THEY INTERPRETED THE MOTIVE BEFORE THEY READ THE RECORD
The Royal Frankish Annals were close to the court. That does not give anyone permission to replace their words with ours.
There is a strange methodological inversion hiding inside modern readings of the Royal Frankish Annals.
It sounds sophisticated.
It sounds cautious.
It sounds like source criticism.
The annalists were connected to the Carolingian court.
Therefore their writing must be understood politically.
They participated in Carolingian legitimacy.
They selected events.
They shaped memory.
They represented power.
They wrote from inside an interconnected elite world.
All of that is worth studying.
Modern scholarship has good reason to investigate it.
The Royal Frankish Annals emerged in close proximity to Carolingian political power, and historians have shown that their political language repeatedly emphasizes not only the king but the Franks, assemblies, consent, collective action, and the relationship between ruler and political community. One major modern treatment describes this as a Carolingian “historiography of consensus.”
Fine.
Now read the record.
Because none of those observations settles a single verb.
None tells us what pugna says in a particular sentence.
None tells us what fides establishes in a particular relationship.
None tells us what an obses is doing in a particular transaction.
None establishes killing where killing was not reported.
None establishes rebellion where rebellion was not named or otherwise established.
None establishes sovereignty before the chronology establishes the relationship.
And none gives a modern historian authority to replace an eighth-century statement with what the historian thinks the eighth-century writer must secretly have meant.
Yet watch how easily the order gets reversed.
The historian begins asking:
WHY DID THE ANNALIST WRITE THIS?
before settling the logically prior question:
WHAT DID THE ANNALIST ACTUALLY WRITE?
That order is backwards.
And once the order is reversed, source criticism can become permission for interpolation.
The author’s presumed political motive begins supplying things the sentence never supplied.
A contest becomes a conventional pitched battle.
A breach becomes rebellion.
A guarantee becomes kidnapping.
Movement becomes invasion.
A relationship being established becomes sovereignty already possessed.
A terse chronological statement becomes a cinematic scene.
And then the resulting reconstruction is defended by pointing back to the writer’s proximity to the court.
That does not solve the source.
It bypasses it.
The proper order is brutally simple:
SOURCE.
WORD.
SYNTAX.
SEQUENCE.
INTERNAL COMPARISON.
EXTERNAL EVIDENCE.
CONTEXT.
ONLY THEN MOTIVE.
Motive is downstream of evidence.
It cannot be permitted upstream.
And if we are actually serious about historical method, there is an even more fundamental law beneath that one:
THE SOURCE RETAINS JURISDICTION OVER ITS OWN TESTIMONY
If you use a historical source, begin with what that source actually says.
If you do not accept the source, say so.
Challenge it.
Reject it.
Contradict it with another witness.
Contradict it with archaeology.
Show that its chronology fails.
Show that an independent source gives an incompatible account.
Show precisely where its statement becomes impossible.
But do not perform the fraudulent middle operation:
cite the source, alter its testimony, and then keep the citation attached to your replacement.
That is not source criticism.
That is borrowing authority from an object after removing the thing the object actually said.
The distinction is fundamental.
A source is evidence.
An interpretation is a claim about evidence.
Those are different objects.
If an annalist writes A, then whatever else may eventually be concluded about the historical event, one fact is established directly:
THE ANNALIST WROTE A.
You may later conclude that A was incomplete.
You may prove A was mistaken.
You may demonstrate that A contradicts another source.
You may show that A represented a political interest.
You may discover evidence the annalist did not possess.
But none of those operations changes the testimony itself.
A remains A.
You do not acquire authority to turn it into B because you believe you understand the writer better than the writer’s surviving words.
This matters because historical writing often hides that substitution behind an apparently cautious sentence:
“We cannot simply trust the source.”
Trust what?
That the manuscript contains these words?
That the writer reports an event?
That his causal explanation is correct?
That a number is accurate?
That an attributed motive is accurate?
That reported speech is verbatim?
These are separate claims.
Test them separately.
Do not dissolve all of them into a vague atmosphere called “bias.”
Bias is not evidence for the historian’s preferred alternative.
Suspicion is not testimony.
Silence is not permission.
And interpretation does not become a witness merely because the witness has been declared unreliable.
If the source reports killing, then the source reports killing.
