Federated Suns Reforged

[Docwade]

Chapter Twenty-Two

3010: The Shape of Strength​

Strength had always been easier to count than to understand.

BattleMechs could be counted.

Regiments could be counted.

Factories could be counted.

DropShips, fighters, artillery pieces, tanks, tons of ammunition and graduates from military academies could all be placed into columns where someone on New Avalon could total them and someone in Sian, Luthien, Tharkad or Atreus could steal the total six months later.

That was why intelligence services liked numbers.

Numbers behaved.

Institutions did not.

On the last night of 3009, Ian Davion had opened an old Halstead file because memory was useful. The Federated Suns had entered the new year with more ways to become stronger without waiting for a crisis to explain why. That had been the achievement.

January began by demonstrating the danger.

The more machinery a system could produce, the easier it became to mistake machinery for strength.

The more formations it could raise, the easier it became to mistake names for armies.

The more information it could collect, the easier it became to mistake information for judgment.

By 3010, the Federated Suns had enough of all three to become dangerous to itself if it forgot the lessons that had made the growth possible.

The year therefore opened with a color nobody particularly liked.

Orange.

First Quarter

January - Orange

Colonel David Davion stared at the readiness display for almost ten seconds before asking the question everyone else had learned not to ask in quite that tone.

"Why is that orange?"

The staff officer beside the display looked relieved.

He had apparently been waiting for someone to notice.

The symbol represented the 4th Dragonlords.

The unit existed.

That was not in dispute.

Its BattleMech regiment existed. Its armor existed. Infantry existed. Artillery existed. Aerospace strength existed. Engineers, logistics, signals, medical and recovery elements all appeared on property books somewhere in the Federated Suns.

The colors had been presented in December.

It had a commander.

It had a headquarters.

It was also not ready for war.

The officer answered carefully.

"Yellow understates the integration problems."

David folded his arms.

"Then make it red."

"Red implies the formation cannot conduct useful combat operations."

"Can it?"

"Parts of it can."

David looked at the symbol again.

"So orange."

"Yes, sir."

Field Marshal Hall sat at the far end of the conference table, reading another report.

Without looking up, he said, "Orange is yellow that becomes red if people start pretending it is yellow."

David considered that.

"I hate that."

"Good."

Hanse Davion leaned back in his chair.

"I like it."

"No one asked you."

"I am wounded."

"You will recover."

Ian Davion ignored both of them.

"What does orange require?"

The staff officer changed the display.

The 4th Dragonlords opened into subordinate formations.

Green.

Yellow.

Green.

Red.

More yellow.

Some units had trained together for months. Others had only recently received final equipment. The artillery regiment was competent. Two infantry formations were individually Regular. One armor regiment contained a Veteran battalion. Several BattleMech companies had pilots whose individual skill exceeded the collective rating of the units around them.

None of that answered whether the RCT could fight as an RCT.

"Novaya Zemlya," Hall said.

Ian nodded.

The training world had stopped being a location and become a verb inside the AFFS.

Formations went to Novaya to discover what they actually were.

The 3rd Dragonlords had gone through the pipeline.

Now the 4th would go.

The 5th would begin forming behind it.

David looked at another symbol.

The 2nd Dragonlords.

Full reinforced RCT strength.

Stationed on Lindsay.

Green.

Not green readiness.

Green skill.

That was worse.

The 2nd Dragonlords had become an administrative accident.

It had formed during the years when the Dragonlords concept was still changing faster than the paperwork defining it. The 1st had never needed the later Novaya pipeline in quite the same way. It had drawn heavily from veteran assault formations and the Brigade of Guards, people who already knew how to fight and needed to learn how to become one new formation.

The 2nd had slipped through for the opposite reason.

It had grown.

Filled.

Equipped.

Moved to Lindsay.

Passed inspections built for an older system.

No one had lied.

That almost made it worse.

The RCT was exactly what its records claimed.

Full strength.

Green.

Ian looked at Hall.

"Can Novaya take them?"

"No."

Hanse frowned.

"Because?"

"Because the Fourth is going through it. After that the Fifth will need the pipeline."

David looked toward Hall.

"Then somewhere else."

Hall nodded.

"Training Command's recommendation is Point Barrow. Existing Combat Training Center. Enough range space for a reinforced RCT if we expand the exercise footprint temporarily. Dedicated Professional OPFOR can be brought in for the intensive phases. Veteran and Elite formations can rotate through selected exercises."

The map shifted rimward.

Point Barrow.

A world far enough from the principal fronts that an RCT could be trained without becoming an immediate operational temptation, close enough to the Taurian frontier that the training would never feel entirely academic.

David studied the movement estimate.

"How long?"

Hall shrugged slightly.

"Until they are Regular."

One officer cleared his throat.

"The unit is already operational."

Hall looked at him.

The officer reconsidered the sentence.

"Administratively."

"Good correction."

Ian asked, "Certification standard?"

"Regular. Minimum."

"No combat certification below that."

The officer looked uncomfortable.

"Highness, that may take most of the year."

"Yes."

"The Second is a reinforced RCT."

"I know."

"It is stationed on Lindsay for a reason."

"I know that too."

Ian waited.

The officer had nowhere useful to go.

Finally Ian said, "If the Combine attacks Lindsay tomorrow, they fight."

"Yes, Highness."

"If the Taurians attack, they fight."

"Yes."

"If I need every armed citizen on the planet, they fight."

"Yes."

"But if I have a choice?"

The officer understood.

"We do not deliberately commit a Green RCT."

Ian nodded.

"That is what the rating is for."

There was no speech afterward.

The order went out.

That was one of the ways the AFFS had changed.

Readiness had stopped being an accusation.

It was becoming information.

The 2nd Dragonlords received the movement order six days later.

Its commanding general read it twice.

The first time as a commander.

The second as a man whose formation had just been told that several years of work were insufficient.

His chief of staff watched him carefully.

"Well?"

"We go to Point Barrow."

"That is all?"

"No."

The general put the order down.

"We become Regular."

The chief of staff smiled.

"That sounded personal."

"It is."

That was acceptable.

Training Command did not require officers to enjoy correction.

Only to survive it honestly.

The rest of the January readiness board produced less dramatic arguments and equally useful results.

The 2nd Federated Suns Armored Cavalry had activated with Veteran companies and a Regular regiment. Its commander kept the rating. Cohesion was not arithmetic.

The 1st Crucis Cavaliers remained a deliberately independent BattleMech regiment. Its five machines still awaiting final acceptance remained in maintenance rather than being dragged onto parade grounds to satisfy photographs. Hanse continued to like the commander for that.

The 1st Crucis Cavaliers were not an incomplete RCT.

They were a BattleMech regiment intended to be movable, attachable and capable of integrating rapidly beneath a larger headquarters.

The distinction mattered.

So did the recurring answer that followed nearly every discussion.

Not everything had to become the biggest version of itself to be useful.

Time

The proposal that became the Avalon Borderers began with a question about distance.

It did not begin with BattleMechs.

That irritated several BattleMech officers.

Field Marshal Hall placed a map of the Federated Suns across the main table.

Not the political map.

The movement map.

Jump routes.

Recharge intervals.

Transport pools.

Warning times.

Likely raid corridors.

Regional reserves.

March Militia locations.

RCT stations.

The map made the Federated Suns look less like a nation than an argument with geometry.

Hall pointed toward a border world.

"Enemy arrives here."

Another symbol appeared.

"Nearest major RCT capable of decisive response is here."

A line connected them.

Then another.

Then another.

Time accumulated.

"Question," Hall said. "What do we buy?"

Someone answered, "More RCTs."

Hall nodded.

"Yes."

Then waited.

No one liked when he did that.

David finally said, "That isn't the whole answer."

"No."

Hanse looked across the map.

"More transport."

"Yes."

"Better warning."

"Yes."

"Local militia."

"Yes."

Hall still waited.

David saw it.

"Time."

Hall smiled.

"Now we have the problem."

The proposed formation did not need to defeat an invading RCT.

That was the first thing written on the board.

It did not need to hold every city.

It did not need to seek decisive battle.

It did not need enough BattleMechs to look impressive in an annual review.

It needed to make an attacker spend something.

Fuel.

Ammunition.

Time.

Attention.

Bridges.

Reconnaissance.

Security forces.

Sleep.

A commander who ignored it should suffer.

A commander who pursued it should lose time.

A commander who tried to destroy it should discover that it had already moved.

Hall wrote four words.

DELAY. DISRUPT. MISDIRECT. SURVIVE.

​

A younger officer stared at the last one.

"Survive?"

Hall looked toward him.

"Yes."

The officer hesitated.

"If the mission is delay -"

"Delay is not martyrdom."

Hall tapped the map.

"A formation that dies buying six hours cannot buy another six tomorrow."

David nodded.

The discussion moved from principle into organization.

That was where principles usually became expensive.

A reinforced BattleMech battalion gave the formation enough armored mass to threaten reconnaissance forces, punish unsupported advances and create tactical problems without pretending to be an RCT.

Four armor battalions gave it frontage and endurance. Two cavalry oriented. Two heavier direct-fire battalions.

Two mechanized or motorized infantry regiments gave it people.

That sounded obvious until someone remembered how much terrain people could hold when BattleMechs could not be everywhere.

Artillery.

Aerospace.

VTOLs.

Engineers.

Signals.

Medical.

Recovery.

Logistics.

Fuel.

Ammunition.

More trucks than anyone making recruiting posters would ever photograph.

Hanse looked over the growing table.

"This is not particularly light."

Hall smiled.

"Light compared with an RCT."

David said, "Not compared with the transport officer carrying it."

"No."

"How many can we support?"

That changed the meeting.

The answer was not twenty.

It was not ten.

It was not even five without qualification.

Training Command's estimate was blunt. Two or three Light Combat Teams consumed approximately the institutional burden of one conventional RCT, depending upon composition. A reinforced Dragonlords RCT could consume still more.

Ian studied the numbers.

"So we are not creating cheap RCTs."

"No, Highness."

"Good."

Several people looked surprised.

Ian pointed at the map.

"If we call them cheap RCTs, somebody will eventually try to use one like an RCT."

That sentence survived every later revision of the doctrine.

The Light Combat Team existed to create time.

The RCT existed to use it.

Twelve Centers

The militia problem was not the absence of training centers.

There were twelve.

That was enough to make the next problem scheduling.

Twenty-seven March Militia commands could not all disappear into training areas at once. Border defense still existed while people learned how to defend borders better.

Training Command therefore built the rotation like a shipping schedule.

One formation in.

One formation out.

Some cycles accepted two training formations simultaneously under standard CTC conditions. Intensive cycles using a dedicated Professional OPFOR usually concentrated on one formation because the aggressor package consumed range space, control staff, observers and exercise bandwidth that ordinary training did not.

No March was permitted to strip an entire defensive zone for convenience.

Units rotated by echelon.

Headquarters staffs came through when subordinate formations remained home under alternate command arrangements.

Battalions trained without dragging every regiment with them.

LCT packages practiced mobilization as part of the movement itself.

The system became complicated enough that one scheduler described it as "a mobilization exercise that never stops."

Hall approved the phrase.

That was precisely what he wanted.

A militia formation that only learned how to fight after it had already arrived at the range had missed half the lesson.

It needed to alert.

Load.

Move.

Draw ammunition.

Establish communications.

Accept attachments.

Lose attachments.

Change routes.

Operate under imperfect information.

Fight.

Then return without leaving its home district defenseless for three months.

The first quarter schedule went out with almost no public attention.

To the AFFS, it was administration.

To several foreign intelligence services, it would eventually become one of the most worrying documents of the year.

Because it showed that training was no longer being treated as an event.

It was becoming traffic.

The schedule became more complicated when Training Command formalized what the instructors had already been doing unofficially. Each CTC maintained two professional OPFOR commands, not because every exercise required two aggressor formations, but because one command could be rebuilding, retraining, maintaining equipment, or supporting a neighboring center while the other carried the primary exercise load. Twenty-four aggressor commands across twelve centers gave the system depth without pretending every range could run maximum-intensity opposition continuously.

The OPFOR commands developed personalities. Some became specialists in reconnaissance and delay. Others were particularly good at logistics raids, night movement, electronic deception, urban defense, or making artillery observers miserable. Hall resisted attempts to standardize those differences away. Standard communications and safety rules mattered. Standard imagination did not.

"If every opponent fights the same way," he told Training Command, "we have built choreography."

The scheduling staff learned the same lesson from a different direction.

In March, a Draconis March militia battalion scheduled for a CTC cycle received a real border alert forty-eight hours before movement. Ten years earlier, someone would have cancelled the training block, left an empty range, and congratulated the militia for remaining available. Instead the battalion stayed home, a Crucis March unit moved forward one slot, and the transportation package already heading toward the CTC was reassigned without being turned around.

Nobody called that flexibility before it happened.

Afterward everyone did.

The substitute unit arrived with less preparation than planned. Its commander complained that the movement packet had been written for someone else.

The CTC commander agreed.

"Good."

"Good?"

"The enemy will not mail you a packet written specifically for your convenience either."

The unit spent the first day discovering which parts of the borrowed plan transferred and which did not. That became training before the first simulated shot.

The original Draconis March battalion stood down from alert nine days later. It entered the next available cycle instead of being quietly forgotten.

Training Command tracked that too.

Cancelled was not the same as completed later.

