilled management.
He wrote: the skill of the management is not evidence of the rightness of what is being managed. Skilled management of a concealment is still management of a concealment. Caerion is skilled. He is also wrong. Both things are true.
He wrote: in nine years I have found that the most effective concealments are usually run by the most competent people. Incompetent people make mistakes. Competent people make almost no mistakes. But they cannot make zero mistakes. They always leave something. I have always found the something. I will find it here.
He returned to the classification revision records one more time, on the afternoon of the second day of working through them, because something in the pattern had been bothering him and he had not yet identified what.
He read through the full list again. He looked at the dates. He looked at the document categories. He looked at the destination notations.
Then he saw it.
Twenty-three of the three hundred and forty-seven documents had been reclassified not to the restricted archive but to a third location: the Oracle's personal archive, a collection held not in the Archive building at all but in the Oracle's residence. This was a classification tier that did not appear in the public catalog at all, because the public catalog only covered the Archive building. He had only found it because the revision record listed destinations, and the revision record was one of the administrative documents he had reviewed in his first week.
Twenty-three documents, held in a private archive that was not subject to any of the oversight procedures that governed the main Archive. No catalog. No access log. No external review.
He wrote: the twenty-three documents in the Oracle's personal archive are not accessible through any institutional mechanism I currently have. They are not accessible to Lysera in her role as Archive Keeper. They are not accessible to Miron. They are not accessible to me.
They are accessible to Caerion.
He wrote: the most sensitive materials were not placed in the restricted archive but in a collection entirely outside the institutional review structure. This is the most complete concealment of the three categories: the restricted archive can at least be requested through official channels, even if the requests are denied. The Oracle's personal archive cannot be requested. It simply does not exist in any accessible documentation.
He wrote: I need to determine what those twenty-three documents contain. I cannot determine it through the institutional channels I have. I need to think about what other options exist.
He spent the afternoon of the third day in the classification revision records working through the content descriptions for the reclassified materials.
Most of the descriptions were terse: administrative record, third tier; correspondence file, sealed; ceremonial procedure document, restricted. The descriptions were designed to indicate the category without describing the content in enough detail to allow a reader to know what specifically was being restricted.
But occasionally a description was slightly less terse than the others. A description that included a date range, or a reference to a specific person, or a notation about the document's relationship to another document. He collected these. Twelve such descriptions, out of the three hundred and forty-seven.
He cross-referenced the twelve with the dates and categories he had already established.
Four of the twelve were from the original covenant period: documents dated to within thirty years of the covenant's initial establishment, before the modification, dealing with administrative topics that should have been routine.
Four were from the modification period itself: correspondence from within the three years during which the modification appeared to have been carried out.
Four were from the post-modification concealment period: administrative records dealing with the reclassification process itself, including two that appeared to be the instructions for how the reclassification was to be organized.
The instructions for the reclassification. He wrote this carefully. The person who had organized the three-hundred-and-twelve-year concealment had left a record of how they organized it. That record was in the restricted archive. He could not access it through the channels available to him.
But it existed. And it would, if he could access it, establish not just what had been done but the specific thought process behind the doing. The concealment plan itself, written down, by the person who had designed it.
He needed that document.
He wrote, at the end of the third day in the classification records: the concealment was designed by someone who understood both the institutional structure they were working within and the nature of the evidence they needed to hide. The design is sophisticated. It is also now fully mapped. Every element of the concealment program has been documented. The three hundred and forty-seven reclassified documents, their destinations, the dates of the reclassification, the instructions for the reclassification process. I do not have access to all of it. I have documented all of it that is accessible and I know what is not accessible and where it is. When the time comes to present this, I will be presenting a complete picture of a deliberate program. Not a collection of anomalies. A program. That is different. That is what turns strong circumstantial evidence into something that cannot be explained away. He was nearly done. He had one more category to cross-reference and then the classification revision picture would be complete. He would finish it in the morning. He closed the file. He turned off the lamp over his section of the reading room. The Archive settled around him into its nighttime quiet, patient and old and full of things it had been waiting to say.
You've reached the end of this chapter