Whether that killing happened exactly as described is another historical question.
If the source does not report killing in some other passage, the historian cannot manufacture killing in that passage merely by saying:
“Well, the source was biased.”
Bias does not magically contain the missing verb.
If you think killing nevertheless occurred, bring the independent evidence.
That is history.
Put the objects beside one another.
Here is source A.
Here is source B.
Here is archaeology.
Here is the chronology.
Here is where they agree.
Here is where they conflict.
Here is what remains uncertain.
Reason from the evidence.
But never replace testimony with interpretation and continue citing the testimony as though nothing happened.
THE ORIGINAL RECORD ALREADY KNOWS HOW TO ACCUSE
This becomes devastatingly clear when we stop speaking abstractly and actually read the Royal Frankish Annals.
Look at 782.
The original text does not require a historian to invent the political category of rebellion.
It gives it.
The Latin says:
Saxones solito more rebellati sunt.
The Saxons rebelled again in their accustomed manner.
Then the entry continues.
The Frankish force encounters the Saxons.
The record does not merely say pugna and leave death hidden inside the word.
It explicitly adds:
multos Saxones interimentes
—many Saxons being killed.
Then it names two Frankish royal envoys who fell.
Later Charles reaches the Aller and Weser. The Saxons assemble, submit themselves under the king’s authority, and hand over those identified by the record as principally responsible for the rebellion for killing. The entry then says that this was done. The original Latin carries all of those distinctions.
Stop.
Read what the object just demonstrated.
The writer knows how to say:
REBELLION.
The writer knows how to say:
KILLING.
The writer knows how to report Frankish deaths.
The writer knows how to describe submission under authority.
The writer knows how to identify responsibility.
The writer knows how to record an order resulting in death.
There is no lexical helplessness here.
No mysterious inability to say that people died.
No hidden convention requiring modern historians to imagine death inside every occurrence of pugna.
The annalist has words.
He uses them.
And that means the differences between his words matter.
The conclusion is not:
Nobody could possibly have died wherever the annalist merely writes pugna.
That would outrun the evidence in the opposite direction.
The conclusion is narrower and stronger:
PUGNA DOES NOT ITSELF REPORT DEATH.
We know this from the writer’s own usage.
Because when the record wants death carried in the sentence, it can add death.
The difference appears again in 783.
The source reports a pugna.
Then separately tells us that a very great multitude of Saxons fell, with few escaping.
A second pugna follows.
Again the source separately says that no smaller number of Saxons fell.
That is an internal lexical control.
You do not need a modern theory.
You do not need to crawl into the author’s psychology.
You do not need a painting.
You do not need to say:
“Well, obviously pugna meant a bloodbath because that’s what medieval warfare looked like.”
The record itself gives you the distinction.
pugna
and
people falling or being killed
can occur together.
They are not therefore identical propositions.
One word cannot simply be forced to carry another word’s evidentiary burden.
This is what reading means.
WHEN THE RECORD MEANS REBELLION, IT CAN SAY REBELLION
The same control destroys another lazy substitution.
Not every contradiction is automatically rebellion.
Not every contest is automatically rebellion.
Not every breach is automatically rebellion.
Because the annalist possesses explicit rebellion vocabulary.
Again, 782 gives us rebellati sunt.
783 again describes the Saxons as rebellious.
784 again reports rebellion.
The record is perfectly capable of placing that political characterization into the story.
Good.
Preserve it.
Do not erase rebellion where the source says rebellion merely because it complicates a preferred interpretation.
But apply the same standard in both directions.
Do not add rebellion where the source does not say rebellion merely because rebellion fits the modern summary.
That distinction matters because rebellion is not simply “fighting.”
Rebellion contains standing.
For rebellion to exist as a political category, there must be some antecedent authority, obligation, fidelity, settlement, or recognized relationship against which rebellion becomes intelligible.
Therefore chronology matters.
When does rebellion language appear?
What happened before it?
What oath was sworn?
What agreement existed?
What standing had been acknowledged?
Who had entered whose authority?
What was subsequently violated?
Those are not trivial semantic questions.
They determine what kind of political relationship the source is actually describing.
Suppose instead we begin with the modern assumption:
CHARLEMAGNE ALREADY SOVEREIGNLY RULED THE SAXONS.