By spring, the twelve centers were no longer merely places where formations went to become better at tactics. They were forcing the AFFS to practice the bureaucracy of readiness. Transportation had to know what could move. Quartermasters had to know what ammunition could follow. Regional headquarters had to cover the gap left by a battalion that was away. Commanders had to accept temporary attachments from units they did not know. Families had to know when people would return. Equipment had to come home in better condition than a careless training calendar would allow.

The centers also created a new kind of volunteer.

Professional OPFOR duty had once sounded like a punishment to officers who imagined the assignment meant pretending to lose for students.

The first returning veterans corrected that misconception aggressively.

OPFOR units did not pretend to lose.

They learned other formations quickly, learned terrain quickly, learned new equipment quickly, and became comfortable receiving a mission on Monday with attachments they had never met on Sunday. Their people became unusually good at recognizing what mattered inside an unfamiliar order of battle.

That made OPFOR duty attractive.

Young officers applied because they wanted repetitions.

NCOs applied because the ranges offered access to equipment, instructors and units from across the Federation.

Technicians applied because keeping aggressor machines running through repeated exercises taught failure modes no peacetime depot could reproduce as cheaply.

Hall read the first personnel report showing more qualified applicants than open OPFOR billets.

"That took less time than expected."

The personnel officer nodded.

"People noticed where the learning is."

Hall looked at the twelve CTC symbols.

That would matter later.

IRON MIRROR

The New Avalon Combat Training Center received the first full operational test of the Light Combat Team concept before the concept possessed an official permanent formation table.

Exercise IRON MIRROR used two provisional OPFOR Light Combat Teams.

They were not permanent formations.

That was part of the test.

Training Command built them from professional aggressors, rotating infantry and armor, engineers, artillery, aviation, signals, medical and logistics elements, and enough staff officers to make the organizations behave like formations rather than collections of weapon systems. Each provisional LCT possessed a reinforced BattleMech battalion, four armor battalions split between cavalry and heavier direct-fire roles, two mechanized infantry regiments, artillery, aerospace support, VTOL reconnaissance, engineers and a mobile support group deliberately forced to operate with less fixed infrastructure than most AFFS formations preferred.

Their equipment was intentionally ordinary.

No experimental BattleMechs.

No secret NAIS weapons.

No hand-picked collection of machines selected because someone wanted the opposing force to look frightening in photographs.

Clints, Centurions, Shadow Hawks, Swordsmen, Javelins, Dervishes, older rebuilt machines, Tiger IIs, Manticores, J. Edgars, Pegasus hover tanks, Hunters, weapons carriers and trucks.

The sort of equipment the AFFS expected to have in quantity rather than the sort a procurement office displayed beneath bright lights.

No one wanted the exercise dismissed afterward because the OPFOR had been given miraculous hardware.

The opponents were not selected for mercy.

The 1st Davion Guards and Davion Heavy Guards received the mission.

Two RCTs.

Both experienced.

Both accustomed to being the formation other people called when a problem had become too important to leave alone.

Their mission was operational rather than tactical.

Restore communications across the exercise zone.

Reopen the principal logistics route.

Secure Fort Edward.

Five days.

Exercise Control would not extend the clock because a bridge was inconvenient, a convoy was late, a battalion had suffered simulated losses or a commander believed another twelve hours would make the result look better.

At one hundred twenty hours the exercise ended.

The OPFOR mission was simpler.

Prevent them.

Colonel David Davion received command of both provisional LCTs.

That surprised fewer people than it would have several years earlier.

He had spent enough time inside exercises, command-system trials, logistics studies and temporary operational groups that Training Command no longer needed to explain why the assignment made sense. What interested Hall was not whether David could devise an ambush.

Most competent officers could devise an ambush.

The question was whether he could command two mobile formations for five days without turning himself into the most overworked company commander in the exercise.

Leftenant Colonel Jasmine Rashid received the combined OPFOR BattleMech group.

Administratively, the two reinforced BattleMech battalions still belonged to their LCTs. Operationally, Rashid coordinated them as a maneuver group whenever David needed BattleMech weight concentrated across the boundary between the two teams. The battalion commanders retained their battalions. Company commanders retained their companies. Jasmine was not there to steer scores of machines one at a time.

She was there to make them arrive where the Guards did not want them, remain just long enough to create a problem, and leave before superior combat power turned the problem into a graveyard.

The arrangement pleased her immediately.

The briefing did not.

David stood in front of the OPFOR command group the evening before the exercise opened. Two LCT commanders. Jasmine. Artillery. Engineers. Aviation. Logistics. Signals. Intelligence. Senior enlisted leaders. Exercise-control liaisons.

He put the mission on the screen.

Five days.

Fort Edward.

Then he removed almost everything else.

A blank operational map remained.

"You have the detailed plan," he said. "Forget it for a minute."

No one moved.

"What are we buying?"

One of the LCT commanders answered first.

"Time."

"With what?"

"Ground."

"Sometimes."

The engineer commander said, "Bridges."

"Sometimes."

Jasmine leaned back in her chair.

"Their patience."

David looked at her.

"That too."

He restored the map.

"We are not defending Fort Edward in a line. We are not fighting two Guards RCTs because someone gave us a heroic speech and five days of ammunition. We are making them solve problems in sequence. Every problem they solve should reveal another one. If a position stops buying time, leave it. If a route becomes predictable, stop using it. If you can make them deploy a battalion to kill a platoon that is already gone, do it."

He looked toward the BattleMech officers.

"If you get a good fight, ask whether it is also a useful fight. Those are not the same thing."

Jasmine smiled faintly.

"That part was for me."

"It was for everyone."

"But especially me."

David ignored her.

"No one dies for an exercise objective. No one stays because withdrawing feels embarrassing. Exercise Control can kill a unit with a marker. Real terrain can kill people. The second one matters more."

Then he pointed toward the two RCT headquarters symbols.

"Headquarters are valid targets under exercise rules. If one becomes isolated, report it. Do not attack without confirmation from me."

That did attract attention.

One LCT commander asked, "Why?"

"Because I do not yet know whether killing one helps the exercise."

"In war it would."

"Yes."

"Then shouldn't we train that?"

"We are. They will know when they exposed themselves. I may still kill them. But if we decapitate an RCT on the first day, the remaining four days become a command-succession exercise. That may be useful. It may not be the exercise Hall asked us to run."

Hall, observing from the rear, said nothing.

David continued.

"Your authority is broad. The LCT commanders own their sectors and their formations. Rashid owns the BattleMech fight I assign her. Artillery decides firing positions inside the fire plan. Engineers decide how to create the effect I ask for unless the effect itself needs approval. Logistics moves before it becomes a target."

He paused.

"If I am telling a company commander which hill to stand on, either the situation is very unusual or one of us has failed."

The command sergeant major of the first provisional LCT nodded once.

That was as close to applause as the briefing received.

The final slide contained four words.

DELAY. DISRUPT. MISDIRECT. SURVIVE.

Jasmine read them.

"Which one matters most?"

David answered immediately.

"The last one. Otherwise the first three only happen once."

The exercise began at 0500.

The Guards expected ambushes.

They got paperwork.

The first problem appeared at Bridge Eleven.

The bridge crossed a broad training river that was shallow enough for infantry in several places and deep enough to make heavy vehicles dislike improvisation. It carried one of the few roads rated for continuous heavy transport between the Guards' initial assembly area and the central exercise zone.

OPFOR engineers did not destroy it.

That would have been too simple.

They placed simulated charges against one support span and conducted a partial demolition calculated to leave the structure standing while reducing its certified load.

When the first Guards engineers arrived, Exercise Control handed them the result.

Bridge passable to light and selected medium vehicles.

Heavy transport prohibited pending inspection and reinforcement.

The Guards now possessed a bridge they could use for some traffic, a bridge they could not use for the traffic they most wanted, and an engineering problem that refused to become either open or closed.

They could repair it.

They could bypass it.

They could ford selected vehicles elsewhere.

They could split traffic.

Every answer consumed something.

Time was one of those things.

The second problem appeared seventeen kilometers north when a traffic-control node began issuing contradictory routing information using valid AFFS authentication.

The node was not captured.

Its data package had been.

A convoy obeyed an authenticated route update and discovered ten minutes later that the depot at the end of the route did not exist.

It existed on the exercise network.

That was worse.

By the time the convoy turned around, the road behind it held another column moving in the opposite direction.

No shots had been fired.

The Guards had already lost forty-three minutes of road capacity.

The first actual contact came from Jasmine.

David did not tell her where to make it.

He sent one message.

I need the western axis to look dangerous until 0730. Do not become fixed west of the river.

Jasmine read it from the command cockpit of her BattleMech and passed three orders.

One company moved visibly.

A second moved quietly.

The third did neither.

It waited.

The visible company used Clints and faster mediums to show itself at the edge of reconnaissance range, fired enough to be taken seriously and withdrew before the Guards could organize a proper response.

The quiet company moved along a folded line of low terrain south of the road.

The waiting company remained behind a wooded ridge with engines idling low and weapons cold.

The 1st Davion Guards did what professional soldiers did when an enemy BattleMech force appeared on the flank of a vulnerable route.

They deployed against it.

Not recklessly.

Not in panic.

A reconnaissance element pushed forward.

A line battalion adjusted.

Artillery prepared fires.

Aerospace controllers requested an updated picture.

That was all Jasmine needed.

The quiet company appeared against the reconnaissance element's opposite flank.

Not close enough for a decisive engagement.

Close enough to force another decision.

The Guards turned part of the screen.

The waiting company fired LRMs over the ridge, less to inflict damage than to make the Guards wonder how much force remained unseen.

Then Jasmine left.

One Guards company pursued for six minutes.

A Clint lance kept itself visible just long enough to encourage the mistake.

Then it accelerated away toward terrain already registered by OPFOR artillery.

The Guards company commander saw the danger before crossing the final rise and stopped.

Good.

Jasmine did not try to tempt him twice.

At 0723, the western axis went quiet.

At 0728, Guard reconnaissance reported no significant BattleMech force west of the river.

At 0730, OPFOR engineers completed the work they had needed covered.

A secondary crossing went out of service.

Jasmine reported only:

WESTERN TASK COMPLETE. TWO COMPANIES GREEN. ONE AMBER FOR AMMUNITION. DISPLACING.

David read it and moved to the next problem.

He did not ask which lance had fired from which ridge.

He did not need to know.

The first morning continued that way.

The second provisional LCT used cavalry armor to threaten the flank of a logistics column, then broke contact before the Guards could bring heavier armor into the fight.

Infantry occupied a village long enough to force a deliberate reconnaissance and left through prepared rear exits before the first assault element entered.

Artillery fired six missions from three positions and displaced after each one.

The fourth firing position was a decoy.

The Guards counterbattery section found it and killed twelve inflatable gun signatures with admirable precision.

The OPFOR artillery commander sent them a congratulatory message after the exercise.

During the exercise she kept the joke to herself.

Near noon, one OPFOR BattleMech company stayed too long.

Its commander had an opportunity against a Guards armored battalion moving without full BattleMech support.

He asked Jasmine for permission.

She gave it.

The first exchange went well.

The second went better.

The third lasted thirty seconds too long.

A Guards BattleMech battalion appeared on the company's northern flank while artillery boxed the southern withdrawal route.

The OPFOR commander called for assistance.

Jasmine saw the trap.

So did David's staff.

An operations officer looked toward him.

"We can order Red Two's reserve company across the river."

David watched the display.

"What did Rashid order?"

"She told the company to break east by platoons and ordered her reserve to demonstrate north."

"Then she has it."

The officer hesitated.

"They may lose half the company."

"Yes."

"We could reduce that."

David looked toward him.

"By doing whose job?"

The officer stopped.

The answer was obvious.

Jasmine was commanding the BattleMech group.

The trapped company commander had a withdrawal mission.

The reserve commander had an intent.

David had no decision to make yet.

He made himself sit back.

That was harder than it looked.

The trapped company escaped with simulated casualties approaching fifty percent.

The reserve lost two machines in the demonstration.

The Guards owned the field.

Forty-seven minutes had passed.

Thirty-five had been required by OPFOR engineers at the crossing behind them.

The bridge came down on schedule.

At the first evening update, the company commander expected criticism.

He received it.

Not for losing machines.

For staying through the third exchange without knowing what enemy formation could reach his flank.

Then Jasmine added:

"You also bought twelve more minutes than we required."

The commander frowned.

"So was it worth it?"

"No."

He looked surprised.

"We needed thirty-five. You gave us forty-seven and lost half a company doing it. Extra time is not free just because we like time."

David heard that from across the command post and said nothing.

He did not need to.

That was why Jasmine was commanding the BattleMechs.

The first night belonged to infantry.

The Guards expected that too.

They strengthened perimeter security around headquarters, ammunition and fuel.

The OPFOR did not attack any of those first.

It attacked recovery.

A mechanized infantry platoon dismounted four kilometers from the Heavy Guards maintenance park and moved on foot through terrain most BattleMech officers regarded as irrelevant because nothing weighing fifty tons could use it.

The platoon carried exercise markers, wire cutters, passive beacons and no desire to die proving bravery.

A diversion began twenty minutes before they reached the perimeter.

Jasmine's BattleMech group appeared on sensors twelve kilometers west.

Not all of it.

Enough.

Heavy signatures moved along a ridge.

Aerospace reconnaissance caught two companies changing position.

Artillery fired smoke.

The Guards watched the obvious threat.

The infantry entered from the opposite side.

They did not attack BattleMechs.

They marked three heavy transporters, two tracked recovery vehicles and a mobile machine shop as destroyed.

Then they tried to leave.

Military police caught most of them.

The MPs were pleased.

Exercise Control was less helpful.

The marked recovery vehicles remained dead.

The next morning, a Guards battalion entered combat with BattleMechs capable of fighting and fewer ways to recover them if they stopped.

The battalion commander protested that the vehicles had never been near the line.

The CTC observer answered, "That appears to have been the point."

Day Two began with fatigue.

Not dramatic fatigue.

The more dangerous kind.

Drivers had slept four hours.

Headquarters staff had slept less.

Maintenance crews had worked through darkness repairing simulation damage that did not stop being mechanically real merely because the weapons were firing training effects.

The Guards remained stronger.

The OPFOR remained more tired than David liked.

That was also part of the test.

An LCT that could delay for six hours and then require two days to recover was not a useful operational formation.

At 0610, Red One's logistics officer recommended pulling two ammunition vehicles farther east before the day's movement began.

The operations staff objected.

The vehicles were farther from the units they supported if moved.

David listened for twenty seconds.

Then asked, "Who owns the decision?"

The room stopped.

The logistics officer said, "Red One commander, sir."

"Then why am I hearing it?"

The operations officer began, "Because movement affects—"

"Everything affects something. Does it cross an exercise restriction?"

"No."

"Does it change my reserve?"

"No."

"Does it prevent Red Two from using the route?"

"No."

David looked toward the liaison screen.

"Then let Red One command Red One."

The logistics vehicles moved.

Three hours later Guards reconnaissance found their previous location and called artillery onto an empty field.

The point was not that David had predicted the strike.

He had not.

The point was that he had not delayed a subordinate's useful decision while trying to make it his own.

The first headquarters isolation occurred at 0942.

It happened accidentally enough that everyone recognized afterward how easily it could have happened in war.

The Heavy Guards headquarters displaced forward after communications were restored along one portion of the main route. The move was reasonable. Remaining farther back would have increased communications delay and separated the command group from the engineer effort controlling the next crossing.

Red Two's cavalry found the movement.

Jasmine's scouts found something more interesting.

The headquarters security battalion had been delayed behind a weight-restricted bridge while the headquarters group used a secondary road suitable for lighter command and communications vehicles.

An OPFOR engineer team then closed the secondary road behind them with a simulated crater.

Not planned specifically for the headquarters.

The opportunity simply appeared.

For twenty-two minutes the Heavy Guards command group sat east of one obstacle, west of another, with most of its dedicated security force on the wrong side of the river and only a reinforced company immediately available.

Jasmine called David.

Her voice was calm.

"I have the Heavy Guards headquarters."

David looked at the map.

"You have eyes?"

"Two sources. One visual. One signals."

"Security?"

"Reinforced company close. More moving. I can put a battalion on them in twelve minutes. Two companies in seven."

The command post became very quiet.

The Heavy Guards headquarters symbol sat in a pocket of terrain that suddenly looked much smaller than it had five minutes earlier.

David asked, "What are they using for fuel on the east axis?"

Jasmine paused.

"That wasn't the question I expected."

"Answer it anyway."

The intelligence officer brought up the route.

A forward fuel distribution point eight kilometers southeast.

Guard armor depended on it for the next day's planned tempo.

David marked it.

"Take that."

Jasmine said nothing for two seconds.

"David."

He noticed she had stopped using rank.

That meant she thought he was being especially unreasonable.

"Yes."

"I can kill their headquarters."

"In a real fight, yes."

"This is a combat exercise."

"And if you kill it, what do we train for the remaining three and a half days?"

"Their succession of command."

"Useful. Not today's question."

She exhaled.

"The fuel point is less satisfying."

"Yes."

"They will recover."

"Eventually."

"They will know we could have hit the headquarters."

"Eventually."

A smile entered her voice.

"That may be more satisfying."

David looked toward the LCT commanders.

"No one closes the headquarters pocket. Let them reopen it. Rashid takes the fuel point. Red Two makes the road feel dangerous and then leaves."

He did not tell Jasmine which companies to use.

She took one.

Not the two she had prepared for the headquarters strike.

One company attacked the fuel point from two directions while a second showed itself far enough north to make the Guards believe the heavier blow was still coming toward the command group.

The fuel point died in simulation twenty-nine minutes later.

The Heavy Guards headquarters survived.

Its eastern armor schedule did not.

When the security battalion finally reopened contact with its headquarters, the Guards commander received a private exercise-control note.

HEADQUARTERS WAS TACTICALLY ISOLATED FOR TWENTY-TWO MINUTES. OPFOR POSSESSED SUFFICIENT FORCE TO ATTEMPT DECAPITATION STRIKE. NO STRIKE ORDERED.

The commander read it twice.

Then asked his chief of staff one question.

"Why?"

The chief could not answer.

That bothered both of them more than a simulated headquarters kill might have.

By noon, the Guards had changed their movement discipline.

Security no longer followed headquarters as a separate problem.

Route status updates became mandatory before each displacement.

The correction cost staff time.

It also made them better.

That afternoon, the OPFOR hit them repeatedly and accomplished very little tactically.

Jasmine's BattleMechs appeared against one flank, fired, displaced and reappeared forty minutes later somewhere else.

One company used a wooded drainage line to approach within effective range of a Guards artillery security element, fired two volleys, then disappeared before the artillery's BattleMech escort could close.

Another company made a deliberate show of crossing a ridge while Red One's armor raided a convoy on the opposite side of the exercise zone.

The Guards stopped chasing individual contacts.

That was the first sign David saw that IRON MIRROR was working.

He was pleased.

It made his problem harder.

By the evening of day two, the first day's easy delays were gone.

The Guards had begun asking not "where is the OPFOR?" but "what does the OPFOR want us to look at?"

Jasmine read the same change.

She called at 1830.

"They're not biting."

"Good."

"You keep saying that when my life becomes difficult."

"Training Command has corrupted me."

"They are starting to screen before they move."

"Good."

"And they are holding reserve instead of committing to every contact."

"Good."

"I may stop giving you reports."

David smiled.

"What do you need?"

Her tone changed immediately.

"Permission to shift the combined BattleMech group south tonight. I want them to see us west at dusk and lose us before midnight."

"Effect?"

"Tomorrow they wake expecting pressure on the west route. I want to hit the central engineer train when it starts moving."

"Restrictions?"

"None. I need Red One's cavalry out of my way by 0300 and one artillery battery for smoke if the break goes bad."

David looked at the map.

The move crossed both LCT boundaries.

That made it his decision.

"Approved. Coordinate with both LCT commanders. Artillery support on call, not automatic."

"Understood."

He did not ask for her route.

At 0215, his operations officer asked where Jasmine's lead company was.

David looked at the display.

"Somewhere south."

The officer blinked.

"Sir?"

"Does she need us to know more?"

"For deconfliction—"

"Artillery has her restricted boxes. Aviation has her corridors. Red One and Red Two have boundary coordination. Does headquarters need the location of every company?"

The officer looked at the screen.

"No."

"Then let them maneuver."

That answer would return to David months later during B-3000 work.

At the time it merely kept him from spending the night steering other people's BattleMechs.

Day Three opened with Jasmine's attack on the engineer train.

The Guards were ready for something.

They were not ready for that.

The western route received enough electronic traffic, artillery registration and visible reconnaissance to look like the obvious morning fight.

The combined OPFOR BattleMech group came from the south instead.

Not in one mass.

Three companies attacked sequentially.

The first struck the route-security screen and withdrew.

The second arrived as the Guards repositioned to pursue the first.

The third never attacked the screen at all. It went around and hit the simulated bridging equipment.

Jasmine broke contact before the Guards could mass two battalions against her.

The result was not spectacular.

Three OPFOR BattleMechs were marked destroyed.

Five more suffered simulated damage.

The Guards lost fewer machines.

The engineer train lost two bridging sections and forty-one minutes.

Jasmine considered that a good morning.

The Guards considered it irritating.

Both were correct.

The second headquarters isolation occurred at 1118.

This time it belonged to the 1st Davion Guards.

The opportunity came from their success.

The Guards had restored enough of the central communications network that their headquarters displaced into a temporary command site closer to Fort Edward. The location was tactically sensible and electronically cleaner than the previous site.

It was also adjacent to a road network the OPFOR had spent two days studying.

A Red One engineer platoon had already prepared two road denial points.

A cavalry troop occupied the third exit for eleven minutes before withdrawing.

Jasmine's reconnaissance section found the command vehicles.

The headquarters was not helpless.

No RCT headquarters was.

It possessed security, BattleMechs close enough to matter and artillery that could make an attack expensive.

But it was isolated from the nearest full maneuver battalion for a window long enough to attempt the strike.

Jasmine called.

"Again."

David knew what she meant.

"Which one?"

"First Guards. Temporary site. Seventeen-minute window, maybe twenty if their engineers are slow."

David's intelligence officer confirmed it.

The staff expected the same answer as the previous day.

David did not give it immediately.

He looked at the exercise clock.

Fifty-four hours elapsed.

Sixty-six remained.

He looked at the Guards' pattern of adaptation.

Then at Fort Edward.

Then at the logistics overlays.

The 1st Guards had created a new maintenance route to support the forward move.

Its security was thin because the route had only existed for six hours.

David keyed Jasmine.

"Can you reach the headquarters?"

"Yes."

"Can you reach their maintenance route?"

A pause.

"Yes."

"Which one lasts longer after you leave?"

Jasmine laughed once.

"You are a deeply unpleasant man."

"Maintenance route."

"Understood."

The OPFOR never entered direct-fire range of the 1st Guards headquarters.

Instead, two BattleMech companies cut the maintenance route at separate points while Red One infantry struck the temporary vehicle park feeding it.

The headquarters remained alive.

The battalions it commanded began receiving increasingly ugly readiness forecasts.

The 1st Guards commander received the same private exercise-control message the Heavy Guards commander had received the day before.

HEADQUARTERS ISOLATED. OPFOR STRIKE FEASIBLE. NO STRIKE ORDERED.

He called the Heavy Guards commander.

The conversation was not recorded in the official report.

It probably should have been.

The operational consequences were.

Both RCTs began treating headquarters movement as a combat maneuver rather than administrative relocation.

Security elements moved first.

Alternative routes were verified.

Communications were established before the main command group displaced.

The lesson had cost no simulated headquarters.

It had still been learned.

Day three also exposed the weakness in David's own command arrangement.

Two LCTs generated too much information.

Not too much information to collect.

Too much information to treat as equally important.

The command post could tell him the ammunition state of a Tiger II platoon, the repair status of a Clint in Jasmine's second battalion, the fuel condition of a mechanized infantry company, the estimated bridge load at three crossings and the last reported location of every Guards battalion.

The temptation was to read all of it.

At 1430 he tried.

For seven minutes David disappeared into the display.

He opened a vehicle readiness exception.

Then a communications fault.

Then an artillery ammunition warning.

Then a report from Jasmine's group showing one BattleMech company displaced farther east than planned.

By the time he looked up, Red Two had already solved the communications fault, artillery had transferred ammunition between batteries and Jasmine had redirected the displaced company around a Guards screen.

The one issue that actually required David had waited.

Red One wanted to collapse a route that Red Two intended to use six hours later.

His chief of staff finally said, "Colonel."

David looked up.

"What?"

"What decisions do you need to make?"

The words landed harder because David had used them on other people.

He stared at the display.

Then laughed once.

"Fair."

The staff filtered.

Three hundred icons became twenty-eight.

Twenty-eight became six decisions.

One belonged to Jasmine.

Two belonged to the LCT commanders.

One belonged to artillery.

One belonged to logistics.

The last belonged to David.

He preserved the route Red Two needed and ordered Red One to create a different delay.

The command post moved again.

Afterward David wrote one sentence in his notebook.

THE ABILITY TO SEE A PROBLEM DOES NOT MAKE IT MINE.

He underlined it.

Day three ended with neither side comfortable.

That was probably the best measure of the exercise so far.

The Guards were behind schedule but adapting.

The OPFOR was ahead of its timing requirement but accumulating losses.

Artillery crews were tired.

Infantry had marched farther than expected because vehicle routes kept becoming unsafe.

Jasmine's BattleMech group had enough ammunition but less maintenance margin than she wanted.

One company had fought too often and needed to disappear for twelve hours simply because its pilots and technicians were becoming slow.

She pulled it.

No one objected.

The third headquarters isolation came before dawn on day four.

Unlike the first two, David's staff deliberately created this one.

The Heavy Guards had learned from day two.

Their headquarters moved with security.

Routes were checked.

Alternate routes were checked too.

So Red Two stopped attacking routes.

It attacked certainty.

A signals team generated evidence of an OPFOR command post east of the main river.

Not enough evidence to be obvious bait.

Enough to make the location worth watching.

Jasmine showed a BattleMech company in support.

The Heavy Guards did not pursue.

Good.

They sent reconnaissance.

Also good.

The reconnaissance found what looked like a lightly protected OPFOR communications node withdrawing south.

The Guards shifted one security battalion to close the route.

That was the decision Red Two wanted.

OPFOR engineers then closed a different road behind the Heavy Guards headquarters while cavalry demonstrated against the remaining exit.

For fourteen minutes the RCT headquarters was isolated again.

Not as badly as on day two.

Not as long.

Enough.

Jasmine's lead BattleMech battalion was nine minutes away.

She called David.

"Third time."

"I know."

"This one we made."

"I know."

"You are still going to say no."

David studied the map.

The Heavy Guards had built a reserve specifically to answer this kind of problem.

If Jasmine attacked, she might kill the headquarters in simulation.

She might also lose a battalion to the reserve afterward.