Then every later Saxon contradiction becomes:
REBELLION.
And afterward the historian points to all these “rebellions” as proof that Charlemagne already sovereignly ruled the Saxons.
Do you see the circle?
Assume the authority.
Use the assumed authority to name opposition rebellion.
Then use the word rebellion as evidence for the authority originally assumed.
That is not proof.
The conclusion entered through the front door wearing the disguise of a premise and emerged through the back door pretending to be evidence.
The cure is chronology.
Let the relationship establish itself.
Do not read the later standing backward.
LOOK AT TASSILO
The original annalistic material becomes even more revealing when it wants to construct an accusation against someone whose prior obligations matter.
Tassilo is not presented merely as:
“a ruler Charles disliked.”
The record builds the case procedurally.
In 781 the text recalls prior oaths.
Tassilo renews his oaths.
He gives twelve obsides.
The record then states that he did not long preserve the promises he had made.
Then the relationship develops.
By 787 the record is explicit about the dispute: obedience, fidelity, sworn obligation, the attempted establishment of peace, Tassilo’s refusal, Frankish claims of justice, and Tassilo eventually placing himself in Charles’ hands and renewing oaths.
Then 788.
Now the accusation is almost juridical in its density.
Tassilo’s fidelity is challenged.
Fraud is alleged.
The record says he confessed to communicating with the Avars.
It reports accusations concerning plots against royal vassals.
It says he instructed men to swear deceptively.
It recalls prior conduct.
An assembly judges him.
A capital judgment is demanded.
And then Charles restrains that judgment.
Why?
The record itself tells us:
moved by mercy, for the love of God, and because Tassilo was his kinsman.
The theological explanation is inside the source. The accusation is inside the source. The fidelity is inside the source. The oath is inside the source. The judgment is inside the source.
We do not need to invent them.
That is what explicit moral-political narration looks like.
And this should permanently end the claim that because these writers operated near royal power, we should assume their vocabulary is merely an opaque shell concealing the reality that modern historians must reconstruct for them.
No.
These people could accuse.
They could praise.
They could say fidelity.
They could say infidelity.
They could invoke justice.
They could invoke mercy.
They could invoke God.
They could report killing.
They could report rebellion.
They could distinguish all of these categories because their political language depended upon distinctions.
Preserve them.
THE COURT WAS NOT GOD
Now we reach the deeper problem with reducing all of this to “court propaganda.”
Yes, the annalistic project participated in Carolingian political representation.
Nobody needs to deny that.
Modern scholarship has demonstrated the affirmative political language of the text and its emphasis on Carolingian rule, the Franks, and consensus.
But the writers did not inhabit a modern secular universe in which the highest imaginable authority was:
the man currently occupying the throne.
Their own text refuses that reduction.
God appears as an actor in history.
Victory is repeatedly attributed to divine aid.
Mercy is explained through love of God.
Justice is not merely royal preference.
Oath, fidelity, sin, peace, judgment, mercy and authority occupy a moral-theological field.
Recent scholarship examining the original redaction of the Royal Frankish Annals explicitly argues that divine agency is central to its historical narrative.
That matters.
The conceptual hierarchy represented by these writers cannot be flattened into:
KING
↓
COURT
↓
WRITER
The king himself exists within an order.
God.
Justice.
Oath.
Fidelity.
Obligation.
Peace.
Then political actors.
The king is powerful.
The king may be favored.
The writer may celebrate him.
But the king is not the ontological source of truth simply because he is king.
The text itself places royal conduct within a theological and moral frame larger than the person of the ruler.
So when a modern reader says:
“Well, this was written near the court.”
Yes.
And?
Now read the sentence.
Court proximity is context.
It is not semantics.
Political proximity does not repeal grammar.
Institutional position does not make verbs interchangeable.
Ideology does not insert deaths into sentences.
And knowing that the writer participated in Carolingian political legitimacy does not tell us which mechanism of legitimacy the writer believed mattered.
That has to be discovered from his record.
Perhaps oath mattered.
Perhaps fidelity mattered.
Perhaps assembly mattered.
Perhaps consensus mattered.
Perhaps agreements mattered.
Perhaps guarantees mattered.
Perhaps breach mattered.
Perhaps the durability of political relationship during the ruler’s absence mattered.