The trade could be justified in war.

The exercise was not war.

David looked at the Guards' recovery battalion moving north.

It carried the machines that would keep yesterday's damaged BattleMechs in tomorrow's fight.

"No headquarters."

Jasmine sighed theatrically.

"Of course."

"Take the recovery battalion."

Silence.

Then:

"That is worse."

"For them?"

"Emotionally."

"Good."

She moved.

The Heavy Guards headquarters reopened its route.

The recovery battalion did not escape cleanly.

Jasmine struck it, damaged enough recovery equipment to affect the day's repair schedule and withdrew under smoke before the Guards reserve could close.

The Heavy Guards commander received his second private notice.

He did not need to read the entire message.

He saw the first line.

HEADQUARTERS ISOLATED.

He closed his eyes.

Then called his operations officer.

"New rule."

"Sir?"

"If anyone tells me a route is secure, ask what that means."

The operations officer hesitated.

"In what sense?"

The commander looked at him.

"Exactly."

By midmorning day four, the Guards stopped making the same mistakes.

That was when IRON MIRROR changed.

The first three days had belonged to the OPFOR.

The fourth belonged to learning.

The Guards stopped trying to destroy every OPFOR element they found.

A Clint lance appeared on the flank.

Reconnaissance tracked it.

No battalion turned.

An infantry platoon fired from a village.

The Guards isolated the village, marked it for follow-on clearance and continued toward the operational objective.

An OPFOR armor company threatened a convoy.

The convoy did not stop while every escort vehicle tried to kill the raiders. One security element engaged. The rest kept the convoy moving.

Scouts moved ahead of engineers.

Engineers moved closer to the formations they supported.

Fuel stopped moving on predictable schedules.

Convoys gained decoys.

Command posts gained decoys too.

The first false headquarters almost worked perfectly.

Jasmine's reconnaissance identified the emissions package.

Movement matched a staff displacement.

Security looked hurried.

She committed a cavalry company supported by a BattleMech lance to probe.

The "headquarters" contained enough electronics to look convincing and enough concealed Guards armor to make attraction dangerous.

The OPFOR cavalry company died in simulation.

The BattleMech lance escaped with one machine marked damaged.

Jasmine stared at the replay.

David called.

"Status?"

"They baited me."

"I saw."

"I disliked it."

"Good."

"You are enjoying this too much."

"What did you learn?"

"That their false headquarters is more aggressive than their real one."

David laughed.

Then stopped when she continued.

"And that I wanted the headquarters badly enough to believe evidence I would have questioned yesterday."

That mattered more.

"What now?"

Jasmine looked at the map.

"I stop hunting headquarters."

"Good."

She disconnected before he could say it again.

The Guards' improved discipline began compressing the OPFOR's freedom of movement.

Roads that had been safe on day two were watched on day four.

Artillery registered likely withdrawal corridors instead of simply firing at contacts.

Aerospace reconnaissance stopped following decoys once they crossed thresholds that made them operationally irrelevant.

The 1st Davion Guards began using one battalion not to attack but to deny the OPFOR the space it needed to disengage.

That was much more dangerous.

Jasmine discovered it at 1530 when two of her companies hit a Guards engineer concentration and found the expected withdrawal route occupied by a battalion that had not reacted to the attack at all.

She had three options.

Fight through.

Reverse.

Split.

David watched the tactical display.

His staff watched him.

No one asked whether he wanted to issue orders.

They had learned too.

Jasmine split.

One company withdrew north under smoke.

The other went south, accepted simulated damage crossing an exposed rise and forced a Guards company to choose between pursuing it and maintaining the blocking position.

The Guards commander held.

Correct.

Jasmine lost four BattleMechs in simulation.

Also correct.

The remaining machines escaped.

When she reported, her voice sounded tired.

"They are getting annoying."

David looked at the clock.

"Seventy-eight hours."

"I know what time it is."

"Then you know why annoying is acceptable."

"Ask me again in twelve hours."

The OPFOR spent the fourth night moving.

Not fighting.

Moving.

Artillery batteries displaced.

Field kitchens displaced.

Medical stations displaced.

The two LCT headquarters displaced independently to reduce the chance that a single Guards thrust could compromise both.

David's mobile headquarters moved last.

He wanted the Guards to see enough traffic to believe a larger movement was occurring north.

Most of the meaningful combat power moved south.

Jasmine's BattleMechs refueled in darkness under light discipline.

Technicians worked with red lamps and profanity.

One Clint needed an actuator replacement.

A Centurion had a simulated weapons casualty but a real coolant leak discovered during inspection.

Exercise Control would have permitted the machine to continue if the leak could be isolated.

Jasmine's maintenance chief grounded it.

A young pilot objected.

"It's an exercise."

The chief stared at him.

"The coolant does not know that."

The Centurion stayed down.

By dawn, OPFOR strength looked ugly.

That pleased no one.

It also meant the numbers were honest.

Day Five began at 0500 with forty-eight OPFOR BattleMechs still fully combat effective across Jasmine's group, additional machines degraded or restricted, armor losses severe enough that several battalions existed more convincingly on the map than on the road, and infantry regiments that had spent four days proving human beings remained useful in terrain designed to make machines important.

The Guards were stronger.

The Guards were also late.

Fort Edward remained outside their control.

The principal logistics route was open only in sections.

Communications had been restored across most of the exercise zone but remained vulnerable along two relay chains.

One hundred twenty hours no longer looked generous.

The Guards knew it.

So did David.

The final day stopped resembling the first.

There were fewer ambushes because the Guards no longer entered obvious ambushes.

There were fewer raids because the Guards had learned to protect what mattered and ignore what did not.

The fight became geometry.

The Heavy Guards pushed along the northern route with engineers integrated directly into the advance.

The 1st Davion Guards controlled the central corridor and used artillery to deny likely OPFOR withdrawal lines.

The two RCTs stopped behaving like two separate hammers and began handing problems between them.

David watched the change and felt the uncomfortable satisfaction of an instructor whose students were becoming capable of beating him.

At 0715, one of the LCT commanders proposed a counterattack against the central corridor.

The staff liked it.

The opportunity was real.

The attack could probably destroy a Guards battalion that had moved ahead of its heavier support.

David asked one question.

"What does it buy?"

Silence.

The LCT commander looked at the map.

"Two hours. Maybe."

"Cost?"

"A third of my remaining armor if they respond quickly."

"And if they don't?"

"Less."

"Do we need two hours there?"

Another pause.

"No."

David nodded.

"Then don't buy time we cannot spend."

The counterattack disappeared.

Instead, Red One withdrew through the sector and left engineers behind to create three smaller obstacles.

None delayed the Guards two hours.

Together they delayed them fifty-three minutes at much lower cost.

Jasmine spent the morning on the western flank.

David's instructions to her were short.

KEEP THEM LOOKING WEST UNTIL 1030. REMAIN CAPABLE OF MOVING SOUTH AFTERWARD. DO NOT GET FIXED.

She answered:

YOU MAKE EVERYTHING SOUND EASY.

He replied:

THAT IS WHY YOU GET THE BATTLEMECHS.

Her first attack began at 0812.

A company emerged against the Heavy Guards flank, fired at range and withdrew.

The Guards tracked but did not pursue.

Her second attack came from a different axis twenty-six minutes later.

Again the Guards refused to chase.

Jasmine changed the problem.

At 0910 she put a heavier company in a position from which it could threaten the engineer vehicles supporting the northern route.

Now the Guards had to react.

Not because they wanted the BattleMechs.

Because the engineers mattered.

A Guards battalion shifted.

Jasmine withdrew.

The battalion stopped.

She reappeared twenty minutes later.

The Guards held reserve.

She fired artillery through an attached observer instead and moved again.

The BattleMech exchange itself remained inconclusive.

The route did not.

The engineers lost time every time they stopped to protect themselves.

At 1026, Jasmine reported:

WESTERN TASK COMPLETE. DISPLACING SOUTH.

Four minutes early.

David did not congratulate her.

He gave her the next job.

Fort Edward's western approaches contained three routes suitable for heavy BattleMechs.

The Guards needed only one.

David needed all three to look dangerous.

Jasmine understood immediately.

"You don't need me to hold them."

"No."

"You need them to clear them."

"Yes."

"That is much cheaper."

"Try to keep it that way."

She broke the BattleMech group into smaller elements.

Not random detachments.

Purposeful problems.

A company occupied a tree line long enough to be detected and left before contact.

A lance fired from an old quarry, then withdrew through a route seeded with false mine markers.

Another company used passive emissions to suggest a battalion waiting behind high ground.

The Guards did not believe all of it.

They did not need to.

They had to verify enough of it that the advance remained deliberate.

At 1130, Fort Edward was still outside their control.

At 1200, it was closer.

At 1300, the 1st Davion Guards had forced one western route and the Heavy Guards were pushing engineers through the northern approach.

The OPFOR was running out of room.

That was expected.

What David had not expected was how quickly his own headquarters would become part of the problem.

At 1417, a Guards reconnaissance element found a relay vehicle that had transmitted twelve seconds too long.

The relay itself was expendable.

The traffic pattern behind it was not.

Within twenty minutes the Guards had a probable location for David's mobile headquarters.

The first warning came from signals.

The second came from a cavalry screen disappearing from the map.

The third was visual.

A Heavy Guards BattleMech appeared on a distant ridge behind the headquarters convoy.

Jasmine saw the contact through a shared feed.

"David."

"I see it."

"I can turn back."

Her BattleMech group was already moving south toward the Fort Edward approaches.

Turning back would sacrifice the final delay plan to protect the command post.

David looked at his chief of staff.

The evacuation routes were ugly.

One remained open.

"No."

Jasmine's answer came immediately.

"They have a battalion closing."

"I know."

"Your security company will not hold them."

"It doesn't have to."

"David—"

"Your mission is Fort Edward. Mine is moving."

A pause.

Then her voice flattened into professional calm.

"Understood."

She did not turn back.

David's headquarters displaced under pressure.

The command vehicles left first.

Signals destroyed exercise-sensitive equipment they could not move.

The security company fought a short delaying action and withdrew by sections.

One mobile headquarters vehicle took a simulated kill.

David's did not.

The convoy cleared the last junction seven minutes before Guards armor reached it.

Jasmine called once the channel was secure.

"That was close."

David was looking at the clock.

"Good."

"You have an unusual definition of good."

"They are learning."

"They are about to kill us."

"Also learning."

The OPFOR headquarters survived.

Barely.

The final six hours belonged to the Guards.

That was important.

They had earned them.

They bypassed positions that could no longer affect the mission.

They pushed engineers through under direct protection.

Artillery suppressed suspected OPFOR routes instead of chasing individual firing points.

Aerospace reconnaissance maintained contact rather than trying to turn every detection into a strike.

The Guards began controlling the geometry of the battlefield.

The OPFOR stopped being able to choose where every engagement occurred.

Jasmine's BattleMechs fought shorter and shorter actions.

One company appeared near Fort Edward, fired, withdrew and found the withdrawal route watched.

It broke east instead, losing three machines in simulation.

Another company tried to threaten the northern engineers and discovered Guards artillery already registered on the likely approach.

It never fired.

Jasmine pulled it away.

At 114 hours, she had enough combat power for one significant attack.

Her operations officer wanted to use it.

So did she.

She called David.

"I have one punch left."

"Where?"

"Anywhere you need it."

David looked at Fort Edward.

The Guards needed the final southern access road.

Red Two infantry still held observation over it but could not stop a full RCT assault.

The road itself had been cleared once already.

He could spend Jasmine's remaining force there.

The BattleMechs might buy forty minutes.

Maybe an hour.

They might also cease to exist as an effective group.

David looked at the clock.

Six hours remained.

Then at the northern route.

The Heavy Guards were closer than he liked but still dealing with an engineer obstacle and damaged recovery capacity.

The central route had congestion.

The southern road did not need an hour.

It needed uncertainty.

"Do not attack."

Jasmine was silent.

"Say again."

"Do not attack. Show them enough that they cannot assume the road is clean. Then move west."

"You are saving the reserve."

"I am saving the option."

"For what?"

"I don't know yet."

That answer satisfied her more than a false certainty would have.

The BattleMechs appeared.

Sensors saw them.

The Guards slowed.

Reconnaissance moved.

Artillery prepared.

By the time the Guards proved the road was not defended by the full OPFOR group, Jasmine was gone.

Thirty-one minutes had passed.

The reserve still existed.

At 117 hours, David finally spent part of it.

Not against the main body.

Against the engineer element that had moved farthest ahead.

Jasmine hit with two companies, fired hard for six minutes and withdrew before the Guards could close the trap.

The engineers survived.

Their equipment did not all remain operational in simulation.

Clearance slowed again.

The Guards kept moving.

At 119 hours, Fort Edward's outer exercise boundary was under Guard control.

The fort itself was not.

Communications were functioning.

The principal logistics route was open to restricted traffic.

The northern approach was being cleared.

The Heavy Guards had enough combat power to crush what remained of Red Two if given time.

The 1st Davion Guards had turned the central route into something the OPFOR could no longer cross safely.

David's staff watched the clock.

119:32.

A Guards company entered the outer fort training area.

119:41.

Engineers opened a lane through the final obstacle.

119:48.

Jasmine's surviving BattleMechs began withdrawing under the preplanned end-state order rather than making a last stand nobody had requested.

119:55.

The Heavy Guards headquarters issued its assault order.

120:00.

Exercise Control froze the network.

Every tactical icon stopped.

For a moment nobody in David's command post spoke.

Then the exercise-result banner appeared.

OPFOR MISSION SUCCESS.

No one cheered.

The OPFOR had won the exercise.

It had not beaten the Davion Guards in battle.

Its armor was wrecked.

Several infantry battalions were combat ineffective.

Artillery crews were exhausted.

Jasmine's BattleMech group had enough functioning machines to remain dangerous and not enough to pretend another two days would end well.

Red One had lost much of its freedom of movement.

Red Two's logistics were approaching the point where another full day would force choices David did not like.

The Guards remained stronger.

On day six they probably would have destroyed the remaining OPFOR formations or forced them out of the exercise area.

There was no day six.

That was the mission.

The after-action review began after food, medical checks and enough sleep to reduce the chance that exhaustion would masquerade as insight.

The main room contained more rank than comfort.

The Guards sat on one side.

The OPFOR command group on the other.

Exercise Control occupied the center like a court everyone claimed not to fear.

Hall stood beside the mission board.

He did not begin with casualties.

He began with the order.

Restore communications.

Reopen the principal logistics route.

Secure Fort Edward.

Five days.

Then he displayed the final state.

Communications restored.

Principal route open under restrictions.

Fort Edward not secured at 120 hours.

A Guards officer spoke first.

"We destroyed almost twice their combat power."

Hall nodded.

"Yes."

"We would have destroyed the rest on day six."

"Probably."

"We were winning."

Hall looked at the mission board.

"No."

The officer's expression tightened.

Hall did not soften.

"You were killing them."

He pointed at Fort Edward.

"You were not doing this."

Then the logistics route.

"This was not fully available."

Then communications.

"This recovered on day four."

He looked back at the room.

"The OPFOR did not need to survive forever. It needed to make five days insufficient."

The Heavy Guards commander leaned forward.

"And if this had been a real operation?"

Hall looked toward David.

David answered.

"We would have been in serious trouble by day six."

"Earlier than that."

"Yes, sir."

The commander changed the display.

Day two.

His headquarters isolation appeared.

"You had me here."

David nodded.

"Yes, sir."

The 1st Davion Guards commander changed it again.

Day three.

"And me here."

"Yes, sir."

The Heavy Guards commander advanced to day four.

"And again."

"Yes, sir."

The room became quieter.

The Heavy Guards commander stared at David.

"Three times?"

"Three useful windows."

"You did not attack any of them."

"No, sir."

"Why?"

David answered without smiling.

"Because you were more useful to the exercise alive."

Silence.

Then someone laughed.

Not the Heavy Guards commander.

Not immediately.

He looked at David for another second.

"Colonel, I am developing a professional dislike for you."

David nodded.

"I've been told that can be healthy."

That got the laugh.

The 1st Guards commander did not let the point disappear into humor.

"In war, you strike."

"Yes, sir."

"No hesitation?"

"None. If I can decapitate an RCT headquarters at acceptable cost, I do it. The exercise-control record should mark all three as catastrophic vulnerabilities."

Hall nodded.

"It does."

David continued.

"But killing the headquarters on day two would have trained your replacement headquarters for three days. We needed to train your headquarters to stop becoming killable."

The Heavy Guards commander looked at the day-four replay.

His second isolation window.

Then at the final day, where his headquarters moved under security and with verified alternatives.

"We learned."

"Yes, sir."

"You could have made the score look better."

David shrugged slightly.

"The score wasn't my mission."

Hall looked pleased.

That made everyone suspicious.

He changed the display to Jasmine's BattleMech group.

"Leftenant Colonel Rashid."

"Sir."

"What did you learn?"

Jasmine did not look at the kill totals.

She looked at the readiness graph.

It declined steadily across five days.

"That surviving is harder than delaying."

A few heads turned.

She continued.

"On day one we could choose most contacts. Day two we could still choose where to become visible. Day three they started asking better questions. Day four they stopped chasing. Day five they were controlling our exits."

She highlighted the final hours.

"Hit-and-run works because there is somewhere to run. Once the enemy controls enough routes, every raid spends mobility. Every firing position becomes more expensive. Every damaged machine becomes harder to recover."

Hall asked, "Could you have held Fort Edward another day?"

"Held? No. Delayed around it? Maybe."

"At what cost?"

Jasmine considered.

"Most of the BattleMech group if the Guards continued improving."

"Then?"

"Then I stop being an LCT BattleMech group and become salvage."

Several officers smiled.

She did not.

"Delay is not magic. We controlled tempo for four days. On the fifth they took most of it back. If the exercise ran another forty-eight hours, we probably lose the formations."

Hall nodded.

"Good."

The Guards officer who had spoken earlier looked irritated again.

"You keep saying good when people describe losing."

Hall turned toward him.

"Because knowing where doctrine stops working is part of doctrine."

He changed the display again.

This time it showed David's headquarters message traffic.

The volume was ugly.

Thousands of reports.

David winced.

Hall noticed.

"Colonel. How many company-level maneuver orders did you issue personally?"

David thought.

"Very few."

"Number."

His chief of staff answered for him.

"Seven that directly changed company movement, sir. Most during headquarters displacement or deconfliction."

Hall looked at David.

"Why?"

"Because I had two LCT commanders and a BattleMech group commander."

"You had the information to issue more."

"Yes."

"You could see individual companies."

"Yes."

"Then why not command them?"

David looked at the display.

"Because seeing them didn't make them mine."

Hall waited.

David continued.

"My job was to decide what the two LCTs needed to accomplish, where their actions conflicted, when to commit cross-boundary assets, what risk we accepted and what information mattered at the combined level. Rashid's job was the BattleMech fight. The LCT commanders owned their formations. If I used the network to reach past them every time I saw something I disliked, I would make their headquarters decorative."

The Heavy Guards commander said, "And slower."

David nodded.

"And slower."

Hall changed the screen to the seven-minute period on day three when David had disappeared into subordinate detail.

"Except here."

David stared at it.

"Yes."

"What happened?"

"I tried to read everything."

"Why?"

"Because I could."

"Useful?"

"No."

"What did you miss?"

"A route-conflict decision that actually required me."

Hall nodded.

"Keep that thought."

David looked suspicious.

Hall ignored him.

The AAR continued for four hours.

No one left thinking the OPFOR had discovered a magical formation.

That was important.

The Light Combat Team had weaknesses.

It could not stand against an RCT indefinitely.

Its lighter logistics package created endurance limits.

Its artillery and armor could be attrited faster than they could be replaced inside the formation.

Once the enemy controlled enough routes, disengagement became progressively harder.

A competent opponent adapted.

The Guards had.

But the exercise also demonstrated what the concept could buy.

Five days in which two RCTs spent ammunition, fuel, engineer effort, maintenance hours, reconnaissance capacity, staff attention and operational time solving a force they significantly outmatched in direct combat.

The Guards won most of the fights.

They lost the timetable.

That distinction became the heart of the report.

Hall asked the final question.

"What was the OPFOR there to do?"

Jasmine answered first.

"Make them spend time."

One LCT commander added, "And stay alive long enough to do it again."

Hall looked toward David.

David studied the mission board.

"We weren't there to win battles."

He looked at Fort Edward.

"We were there to win the campaign."

The sentence went into the report.

So did the warning beneath it.

DO NOT CONFUSE SUCCESSFUL DELAY WITH INFINITE DELAY.

The Light Combat Team concept survived IRON MIRROR.

More importantly, it became less romantic.

Thirty Days

David received the B-3000 orders three days after IRON MIRROR.

Temporary duty.

New Avalon Institute of Science.

AFFS Command Systems Development Group.

B-3000 Field Architecture Project.

Thirty days.

Jasmine read the order over breakfast.

"Thirty?"

"That is what it says."

Major Kara O'Sullivan looked over his shoulder.

"That means ninety."

David looked offended.

"It does not."

Kara continued eating.

Jasmine nodded.

"Ninety."

"I have been ordered for thirty days."

"You have been ordered to look at a command system."

"Yes."

Kara looked at Jasmine.

"Ninety."

David put the order down.

"I live in a hostile household."

"No," Jasmine said. "A predictive one."

He was gone thirty-four days before the first extension.

Neither woman mentioned it.

That was worse.

The B-3000 did not begin with a smaller computer.

It began with more people.

That irritated engineers trained to think of progress as miniaturization.

The original Star League B-2000 command computer had fitted inside a Cyclops cockpit because Star League engineers had solved packaging problems the Federated Suns had not yet recreated.

The AFFS had taken a different route.

B-250 for company command.

B-500 for battalion.

B-1000 for regiment.

B-2000V for brigade or reinforced regimental headquarters, carried in a Heavy Mobile Headquarters vehicle because people required volume even when electronics eventually did not.

The next question was an RCT.

Not merely more units.

More kinds of units.

BattleMechs.

Armor.

Infantry.

Artillery.

Engineers.

Aerospace.

Transport.

Medical.

Recovery.

Signals.

Logistics.

And the increasingly dangerous assumption that because a headquarters could receive all of that information, the commander should see all of it.

David had already learned what happened when a command display became a very expensive way to drown.

The B-3000 project existed to make drowning faster unless they solved that problem first.

One engineer demonstrated a prototype operations screen.

Three hundred and eleven active tracks.

Forty-two subordinate reports.

Weather.

Fuel.

Ammunition.

Casualties.

Airspace.

Artillery availability.

Electronic warfare.

Repair status.

Transport.

Engineering.

David watched for almost a minute.

Then asked, "What am I supposed to do with this?"

The engineer hesitated.

"Command."

"No."

The engineer looked confused.

David pointed toward the display.

"This tells me everything."

"Yes."

"That's the problem."

The room went silent.

David called over one of the staff officers assigned from Training Command.

"What do I need to know?"

The officer looked at the screen.

"Which part?"

"Exactly."

They spent the next six hours removing things.

Not deleting them.

Moving them.

Company data belonged primarily with battalions.

Battalion detail belonged primarily with regiments.

Regimental detail moved upward only when it crossed thresholds.

Green did not need to become a general's problem merely because a sensor could report it.

Amber might.

Red did.

Uncertainty also did.

David wrote a sentence across the top of the working board.

SYSTEM MUST SUPPORT DECISION. NOT REPLACE JUDGMENT.

​

No one erased it.

The first B-3000 field simulation failed for a different reason.

The system performed well.

The staff did not.

Every section discovered that it could see everyone else's work.

Operations began correcting logistics priorities.

Logistics began arguing with movement routes before Operations had finished designing them.

Intelligence opened artillery overlays and started recommending fires.

Artillery officers responded by asking Intelligence to stop doing artillery.

The commander tried to settle all of it.

Within twenty-three minutes, he had become the busiest person in the headquarters and the least useful.

David stopped the exercise.

The commander looked irritated.

"The system worked."

"Yes."

"Then why stop?"

David pointed around the headquarters.

"Because everyone else stopped."

The commander looked at the staff.

They were waiting for him.

Every one of them.

David said, "You turned a headquarters into one man with several thousand assistants."

The commander stared at the display.

"I had the information."

"So did they."

That became the second lesson.

The B-3000 was not valuable because it let the commander know everything.

It was valuable because it let the headquarters know enough of the same things to work together.

That required boundaries.

Authority.

Trust.

And the discipline to leave a subordinate problem subordinate until it stopped being one.

Jasmine's War

While David argued with computers, Jasmine inherited the OPFOR program.

Not officially forever.

Officially while Colonel Davion was detached.

Training Command had learned that temporary arrangements were one of the easiest ways to discover permanent talent.

Jasmine did not run the OPFOR the way David did.

That pleased Hall.

A copy would have been less useful.

David liked logistics traps.

Jasmine liked dilemmas.

She built exercises where every choice made sense and every choice cost something.

A battalion could reinforce the threatened bridge.

Doing so exposed the ammunition route.

It could secure the ammunition route.

Doing so gave the OPFOR freedom on the bridge.

It could split.

The OPFOR wanted that.

The first unit through her exercise complained that she had designed an impossible problem.

Jasmine read the complaint.

Then the AAR.

Then asked one question.

"Did they accomplish the mission?"

Hall said, "Yes."

"Then it wasn't impossible."

"They lost twenty-eight percent simulated strength."

"They would have lost forty if they tried the obvious plan."

Hall smiled.

"You sound pleased."

"No."

She looked at the casualty table.

"I sound like I'm learning."

Hall nodded.

That was the correct answer.

David returned from NAIS during the third exercise.

He sat in the observation room and said nothing for twenty minutes.

Jasmine knew he was there.

She did not turn.

Her staff made a recommendation.

She listened.

Asked two questions.

Approved it.

Ten minutes later the training formation changed axis and avoided the trap Jasmine had expected them to enter.

One of her operations officers frowned.

"We missed them."

"No."

Jasmine changed the map.

"They learned."

Behind her, David smiled.

Hall noticed.

"Your lesson survives."

David looked at him.

Hall pointed toward Jasmine.

"We're trying to teach her now."

David nodded.

"That seems fair."

Jasmine's second exercise looked nothing like David's.

Her staff expected another elaborate trap. Instead she gave the aggressor battalions broad boundaries, three protected logistics targets, and an instruction that irritated every officer who wanted to know the plan in advance.

"Make the training unit choose."

Her operations major asked, "Choose what?"

"Exactly."

The aggressor commanders went away unhappy and returned three hours later with four different ideas. One wanted to threaten the bridge. One wanted to raid the fuel farm. One wanted to make enough noise on the western axis that the training unit committed its reserve there. The fourth wanted to do almost nothing and wait for the training commander to create an opening through impatience.