You cannot determine any of that by announcing “court propaganda” and walking away.
You have to read.
THEY INTERPRETED THE MOTIVE BEFORE THEY UNDERSTOOD THE MECHANISM
This is the deepest methodological inversion.
The motive-first chain looks like this:
We know what kind of writer this was.
Therefore:
We know his political interest.
Therefore:
We know what kind of story he wanted to construct.
Therefore:
We know why he selected these words.
Therefore:
We know what reality really existed beneath those words.
Every step moves farther from the object while pretending to move closer to the event.
Eventually the historian possesses more confidence about the writer’s hidden meaning than the surviving words warrant.
That is interpretation acquiring counterfeit standing.
The admissible chain goes the other direction:
HERE IS THE WORD.
HERE IS THE SENTENCE.
HERE IS THE ENTRY.
HERE IS HOW THE SAME RECORD USES DIFFERENT LANGUAGE WHEN DIFFERENT THINGS OCCUR.
HERE IS THE CHRONOLOGY.
HERE IS THE RELATIONSHIP BEFORE THE EVENT.
HERE IS WHAT CHANGED AFTERWARD.
HERE IS AN INDEPENDENT SOURCE, IF ONE EXISTS.
NOW LET US DISCUSS MOTIVE.
That is not naïve literalism.
That is evidentiary order.
ERASE “THE SAXON WARS” FOR FIVE MINUTES
This is where the inherited historical picture begins wobbling.
The phrase already contains a story.
THE SAXON WARS.
Once that summary enters the reader’s mind, every annual entry arrives already interpreted.
Charles travels into Saxony?
Military campaign.
pugna?
Battle.
An agreement follows?
Terms imposed after defeat.
obsides?
Hostages seized from the conquered.
Charles leaves?
Campaign complete.
Relationship subsequently fails?
Rebellion.
Charles returns?
Suppression.
Repeat.
After enough entries have been translated through that framework, the reconstructed chronology is presented as proof that the framework was correct.
But the framework participated in producing the chronology.
That is what must be audited.
Erase the summary.
Read the sequence.
Who moved?
Who assembled?
Who contradicted whom?
What exact verb was used?
What relationship existed immediately beforehand?
What agreement followed?
Who furnished obsides?
Who gave fides?
What was sworn?
Was there a placitum?
Who departed?
What happened when Charles was gone?
When did he have to return?
Where is rebellion explicitly named?
Where is killing explicitly named?
Where does devastation appear?
Where is authority explicitly acknowledged?
Where does the record say the road becomes open?
Now let the historical structure emerge from the record.
Do not force the record to impersonate the story inherited from above.
THE ROAD IS NOT DECORATION
This is why the road matters so much.
Look at 785.
The text is not shy about coercion.
Charles sends detachments.
The record says rebellious Saxons were plundered.
Fortifications were taken.
Strong places were entered.
Roads were cleared.
Then comes the extraordinary line.
After the public assembly at Paderborn, Charles proceeds through Saxony:
vias apertas nemini contradicente
—the roads open, with nobody contradicting him—
and travels wherever he wishes.
Then Widukind and Abbi enter the sequence.
They ask for assurances that they will remain unharmed.
Charles returns to Francia.
And what happens next is especially important:
Charles sends obsides to them through his envoy Amalwin.
They receive those obsides and then come to Charles at Attigny.
There they are baptized.
Only then does the entry conclude:
tunc tota Saxonia subiugata est
—then all Saxony was subjected.
Read the sequence.
Do not replace it.
Violence is there.
Plundering is there.
Fortifications are there.
Rebellion is there.
The writer is perfectly willing to tell us these things.
But so are:
roads,
contradiction,
assurances,
obsides moving from Charles toward Widukind and Abbi,
movement,
arrival,
baptism,
and finally the explicit statement of subjection.
That sequence matters.
Especially the obsides.
If somebody approaches the Latin carrying the automatic modern scene:
“the victorious king takes hostages from defeated Saxons,”
785 creates a serious problem.
Because here Charles sends obsides to Widukind and Abbi as the assurance under which they will come to him safely.
That usage forces us to stop pretending the English mental picture supplied by “hostage” automatically exhausts the political function of obses.
The object itself tells us otherwise.
Now the word has to be studied relationally.
Who furnishes the obsides?
To whom?
For what undertaking?
What obligation does the transfer secure?
What happens after the security is furnished?
That is how language acquires meaning from evidence.
Not from inherited imagery.
And the road itself is equally revealing.
Why tell us the roads were open?
Why tell us nobody contradicted him?
Why tell us that he could go wherever he wished?
Because movement itself is evidence of political condition.
Something has changed.
The record tells us that passage, contradiction and standing are connected.
So follow the road.
Do not paint around it.
SOURCE FIRST DOES NOT MEAN SOURCE INFALLIBLE
Someone will immediately object:
“But historical sources can lie.”
Of course they can.
That proposition changes nothing.
A lying witness is still authoritative evidence for what the witness testified.
Then you prove the testimony false.
That distinction is elementary and indispensable.
Suppose a source says:
The king entered the city on Monday.
Another independent contemporary record proves the king was hundreds of miles away until Wednesday.
Excellent.
Now we have contradictory evidence.
We investigate.
What we do not do is rewrite the first source:
“What the author really meant was Wednesday, because he was a court writer.”
No.
The source said Monday.
If Monday is wrong, prove Monday wrong.
Preserve the error.
The error itself becomes evidence.
Perhaps it exposes chronology.
Perhaps motive.
Perhaps copying.
Perhaps retrospective construction.
But the historian does not repair testimony inside the witness’s mouth.
The same applies here.
Reading exactly what a source says is not equivalent to believing that everything it says corresponds perfectly to external reality.
Those are separate operations:
WHAT DID THIS WITNESS SAY?
and:
WHAT ACTUALLY HAPPENED?
The first must precede the second because the second depends upon comparing properly identified evidence.
If historians begin changing the contents of the witnesses before comparison, they destroy the very evidence needed to reconstruct the event.
USE THE SOURCE OR CHALLENGE THE SOURCE
This is the rule that ends the nonsense.
If you use the source:
preserve its testimony.
If you challenge it:
show the contrary evidence.
If you reject it:
say why.
But you cannot do this:
“The Royal Frankish Annals are our principal evidence for this event, but because they were court-connected we cannot trust what they say, so we will reconstruct what they really meant, and then cite the Royal Frankish Annals for our reconstruction.”
No.
You have just hollowed out the source and retained its name.
That is appropriation of standing.
The citation makes the reader believe the historical object carries the proposition.
But the proposition entered through the historian.
That distinction must remain visible.
SOURCE CLAIM
is not
HISTORIAN’S INFERENCE.
Both may be useful.
Both may even be correct.
But they do not possess the same provenance.
AND DO NOT CHEAT BY BORROWING A LATER REVISION
Now we reach the second law.
It is absolute.
A later revision cannot testify from the temporal position of the earlier textual layer.
The Royal Frankish Annals survive in multiple textual forms. Scholarship has long distinguished an earlier annalistic text from a later reworking, and studies of the manuscript tradition emphasize the need to keep those textual states analytically distinct.
Good.
Then actually keep them distinct.
Suppose the earlier textual state says:
A.
A later reviser writes:
A + B.
B has later provenance.
That is not an insult to B.
It is a timestamp.
Maybe the reviser possessed better information.
Maybe B came from another record.
Maybe B clarified something.
Maybe B corrected something.
Maybe B supplied explanation.
Maybe B supplied motive.
Maybe B reshaped the political presentation.
Maybe B improved the prose.
All of those are hypotheses requiring evidence.
But the first uncontestable fact is:
B WAS NOT CARRIED BY A IN THAT EARLIER TEXTUAL STATE.
Therefore B cannot be smuggled backward and cited as though the earlier witness said A+B.
A revision is not a time machine.
WHAT A REVISION ACTUALLY PROVES
A later revision is historically valuable.
Extremely valuable.
But classify the evidence correctly.
The revision is direct evidence that:
SOMEBODY LATER CHANGED THE RECEIVED TEXTUAL OBJECT.
That fact is real.
Now compare.
What changed?
What was added?
What disappeared?
What became explicit?
Where did motive appear?
Where did political characterization change?
Where did narrative detail expand?
Where did the later writer make something legible that the earlier textual state left terse?
Those are excellent historical questions.
But they are questions about the textual history.