Jasmine approved all four.

"Those plans contradict one another," the major said.

"Only if I make them happen at the same time."

The exercise became an argument conducted with battalions.

The training formation reinforced the bridge.

The aggressors touched the fuel route.

The reserve shifted west.

Nothing happened there.

That bothered the training commander enough that he shifted it again.

The fourth OPFOR battalion, which had spent most of the morning hidden and therefore looked useless on the command display, moved through the space the reserve had just vacated.

Jasmine's staff saw the opening immediately.

"Reinforce them?"

She shook her head.

"They found it. Let them use it."

"We can make the penetration decisive."

"Then I would be commanding their battalion for them."

The major stopped.

It was the same disease David had encountered in the B-series, only without a computer to blame.

The OPFOR battalion commander exploited the opening imperfectly. He moved too slowly at first, then too quickly, exposed one flank, lost a company in simulation, corrected, and still forced the training formation to abandon its preferred timetable.

At the AAR he apologized for the mistakes.

Jasmine looked at him.

"Did you need me to tell you where to go?"

"No, ma'am."

"Then I learned something useful."

So had he.

By the time David returned from NAIS, Jasmine's exercises had acquired a reputation among the CTC staff. David's aggressor work had taught formations that logistics and time were targets. Jasmine's taught them that a competent enemy might present several acceptable answers and punish the one they chose badly.

The two approaches were becoming complementary without becoming identical.

Hall considered that a better result than finding two officers who thought the same way.

New Valencia - Leaving Well

Jaime Wolf had already made up his mind before Ian arrived.

That simplified the meeting.

It did not make it easier.

Fort Jaime had become more than a military installation.

Families lived there.

Children had grown there.

Technicians knew Davion procurement officers by name.

Dragoon pilots argued with local mechanics about spare parts in the comfortable way people argued only after years of trust.

The Federated Suns contract had been good.

That was part of why Jaime intended to leave.

Mercenaries who stayed too long in one realm risked becoming something else.

Ian understood.

He disliked understanding.

Hanse tried once anyway.

"Longer term."

Jaime shook his head.

"Not this time."

"Better terms."

"No."

"More independent deployment authority."

Jaime smiled faintly.

"You already give us more than most employers."

"That was supposed to help my argument."

"It did."

Hanse sighed.

Ian looked toward Jaime.

"Is there something we did wrong?"

"No."

The answer came immediately.

Ian believed it.

That made the rest easier.

"Then where?"

"Capellan space."

Hanse's expression changed.

Jaime saw it.

"You asked."

"I did."

"You don't have to like the answer."

"I don't."

"Good."

Ian leaned back.

"We will honor the contract through completion."

"I know."

"No interference with departure."

"I know."

"No intelligence operation attached to your movement."

Jaime looked toward him more carefully.

"I know."

Ian smiled.

"That one you weren't sure of."

"No."

"Fair."

The gift came afterward.

That was important.

Not an inducement.

Not a payment contingent on reconsideration.

A Heavy Mobile Headquarters.

B-2000V.

​

Production-standard where production-standard existed.

Field-modified where it did not.

Dragoon technicians were invited to inspect it before transfer.

Jaime looked at the documentation.

"Why?"

Ian answered.

"Because you helped us learn how to build it."

The Dragoons had carried Star League command experience that the AFFS could study without pretending it owned.

Exercises had exposed weaknesses.

Kerston and other Light Horse officers had demonstrated the value of separating command assistance from the physical limitations of a BattleMech cockpit.

The Dragoons had done the same in different ways for years.

The gift recognized service.

Nothing more.

Nothing less.

Jaime closed the folder.

"Thank you."

Hanse watched him.

"You could still stay."

Jaime laughed.

"No."

Joshua Wolf caused the next problem.

He wanted thirty-six Waylands.

David stared at the request.

"Thirty-six?"

Joshua nodded.

"Yes."

"Do you know how many people currently want Waylands?"

"Yes."

"Apparently not."

Jaime looked amused.

Joshua did not.

"We can use them."

"I know."

"Then -"

"We can give you six."

Joshua blinked.

"Six?"

"By year-end."

"I asked for thirty-six."

"I noticed."

"Can you do twelve?"

"No."

"Eight?"

David looked at him.

Joshua smiled.

"Six."

"Six."

The agreement was signed.

The Dragoons would leave with fewer Waylands than they wanted and considerably more than most mercenary commands in the Inner Sphere could imagine requesting.

That too said something about the relationship.

Later, David walked the Fort Jaime maintenance lines with one of the Dragoon technicians.

He expected to talk about Waylands.

The technician wanted to talk about people.

"Your Ministry is good," he said.

David glanced over.

"They will be pleased to hear that."

"Don't tell them. It will ruin them."

"Naturally."

The technician stopped beside a Marauder undergoing actuator work.

"Most employers think a mercenary contract begins with a combat unit and ends with a payment schedule."

David waited.

"You people finally learned families are part of readiness. Dependents. Schools. Medical. Housing. Replacement parts. The things no one paints on a DropShip."

"Andrew started that argument."

"I know."

"It was unpopular."

"Good ideas often are."

David smiled slightly.

"That sounds suspiciously Davion."

"Don't insult me."

They continued walking.

Leaving well mattered.

If the Dragoons returned someday, it should be because the relationship had remained worth remembering.

If they did not, the Federated Suns should still be able to say it had honored the contract.

That was enough.

The contract closeout took longer than the conversation with Jaime.

That was appropriate.

Leaving well required more paperwork than deciding to leave.

The Ministry of Mercenary Relations sent teams to New Valencia months before the final departure window. Not negotiators trying to reverse Jaime's decision. Accountants. Housing officers. transport planners. medical administrators. quartermasters. claims specialists. People whose job was to ensure that departure did not become a final opportunity for everyone to discover an unresolved promise.

One ledger contained school placements for Dragoon dependents.

Another contained locally hired civilian employees whose contracts ended when specific units departed.

A third listed damaged Davion equipment awaiting reimbursement decisions.

A fourth listed Dragoon property inside AFFS depots.

Nobody put those ledgers on recruiting posters.

Jaime read all of them.

He found three errors.

The Ministry found two he had missed.

That pleased him more than a clean report would have.

One dispute involved a warehouse full of spare armor plate purchased partly with Dragoon money and partly under a Davion emergency-support account during the New Aragon campaign. The simplest solution would have been to split the stock by tonnage.

The warehouse chief objected.

"Half the plates fit machines they aren't taking in this shipment."

The contract officer looked at him.

"Then tonnage is the wrong answer."

They rebuilt the allocation by useful inventory instead.

Jaime heard about the correction afterward.

"Your people could have kept the difference."

Ian shrugged.

"Then the ledger would have lied."

"Most governments survive that."

"I am trying not to make it a habit."

The Heavy MHQ produced a different sort of suspicion.

Dragoon technicians opened panels, checked serials, examined software interfaces, compared production drawings against what had actually been installed, and looked for the hook everyone understood might be there even if politeness prevented anyone from saying so.

There was no remote lockout.

No maintenance dependency disguised as generosity.

No contract clause requiring the Dragoons to return the vehicle if they entered Capellan service.

No intelligence package buried in the communications architecture.

Jaime finally asked Ian directly.

"No conditions?"

"One."

Jaime's expression changed.

Ian let the pause last exactly long enough to become annoying.

"Use it."

Jaime stared.

Hanse laughed.

"He practiced that."

"I did not."

"You absolutely did."

The tension broke.

Later, Jaime found David watching a loading drill near the DropPort.

"You understand why we're leaving?"

David nodded.

"I think so."

"That is not the same as liking it."

"No."

Jaime watched a Dragoon family board a transport shuttle carrying too much luggage and one child who had apparently decided a stuffed animal required its own seat.

"Good employers are dangerous to mercenaries."

David looked at him.

Jaime continued.

"Bad employers remind you what you are. Good ones can make staying feel like belonging."

"Would that be terrible?"

"Not if we chose to stop being mercenaries."

That was the distinction.

The Dragoons had not come to the Federated Suns to become another Davion regiment with unusual paperwork. Their independence mattered because it was theirs.

David nodded toward the shuttle.

"Then leave as friends."

Jaime looked at him for a moment.

"That is the plan."

The last weeks at Fort Jaime became deliberately ordinary. Training continued. Repairs continued. Children went to school until their departure dates. Local merchants complained about losing customers. Davion soldiers attended farewell dinners and then reported for duty the next morning because sentiment did not suspend schedules.

The Federation did not hold a triumphal review for troops it did not own.

It held a departure ceremony small enough that the people leaving could recognize the people saying goodbye.

That was better.

Northwind - Giving Back a Name

Northwind had learned not to trust promises made by Great Houses.

That did not mean it had forgotten how to recognize respectful behavior.

Ian came personally.

No military escort larger than necessary.

No announcement that implied ownership.

No speech about ancient ties designed to turn history into leverage.

He met the Elders.

Then told them what the AFFS intended to do.

The 10th Deneb Light Cavalry would return to Rowe.

The regiment had spent enough time on Northwind.

Its work was complete.

The 9th Deneb Light Cavalry would be restored.

One Elder looked at Ian for a long moment.

"The Tenth carried what remained of the Ninth because someone had to."

Ian nodded.

"Yes."

"Now?"

"Now there are enough people to give it back."

No applause.

That was not how Northwind worked.

One Elder lowered his eyes.

Another nodded.

The decision moved forward.

The 5th Deneb Light Cavalry remained connected to Northwind.

The returning Highlander regiments complicated everything and improved it.

Six regiments had come home.

Not six regiments ready to accept permanent Davion employment.

Ian did not ask them to.

Stuart's Highlanders and the 3rd Kearny Highlanders accepted Davion garrison contracts first.

The others rested.

McCormack's Fusiliers.

The 2nd Kearny Highlanders.

The 1st Kearny Highlanders.

Marion's Highlanders.

Repair.

Recruit.

Rebuild family connections.

Argue with quartermasters.

Drink with people they had not seen in years.

Remember what home meant after service elsewhere.

No one in New Avalon described that as wasted time.

The old AFFS might have looked at four veteran mercenary regiments sitting on Northwind and asked why they were not immediately under contract.

The 3010 AFFS asked whether pushing too hard would destroy the reason they had returned.

That was progress measured in restraint.

The restoration of the 9th Deneb Light Cavalry became a quieter process.

The unit had a history.

That made the work harder, not easier.

A historical name was not permission to invent continuity where none existed.

Training Command and Personnel built the regiment from surviving lineage, returning personnel, Deneb cadre and new recruits.

The 10th transferred records, traditions and equipment that had been carried on behalf of the absent unit.

Not everything transferred.

The 10th had become itself too.

The first argument concerned colors.

The second concerned seniority.

The third concerned which trophies belonged to whom.

Ian stayed out of all three.

The soldiers settled them.

That was healthier.

By summer, the 9th existed as more than a planning line.

By autumn, it possessed a coherent regimental identity.

By December, it would return to the rolls.

No one hurried the calendar because a restoration ceremony would look good beside a Highlander contract announcement.

The first Highlander DropShip landed through weather that would have justified delay and did not require it.

Northwind watched anyway.

Families gathered behind safety lines at the field. Some had names on placards. Some did not need them. A few waited for people they had not seen in years. Others waited knowing the person they wanted would not step down the ramp and had come because the regiment returning home mattered even when one particular soldier did not.

Ian stayed away from the landing area.

That was deliberate.

A First Prince could turn a homecoming into his own event merely by attending it.

He met the Highlander leadership the next day.

One Elder thanked him for that absence before anything else.

Ian understood the compliment.

The six returning regiments created immediate military value and immediate civilian strain. Housing. medical appointments. school places. repair berths. dependents relocating from years of service abroad. veterans who wanted retirement on Northwind but had not yet decided where. technicians who had spent most of their adult lives maintaining equipment under Capellan supply systems and now needed cross-reference catalogs for Davion-standard parts.

The Strategic Refit Center became crowded.

No one complained publicly.

Privately, everyone did.

That was also home.

The contract talks with Stuart's Highlanders and the 3rd Kearny Highlanders took place only after both regiments had completed their first personnel accounting.

The AFFS offered garrison work.

The Highlanders asked who retained operational control.

They asked what happened if New Avalon wanted them off Northwind.

They asked who paid for battle damage, what salvage rules applied, how much notice ended the contract, and whether Davion commanders could detach companies individually without regimental approval.

The Ministry answered each question in writing.

One Highlander colonel read the draft twice.

"You expect us to argue with this."

The Ministry representative nodded.

"I would be worried if you didn't."

They argued for two days.

Then signed.

The other four regiments did not.

An AFFS liaison asked whether Ian wanted another offer prepared immediately.

"No."

"Highness, McCormack's and the Second Kearny are both highly capable formations."

"I know."

"We may lose the opportunity."

Ian looked through the window toward Northwind's hills.

"They came home. Let them be home before we ask what else they can do for us."

The liaison nodded slowly.

That instruction traveled farther than the contract office.

It told the Highlanders that rest was not a negotiating tactic New Avalon intended to punish.

The 9th Deneb restoration carried the same philosophy in miniature.

When Personnel asked for volunteers from the 10th to form the restored regiment's first cadre, it did not simply transfer names from a spreadsheet. Soldiers were asked.

Some volunteered because their families had served in the Ninth.

Some because restoring a lost regiment felt like work worth doing.

Some declined because the Tenth had become their regiment and history did not override present loyalty.

All three answers were accepted.