Do not confuse them with the earlier event.
The chain must remain visible:
EVENT
↓
EARLIER ANNALISTIC TEXT
↓
LATER REWORKING
↓
MODERN EDITION
↓
TRANSLATION
↓
HISTORICAL INTERPRETATION
↓
TEXTBOOK SUMMARY
↓
POPULAR IMAGE
Every layer has standing.
Every layer has provenance.
Every layer has a date.
Collapse them and later interpretation begins walking backward through time.
A characterization supplied later starts sounding eighth-century.
A translator’s choice starts sounding Latin.
A historian’s inference starts sounding annalistic.
A textbook summary starts sounding like eyewitness testimony.
And eventually everyone “knows” a story whose individual propositions nobody has traced back to the objects supposedly carrying them.
That is exactly what provenance prevents.
INTERPRETATIONS HAVE TIMESTAMPS TOO
This is the larger principle.
A source has provenance.
So does a revision.
So does a translation.
So does an interpretation.
So does a historical summary.
A later inference does not inherit the date of the evidence from which it was inferred.
A translation does not automatically inherit the semantic precision of the Latin merely because it appears on the facing page.
A revision does not become earlier testimony merely because modern editors print the textual forms together.
And a modern historian’s sentence does not become an eighth-century fact merely because a footnote underneath it points toward the Royal Frankish Annals.
INTERPRETATIONS HAVE TIMESTAMPS TOO.
That single recognition clears away enormous confusion.
Ask:
When did this proposition first appear?
Which object actually carries it?
Does the earliest record carry it?
Does it first appear in another contemporary source?
Does it enter through a later recension?
Does it appear in a translation?
Does it first appear as an inference in modern historiography?
These questions do not weaken history.
They finally give every historical claim its proper standing.
THIS IS WHERE THE MOTIVE-FIRST APPROACH BECOMES ABSURD
Rosamond McKitterick has made influential arguments about the political construction of Carolingian historical memory. In discussing the rise of Carolingian kingship, she explicitly warned that later historians had accepted an influential Frankish narrative of those political changes and argued that the validity of that narrative needed to be reassessed.
That instinct—to return to the evidentiary basis—is valuable.
Apply it completely.
Do not stop halfway.
Do not say:
“The inherited story may be a construction,”
and then respond by constructing another story from presumed motive before allowing the sentences to establish their own contents.
The antidote to an inherited interpretation is not another interpretation granted superior standing.
The antidote is provenance.
What does the earliest relevant object actually carry?
What does another independent object carry?
Where do they differ?
Where does the later characterization first appear?
That method does not need psychological omniscience.
It needs records.
THE ORIGINAL WRITER’S MOTIVE COMES AFTER THE TEXT
This rule applies even when we strongly suspect political purpose.
Perhaps the writer wanted to legitimate Carolingian authority.
Fine.
Now demonstrate how the text constructs legitimacy.
Do not decide in advance.
Perhaps the mechanism is military victory.
Perhaps it is consensus.
Perhaps it is divine favor.
Perhaps fidelity.
Perhaps rightful judgment.
Perhaps oath.
Perhaps peaceful passage.
Perhaps the capacity to establish an agreement that survives the king’s physical departure.
Perhaps several of these simultaneously.
The text decides which mechanisms it actually contains.
Indeed, scholarship already finds Carolingian consensus and divine agency inside the writing.
Those are not reasons to stop reading.
They are reasons to read more carefully.
Because now we have to understand what the writer thought legitimate political order actually looked like.
Who consents?
Who swears?
Who owes fidelity?
Who breaches it?
Who judges?
Who guarantees?
Who travels?
Who contradicts?
Who may pass?
When does the relationship hold without the king standing physically in front of everyone?
Those may be the political mechanics hiding in plain sight.
THE REVISER’S MOTIVE ALSO COMES AFTER THE CHANGE
And the law does not change merely because we are suspicious of a later reviser.
Do not replace motive-first interpretation of the original writer with motive-first interpretation of the reviser.
The revision itself establishes one minimal proposition:
THE REVISER WANTED THE RECEIVED TEXT TO BE DIFFERENT IN THE PLACES HE ALTERED IT.
Why?
Now investigate.
Maybe correction.
Maybe additional evidence.
Maybe politics.