One sergeant major reviewing the transfer list found two brothers with the same family lineage choosing different units.

He started to ask whether they wanted time to reconsider.

Their father, retired and standing nearby, stopped him.

"They already did."

The brothers shook hands.

One went to the Ninth.

One stayed with the Tenth.

That was how a name returned without stealing another unit's identity to do it.

The Aranos

High Lord Tamati Arano II arrived on New Avalon with a diplomatic party large enough to demonstrate sovereignty and small enough not to pretend the Aurigan Coalition was a Great House.

Ian respected both choices.

Kamea Arano was young enough to find parts of New Avalon impressive that the adults had learned not to notice.

The palace was interesting.

The museums were better.

The simulator hall was best.

Hanse discovered this when he found her standing beside an Albion instructor while a cadet tried unsuccessfully to land a damaged aerospace fighter.

"He is too fast," Kamea said.

The instructor looked down.

"He is."

The simulated fighter hit the runway.

Badly.

Kamea winced.

Hanse said, "He'll do better next time."

She looked at him.

"He broke the landing gear."

"In the simulator."

"Yes."

"Which is why we have simulators."

Kamea considered that.

Then nodded.

"Can I try?"

Hanse looked at the instructor.

The instructor looked at the High Lord's daughter.

Then at Hanse.

Hanse smiled.

"I am not answering that."

The instructor sighed.

Fifteen minutes later Kamea broke the landing gear too.

She was furious.

The instructor reset the simulation.

"Again."

Tamati later asked about education.

Not annexation.

Not dependency.

Education.

Could Aurigan students attend selected Federated Suns institutions?

Could instructors rotate through Coalition schools?

Could technical curricula be exchanged without turning the arrangement into Davion control over Aurigan education?

Ian's answer was yes, provided everyone wrote down what yes meant before students began moving.

Jurisdiction.

Tuition.

Accreditation.

Security restrictions.

Which schools.

Which technical programs.

What information remained sovereign.

What happened when an Aurigan student failed.

What happened when one succeeded spectacularly and received an AFFS recruiting offer.

Tamati appreciated that last question.

"You thought of that quickly."

Ian looked toward David.

"We have learned that success creates administrative problems faster than failure."

David nodded.

"Failure usually explains itself."

Kamea listened to adults discuss tuition, credentials, jurisdiction and exchange quotas.

She became visibly bored.

David noticed.

"What would you rather be doing?"

"Simulator."

Tamati closed his eyes.

Hanse smiled.

The relationship began there.

Not with a treaty.

With a child wanting another attempt at a landing she had failed.

At dinner, Kamea ended up seated near Ardan Sortek, Thomas Davion and Hanse.

The seating had not been planned as a historical moment.

That was another reason it mattered.

Thomas explained why Atlas pilots disliked narrow hangar doors.

Ardan told him that Atlas pilots disliked anything that suggested the universe should not rearrange itself around ninety tons of entitlement.

Kamea laughed hard enough to spill water.

Hanse handed her a napkin.

No one in the room knew what those friendships would mean twelve years later.

That was good.

People should be allowed to become important to one another before history explains why.

Kamea's second simulator attempt lasted longer than the first.

She still broke the landing gear.

Less badly.

The instructor reset the exercise again.

"What did you change?"

"Slower approach."

"What did you forget?"

Kamea frowned at the replay.

"Crosswind correction."

"Again."

On the third attempt she reached the runway intact, bounced once, overcorrected, and nearly left the simulated pavement before stopping.

She stared at the screen.

"That counted?"

The instructor considered the question.

"You and the aircraft are both reusable. I will call that progress."

Kamea smiled.

Tamati heard about the exchange before dinner.

"My daughter has discovered Davion education."

David looked concerned.

"Has she filed a complaint?"

"Not yet. She wants another simulator period tomorrow."

"Then it may be serious."

The education negotiations became more specific after that.

Tamati did not want a generation of Aurigan students trained to believe competence required leaving the Reach permanently. Ian did not want an exchange program that quietly became recruitment by another name.

The first draft therefore included return-service expectations for certain Coalition-funded technical scholarships and explicit limits on AFFS recruiting inside designated Aurigan programs.

One Davion officer objected.

"If an exceptional student wants to join us, why prevent it?"

Hanse answered.

"We are trying to become the place they trust sending children, not the place that keeps the best ones."

"What if the student asks?"

"Then there is a process. A process is different from putting a recruiter outside the classroom."

Tamati approved that distinction.

David became involved when the technical curriculum discussion reached maintenance standards. He brought an Aurigan manual and an AFFS manual to the table and opened both to equivalent inspection procedures.

The terminology differed.

The tolerances sometimes did not.

"This is the useful part," he said.

An Aurigan delegate looked at him.

"The same answer with different words?"

"Sometimes. Sometimes different answers to the same problem. Those are even more useful if we know why."

The proposed exchange therefore included technicians, not merely officer cadets and noble students.

Machine-tool instructors.

Medical technicians.

Logistics specialists.

Aerospace maintainers.

People whose names would never appear in a diplomatic announcement and whose work would determine whether the agreement mattered ten years later.

Kamea escaped part of the afternoon meeting by joining Thomas and Ardan on a supervised tour of an engineering shop.

She asked why one BattleMech actuator assembly had three different inspection tags.

Thomas started explaining the machine.

Ardan interrupted.

"Because three different people are responsible for different ways it can kill you."

The shop chief looked offended.

"That is not technically wrong."

Kamea laughed.

Then asked the chief instead.

He showed her the structural inspection, hydraulic check, and control-system certification.

She listened seriously.

David found them twenty minutes later.

"I thought this was a tour."

Kamea pointed toward the actuator.

"It is."

Thomas said, "She asked a question."

David sighed.

"That is how it starts."

The shop chief understood immediately.

Tamati did too when they returned Kamea an hour late.

The visit ended with no grand alliance announcement.

Trade language remained under negotiation.

Education exchanges moved toward implementation.

Military cooperation stayed bounded by Aurigan invitation and sovereignty.

That was enough.

The relationship had room to grow without needing anyone to pretend they already knew what it would become.

Second Quarter

April - What Already Exists

The UrbanMech proposal reached Kara because no one else wanted it.

That was usually a sign that it would become interesting.

The Strategic Refit Center had six machines in adjacent bays.

All UM-R60s.

All old.

One extremely old.

Kara walked between them with the project engineer and a procurement officer who had already decided the meeting was an embarrassment.

The first UrbanMech stood with its right arm removed.

The second had its autocannon feed system open.

The third was little more than a torso on a maintenance frame, its legs stripped for actuator inspection.

The machines looked exactly like what critics thought they were.

Relics.

Kara stopped beside the oldest.

"How old?"

The engineer checked the plate.

"Original chassis date, 2698. Multiple rebuilds. Current engine block is later. Cockpit replaced twice. Right leg structure probably original in sections."

The procurement officer stared.

"And we are spending money on it."

Kara looked at him.

"Apparently."

The engineer handed her the proposal.

"We are not recommending production."

"Good."

"We are not recommending a speed increase."

"Better."

"We are not recommending a weapons redesign."

Kara stopped.

"What are you recommending?"

The engineer pointed down the list.

Service-life extension.

Standardized wiring harnesses.

Actuator rebuild tolerances.

Replacement seals and pumps.

Updated diagnostic ports.

Modern communications where practical.

Ammunition-feed inspection and refurbishment.

Jump-jet overhaul.

Armor replacement as required.

Selected cockpit display modernization.

One line at the bottom proposed removing an unnecessary external heat sink on standard configurations where inspection confirmed the change was safe, using the ton for additional autocannon ammunition.

Kara read it twice.

"Twenty shots."

"Yes."

"Still slow."

"Yes."

"Still an UrbanMech."

"Yes."

"Excellent."

The procurement officer looked at her as if she had betrayed him personally.

"Major, we are spending money on UrbanMechs."

"Yes."

"We have Swordsmen."

"Yes."

"Javelins."

"Yes."

"Clints are coming into wider service."

"Yes."

"Then why?"

Kara pointed at the UrbanMech in front of them.

"Because this one already exists."

The officer stopped.

Kara continued.

"A new Swordsman needs a factory slot. A new engine. Chassis. Gyro. Weapons. Armor. Acceptance. Transport."

She tapped the old armor.

"This needs parts."

The distinction reached Ian in three paragraphs.

He approved the program.

Not a new production line.

Not a resurrection announcement.

Not a heroic new variant.

A refurbishment package.

The Capellan March Militia possessed enough UrbanMechs for the work to matter.

Planetary militia inventories contained more.

Some would remain ugly.

Most would remain slow.

All would be assigned where slow was acceptable.

Industrial security.

DropPorts.

Cities.

Transport junctions.

Fixed regional defense.

The AFFS still preferred not to fight inside cities.

Refurbishing UrbanMechs did not change that.

It simply acknowledged that enemies did not always ask permission before bringing war to civilians.

The argument appeared again at the procurement review.

One officer asked why the AFFS was investing in a machine designed for the kind of battle Davion doctrine tried to avoid.

Kara answered before anyone else.

"We avoid fighting in our cities too."

The officer paused.

"That's different."

"It shouldn't be."

No one had a useful reply.

MOUT Town

The official schedule called it the Military Operations on Urban Terrain Complex.

Everyone called it MOUT Town.

Maintenance crews called it Lego City whenever they thought officers were not listening.

The name survived because the buildings did.

Not individually.

That was the point.

Wall panels could be replaced.

Windows removed.

Facades rebuilt.

Interior partitions reconfigured.

Streets blocked and reopened.

One exercise could destroy half a block on Monday and another unit could assault the same block in a different configuration on Thursday.

The construction crews had become very good at replacing walls.

They had opinions about BattleMech pilots.

A hand-painted sign outside one maintenance yard read:

BUILDING 42 REPLACEMENT PANELS AVAILABLE: 17

​

Someone had written below it:

CURRENT REQUIREMENT: 23

​

No one erased that either.

The Avalon Hussars company entering MOUT Town had twelve BattleMechs.

The defending force had an infantry battalion.

Four UrbanMechs.

Engineers.

Signals.

Local reconnaissance assets.

The Hussars read the order.

One Leftenant looked at the defender list.

"Four UrbanMechs?"

His captain looked at him.

"Yes."

"That is all?"

"And an infantry battalion."

"Yes, sir, but four UrbanMechs."

The captain understood.

He also privately agreed.

That became embarrassing forty-three minutes later.

The mission was not to destroy the defenders.

It was to open a protected corridor through the town, secure a communications building and allow a simulated civilian evacuation convoy to move along the northern route.

Ninety minutes.

Restricted fires.

No artillery against occupied civilian structures.

No indiscriminate demolition.

The defenders had forty-eight hours to prepare.

The Hussars entered cautiously.

That lasted eleven minutes.

A reconnaissance vehicle hit a simulated mine.

The road ahead became blocked.

Infantry appeared in three windows, fired simulation lasers and disappeared.

An UrbanMech stepped from behind a parking structure.

Its AC/10 fired.

The Wolverine leading the next element registered a right-torso hit.

The UrbanMech vanished.

"Contact west!"

The Wolverine turned.

The captain shouted over the net.

"Do not chase."

The Wolverine pilot stopped.

Barely.

Three infantry SRM teams were waiting in the alley he had been about to enter.

Silence.

Then:

"Understood."

The company moved again.

The defenders did not hold a line.

They held pieces of the town.

An infantry platoon fired from an upper floor, withdrew through a cut passage between adjoining buildings and appeared three blocks away before the Hussars finished clearing the first position.

A signal team used a retransmitter inside an empty shop to make a platoon headquarters appear where no platoon headquarters existed.

Engineers barricaded one road and left another conspicuously open.

The Hussars assumed the open road was trapped.

It was not.

They spent six minutes proving it.

Thirty minutes into the exercise, they had advanced eight blocks.

The Hussars had killed nothing.

They had found four empty fighting positions, two decoys, one retransmission station, several barricades and a mannequin wearing an infantry helmet.

Someone had painted SHOOT ME across the chest.

The Enforcer pilot who found it did not appreciate the humor.

The third UrbanMech fired from inside a parking garage.

Not beside it.

Inside.

Only the upper torso and autocannon appeared between reinforced columns.

The Hussars returned fire.

A modular wall panel exploded.

The UrbanMech stepped backward.

Gone.

The company captain stared at the structure.

"That is disgusting."

His gunner answered.

"Yes, sir."

"I meant professionally."

"I assumed so."

The UrbanMechs did not win.

That mattered.

When the Hussars stopped reacting to them and started solving the mission, the defenders' advantage shrank quickly.

Two BattleMechs maintained overwatch.

Attached infantry began clearing only those structures that actually threatened the route.

Bypassed positions were isolated instead of assaulted.

Engineers moved under protection rather than after the fight.

The Hussars refused bait.

An UrbanMech appeared.

Fired.

Withdrew.

No one followed.

The machine suddenly became much less dangerous.

The Hussars eventually trapped one near the freight terminal and killed it in simulation.

The company channel celebrated.

The captain did not.

Something bothered him.

"Where is the infantry?"

No one answered.

Then the tactical display updated.

The communications center the Hussars had cleared twenty minutes earlier was contested again.

The defenders had moved through utility tunnels, service alleys and pre-cut passages while three Hussar lances concentrated on the UrbanMech.

The captain understood.

"We've been had."

The final thirty minutes were excellent.

They were also too late.

The Hussars retook the communications center.

Opened most of the route.

Bypassed unnecessary resistance.

Then discovered that defending engineers had dropped a modular building across the northern highway.