Maybe theology.
Maybe style.
Maybe memory.
Maybe explanation.
Do not guess.
Compare the objects.
The law applies to everyone.
WHAT THIS DOES TO HISTORICAL STANDING
Now the entire dependency structure becomes visible.
Modern histories of Charlemagne derive tremendous authority from surviving early medieval records.
The Royal Frankish Annals matter precisely because they are among the major narrative objects from which Carolingian political history is reconstructed. They were copied and reused by later Carolingian writers and became part of the larger Frankish historical canon.
Fine.
Then historical claims made in their name remain answerable to them.
You cannot say:
“According to the Royal Frankish Annals…”
and then give me something the annal did not say.
If you are making an inference, label it:
“I infer…”
If another source supplies the information, cite that source.
If a later revision supplies the language, timestamp the revision.
If archaeology supplies it, show the archaeology.
But do not attach the authority of the earliest source to a later proposition.
That is the whole standard.
THE HISTORIAN DOES NOT OUTRANK THE OBJECT
This is the line history needs restored.
The historian may know more than the annalist about some things.
The historian may have more sources.
Better geography.
Archaeology.
Philology.
Manuscript comparison.
Scientific dating.
Comparative law.
None of that gives the historian authority to change the annalist’s testimony.
It gives the historian authority to challenge it with evidence.
That is different.
The historian’s job is not:
make the source say what really happened.
The historian’s job is:
preserve what every source says, establish the standing of each piece of evidence, resolve conflicts where possible, and state clearly which conclusions are testimony and which are inference.
That is history.
Anything else risks becoming narrative laundering.
THE SOURCE SAYS A
Here is how narrative laundering works.
The source says:
A.
The historian thinks:
The writer was attached to the court.
Therefore:
The writer wanted X.
Therefore:
A must really be hiding B.
The translation begins carrying B.
The historical narrative begins saying B.
The textbook compresses B.
The documentary visualizes B.
A century later everyone remembers B.
Then someone finally opens the Latin and asks:
Where is B?
And everyone points toward A.
That is counterfeit standing.
The source never carried B.
The interpretive chain did.
So separate them.
DO NOT CALL THE INTERPRETATION THE SOURCE
This is the demand.
If the source says pugna, show pugna.
If the same sentence reports people being killed, show the killing language.
If another occurrence of pugna contains no death statement, do not silently move the death statement over.
If the source says rebellati sunt, say rebellion.
If the source does not, do not automatically supply rebellion merely because the modern chapter title says “Saxon rebellion.”
If the source says fides, preserve fides long enough to determine what relationship it creates.
If it says sacramentum, preserve the oath.
If obsides move from Charles to Widukind and Abbi to guarantee safe appearance, do not let the English word “hostage” erase the transaction.
If the record says roads were open and nobody contradicted Charles, do not dismiss those statements as narrative debris while supplying some unrecorded modern theory as the real event.
The words are the evidence.
Start there.
FOLLOW THE ROAD
The road gives us the perfect final test because nobody has to speculate about it.
785 says the roads were opened.
It says nobody contradicted Charles.
It says he traveled through Saxony wherever he wished.
It then gives us the negotiation with Widukind and Abbi.
It gives us their request for assurances.
It gives us Charles sending obsides to them.
It gives us their arrival.
It gives us baptism.
And then:
tunc tota Saxonia subiugata est.
Then all Saxony was subjected.
That is a sequence.
Respect it.
Maybe someone will construct a better explanation of the sequence than mine.
Good.
Bring it.
But bring the evidence.
Do not begin with:
“Charlemagne conquered Saxony.”
and then force every preceding verb to mean whatever that summary requires.
Let the source tell us when it considers subjection established.
Let it tell us when rebellion appears.
Let it tell us when killing appears.
Let it tell us when roads are opened.
Let it tell us who furnishes the guarantees.
Let it tell us when agreements hold.
Then argue.
That is the entire challenge.
THIS IS NOT ABOUT BELIEVING THE ANNALIST
It is about preserving evidence.
If tomorrow somebody produces an independent eighth-century witness demonstrating that one of these annalistic statements was false, excellent.
History advances.
Put the two objects together.
If archaeology contradicts the narrative, excellent.
Show it.
If another contemporary source provides missing casualties, excellent.
Cite it.
If a charter establishes a political relationship earlier than the annal does, excellent.
Bring the charter.
Nothing about this method protects the annalist from contradiction.
It protects the evidence from being rewritten before contradiction occurs.
That distinction changes everything.
SOURCE MATERIAL IS WHAT GIVES THE CLAIM STANDING
Without source material, we do not possess an earlier witness.
We possess later reconstruction.
That reconstruction may be excellent.
It may be brilliant.
It may even be correct.
But its standing is different.
Call it what it is.
The historian’s inference is the historian’s inference.
The translator’s rendering is the translator’s rendering.
The reviser’s addition is the reviser’s addition.
The original annal’s statement is the original annal’s statement.
Do not collapse them.
The historical record becomes stronger, not weaker, when every layer keeps its name.
HERE IS THE STANDARD FROM NOW ON
If you claim the Royal Frankish Annals say something:
SHOW THE LATIN.
If the Latin requires lexical interpretation:
SHOW THE USAGE.
If you claim a political relationship:
SHOW THE CHRONOLOGY THAT ESTABLISHES IT.
If you claim rebellion:
SHOW THE REBELLION OR THE ANTECEDENT STANDING THAT MAKES THE CATEGORY WARRANTED.
If you claim killing:
SHOW THE DEATH EVIDENCE.
If you claim obsides means a particular procedure:
SHOW WHO GIVES THEM, WHO RECEIVES THEM, AND WHAT OBLIGATION THEY SECURE.
If you claim motive:
SHOW WHAT IN THE RECORD WARRANTS THE INFERENCE.
If you rely upon another source:
NAME THE OTHER SOURCE.
If you rely upon a later revision:
LABEL IT LATER.
If you rely upon modern reconstruction:
LABEL IT INTERPRETATION.
Do not mix the layers.
Do not hide the timestamp.
Do not borrow standing.
THE FINAL ORDER
Not:
HISTORIAN
↓
THEORY
↓
PRESUMED MOTIVE
↓
REWRITTEN SOURCE
↓
FOOTNOTE
But:
OBJECT
↓
TESTIMONY
↓
LANGUAGE
↓
SEQUENCE
↓
OTHER OBJECTS
↓
CORROBORATION OR CONFLICT
↓
INTERPRETATION
That order restores historical accountability.
It also restores something far simpler:
the right of the dead to finish their own sentences before somebody explains what they supposedly meant.
The Royal Frankish Annals were close to power.
Good.
Read them.
They were written inside a political world.
Good.
Read them.
They participated in Carolingian legitimacy.
Good.
Now determine from the actual vocabulary what legitimacy meant in their record.
They lived inside a theological order.
Good.
Read how God, oath, fidelity, justice and mercy operate in the text.
They may have omitted things.
Produce the evidence for an omission where you claim one.
They may have been wrong.
Prove the contradiction.
They may have been partial.
Compare them against another witness.
But stop using “bias” as permission to write missing sentences.
Stop using “context” as permission to replace grammar.
Stop using “court connection” as permission to manufacture motive.
Stop using later revisions as though they traveled backward in time.
Stop attaching eighth-century standing to propositions that first appear centuries later.
And above all:
STOP SAYING THE SOURCE SAYS SOMETHING THE SOURCE DOES NOT SAY.
That is not a minor translation problem.
That is a provenance failure.
History cannot survive it.
The record must remain the record.
The revision must remain the revision.
The translation must remain the translation.
The interpretation must remain the interpretation.
Every claim gets its proper object.
Every object gets its proper date.
Every witness keeps its own testimony.
Then compare them.
Then test them.
Then challenge them.
Then interpret them.
But first:
READ THE RECORD.
Do not tell me why the annalist wrote it before telling me what he wrote.
Do not tell me what he secretly meant when his own vocabulary is sitting in front of us.
Do not put a later writer’s words into an earlier writer’s mouth.
Do not replace testimony with interpretation and leave the witness’s name in the footnote.
If you think the source is wrong:
PROVE IT WITH EVIDENCE.
If you think another account is better:
BRING THE OTHER OBJECT.
But if you invoke the source:
THE SOURCE GETS TO SAY WHAT THE SOURCE SAID.
Everything after that has its own provenance.
That is not radical.
That is history.