Estimated clearance time:

Seventeen minutes.

Exercise time remaining:

Nine.

MISSION FAILURE.

​

The after-action review began forty minutes later.

The Hussars sat on one side of the room.

The defending infantry battalion sat on the other.

Four UrbanMech pilots occupied the front row.

One had a bandage over his forehead after forgetting that the simulator cockpit hatch was lower than the real one.

No one mentioned it.

The senior CTC instructor stood beside the tactical replay.

"Captain."

"Yes, sir."

"How many UrbanMechs did you destroy?"

"One."

"How many did you engage?"

"All four."

"How many were your mission?"

The captain stared at the map.

"None."

The instructor nodded.

"Good."

He highlighted the evacuation route.

"What was your mission?"

"Open and secure this corridor."

"Did the UrbanMechs prevent that?"

The captain started to answer.

Stopped.

"No, sir."

The instructor pointed to the infantry battalion commander.

"He did."

The infantry major said nothing.

He did not need to.

The instructor continued.

"The UrbanMechs existed to make you look at them."

One of the Hussar lieutenants muttered something.

The instructor heard him.

"Speak."

The lieutenant hesitated.

"They were very good at it, sir."

A few people laughed.

The instructor did not.

"That is their job."

He changed the display.

Each UrbanMech position appeared.

Garage.

Warehouse.

Transit yard.

Industrial alley.

Courtyard.

Parking structure.

None had remained in place long.

Then the infantry routes appeared.

Companies flowed around the Hussar advance like water.

Then engineers.

Signals.

Observation posts.

Obstacles.

The four UrbanMechs suddenly looked very small.

"The BattleMechs were the least important part of the defensive system."

That irritated several BattleMech pilots.

Good.

The instructor faced the room.

"MOUT training does not exist because the AFFS wants to fight through cities."

He pointed toward the town beyond the windows.

"Our preference remains bypass. Isolation. Evacuation where practicable. Reduction without turning the civilian population into terrain."

Then he pointed toward the Hussars.

"But sometimes the enemy chooses."

Then the defenders.

"And sometimes you are defending the city someone else chose."

The surviving UrbanMech lance commander sat quietly.

The Hussar captain looked toward him.

"I spent half the morning trying to kill you."

"Yes, sir."

"You fired at me twice."

"Yes, sir."

"What were you doing the rest of the time?"

"Making sure you knew where I was."

The captain stared at him.

Then laughed.

"That may be worse."

The infantry battalion commander finally spoke.

"It usually is."

The AAR went into the expanding MOUT curriculum.

The refurbishment order continued.

No one proposed that UrbanMechs become fashionable.

That would have ruined them.

Andoran - The Parts Nobody Cares About

The Clint line produced its own argument.

The machine was not a Swordsman.

Everyone agreed.

That agreement became necessary because procurement officers kept trying to make it one.

Both were forty-ton BattleMechs.

Both were intended to become common enough that militia and second-line support mattered.

That was where the similarity ended.

The Clint moved.

Six jump jets.

Speed.

An AC/5.

Twin medium lasers.

A scout that could fight.

A cavalry machine that could find trouble, irritate it, force it to deploy and leave before the answer became too large.

The Swordsman did something else.

More deliberate.

More general.

No jump jets.

A militia machine designed to stay useful across a wider range of ordinary jobs.

The industrial commonality team did not touch the roles.

It touched everything boring.

Actuator bearings.

Armor attachment standards.

Cockpit electronics where possible.

Communications modules.

Environmental components.

Connectors.

Fasteners.

Cable specifications.

Diagnostic fittings.

Maintenance tools.

Access-panel standards.

Myomer procurement grades where engineering allowed.

A young engineer complained.

"We aren't improving the Clint."

The senior tooling specialist looked at him.

"We are."

"We haven't changed a weapon."

"No."

"Engine?"

"No."

"Armor?"

"No."

"Jump system?"

"No."

"Then what changed?"

The specialist held up a connector.

"This."

The younger man stared.

"That is a plug."

"Yes."

"It doesn't matter."

The specialist smiled.

"Exactly."

The report that eventually reached New Avalon carried a handwritten note copied from Andrew-era production guidance:

DO NOT TURN THE CLINT INTO A SWORDSMAN. MAKE THE PARTS NOBODY CARES ABOUT INTERCHANGEABLE.

​

Kara approved it without amendment.

The logic spread farther than the two BattleMechs.

If a connector used on a Clint could also be stocked for a Swordsman, the warehouse needed one bin instead of two.

If the same diagnostic adapter worked on both, a militia technical school needed one training block instead of two.

If cockpit environmental components shared service standards, a planetary depot could support machines from different manufacturers without keeping separate expertise for every trivial subsystem.

No one claimed complete interchangeability.

That would have been another kind of stupidity.

Some systems were unique because engineering required them to be.

The reform goal was not sameness.

It was to stop making uniqueness where uniqueness bought nothing.

Blue Water

By 3010 the Federated Suns did not need to discover carrier aviation.

The Sentinel program had already taught Black Navy crews how much work lived behind a fighter sortie.

Pensacola had been training pilots, aircraft handlers, maintainers, ordnance crews, fuel specialists, flight-control personnel, rescue teams and damage-control sailors for years.

Patrol Space Stations had added another layer.

The Essex program did not begin from ignorance.

It began from a different question.

How did a carrier operate as the center of a task force?

FSS Essex had entered the Avalon Sea the previous year unfinished and honest about it. In 3010 the carrier remained in fitting-out and trials long enough for every department aboard to discover at least one thing the drawings had underestimated.

The Air Group came aboard in stages.

Ashore qualification had ended with an admission that the wing was formed but not yet carrier qualified.

The sea fixed that.

The first recovery cycle proved that painted deck geometry on an airfield had not taught pilots what a moving ship felt like beneath them.

The first rough-weather cycle proved that a one-hundred-thousand-ton ship could still move enough to make people respect the ocean.

The first sustained week of operations proved that salt found machinery with religious devotion.

None of those lessons surprised anyone who had served aboard a wet-navy ship.

They did surprise several Sentinel veterans.

Commander Alana Mercer watched one Black Navy aviation officer stare at a row of aircraft secured on the flight deck before an evening cycle.

"You're leaving them there?" he asked.

"For forty minutes."

He looked at the ocean.

"Outside."

"That is traditionally where the flight deck is."

"In weather."

"Yes."

"On a surface that moves under them."

Mercer smiled.

"You're learning."

The officer stared at the tie-down crews.

"You people are insane."

"You recover fighters into vacuum."

"Vacuum is predictable."

A wave struck the hull hard enough to send spray above the bow.

Mercer looked toward him.

"Point conceded."

The argument lasted exactly as long as it needed to.

They tested tie-down patterns.

Deck-park limits by weather state.

Fueling rules under changing wind.

When to strike aircraft below.

How many fighters could remain spotted without turning the deck into a casualty trap.

The answer changed with mission and weather.

That was normal.

Different Air

The distinction between Blue Navy and Black Navy aviation became harder to ignore once both communities began using the same carrier deck.

Neither side was inexperienced.

That was what made the first arguments useful.

Black Navy aviators came from Sentinels, Patrol Space Stations, DropShips and planetary aerospace commands. They thought in vectors, fuel fractions, intercept volumes and three-dimensional battles whose upper boundary was not an atmosphere but whatever the mission required. They were accustomed to fighters that could leave a world entirely, cross from atmosphere into space and return from directions that made the word altitude feel provincial.

Blue Navy aviators thought differently.

They thought in decks.

Runways.

Wind.

Weather.

Sea state.

Terrain.

The number of aircraft that could be armed, fueled, moved, launched, recovered and made ready again before the enemy returned.

Captain Alana Mercer discovered the difference during the first joint planning conference aboard Essex.

Her counterpart from the Sentinel force, Captain Tomas Havel, had commanded aerospace fighters from ships for most of his career. He had survived launch cycles under thrust, recovery operations after battle damage, vacuum emergencies and one incident involving a disabled fighter that had reached its carrier with so little fuel remaining that the maintenance chief later accused the pilot of landing on fumes and optimism.

Mercer respected him immediately.

Havel respected her approximately fifteen minutes later.

The delay came from a diagram.

Mercer had drawn the next exercise as a layered atmospheric defense. Conventional fighters would operate lower, using ship and shore controllers. Aerospace fighters would hold higher and farther out. A Sentinel control detachment would maintain the broader picture. Essex would coordinate recovery and re-tasking. Surface escorts would provide sensor coverage and simulated missile support.

Havel studied it.

"Why are your conventional fighters that low?"

Mercer looked at him.

"Because they are conventional fighters."

"I know what they are."

"Then you know why."

Havel pointed at the upper intercept lanes.

"You are surrendering altitude before the fight begins."

Mercer pointed toward the coast.

"I am using the atmosphere before the fight begins."

That started it.

Havel argued that altitude was stored energy and that giving it away narrowed the conventional fighters' options. Mercer agreed.

Mercer argued that terrain, weather and surface sensor support were also resources, and that sending conventional fighters high enough to fight aerospace machines on aerospace terms merely made them easier to kill. Havel agreed with that too.

Neither enjoyed discovering agreement inside the argument.

The dispute moved to launch cycles.

Havel wanted the aerospace fighters airborne earlier.

Mercer wanted a portion held on deck.

"Why keep combat power tied down?" he asked.

"Because if I launch everything now, it is all using fuel now."

"They can recover."

"Into what cycle?"

Havel paused.

Mercer brought up the deck plan.

Four aircraft returning with low fuel.

Six preparing to launch.

Two down for maintenance.

A rescue VTOL on alert.

A conventional strike section waiting for ordnance.

"Black Navy thinks first about getting the fighter to the fight," Mercer said. "Blue Navy has to think about getting the fighter back onto the deck while the next fighter is trying to leave it."

Havel looked at the diagram.

"That is unfair."

"The ocean has never shown much interest in fairness."

Havel smiled.

"Neither has vacuum."

That became the beginning rather than the end of the disagreement.

The two CAGs stopped trying to prove that one environment was harder.

Both were hard in different directions.

Black Navy operations punished bad geometry, bad fuel planning and the assumption that a carrier could hide behind a screen in three-dimensional space.

Blue Navy operations punished bad timing, bad deck management, weather complacency and the assumption that an aircraft could always land somewhere else if the deck became unavailable.

Different did not mean better.

Different did not mean worse.

It meant the mistake moved.

The first mixed exercise demonstrated that with enthusiasm.

The aerospace fighters won.

Decisively.

The conventional fighter squadrons had been given an air-defense mission against a Black Navy aggressor package. Their pilots knew the aerospace machines had more performance available and tried to compensate with courage.

That was almost always expensive.

The conventional fighters climbed to meet them.

The aerospace aggressors kept the altitude advantage, accelerated through engagements the conventional pilots wanted to prolong, converted speed into position and repeatedly forced the defenders to choose between breaking formation or accepting attacks from above.

Exercise Control stopped counting individual simulated kills after the result became obvious.

The conventional pilots landed angry.

Mercer let them remain angry until the debrief.

Then she displayed the engagement.

"What did we prove?"

One pilot answered immediately.

"That aerospace fighters are better."

Havel, sitting against the wall, raised an eyebrow.

Mercer shook her head.

"We already knew that."

The room quieted.

She replayed the first intercept.

"You climbed thirty thousand feet to meet aircraft with better thrust and better high-altitude performance. Why?"

"They were coming through our sector."

"Was your sector thirty thousand feet high?"

The pilot stopped.

Mercer changed the display.

The coastline appeared.

Cloud layers.

Surface radar coverage.

Two valleys extending inland from the sea.

A civilian shipping lane the exercise had marked as restricted.

"You abandoned every advantage you had so you could challenge them where they were strongest."

Another pilot looked toward Havel.

"So what are we supposed to do? Hide?"

Havel answered before Mercer could.

"Yes."

Several pilots stared at him.

He shrugged.

"Until hiding becomes useful."

That changed the room.

The Black Navy CAG walked to the display.

"An aerospace fighter can go places you cannot. It can climb faster, leave the atmosphere, attack from more geometries and usually carry the fight into regimes where your machines become progressively less comfortable."

He pointed toward the low-altitude lanes.

"So stop volunteering to follow it there."

Mercer folded her arms.

"Thank you."

Havel ignored her.

"Make us come where you are. Make us descend. Make us spend energy. Make us look down through weather and terrain. Make us worry about ground sensors, ship sensors and another section we cannot see. If you cannot win the aircraft comparison, change the comparison."

One of the conventional pilots looked skeptical.

"And if they refuse?"

Mercer answered.

"Then they are not stopping your mission."

That was the part several pilots had forgotten.

The next exercise was not an air-superiority contest.

The conventional fighters had a strike mission.

The aerospace aggressors had to stop it.

That changed everything.

The conventional strike group launched low and remained low. Surface controllers used escort sensor tracks to keep them away from the first aerospace sweep. Two conventional sections appeared deliberately on another axis, loud enough on sensors to look like the main body.

The Black Navy aggressors saw them.

They turned.

The decoy sections did not fight for very long.

They did not need to.

They broke toward the coast, split beneath a cloud layer and forced the aggressors either to pursue into lower altitude or abandon them.

One aerospace pair stayed high.

Correct.

The second descended.

Also correct.

The problem was that the actual strike package was twenty kilometers farther south.

The aerospace high cover found it late.

Aerospace fighters from Essex met them before the intercept developed cleanly.

For several minutes the sky became exactly what the exercise designers wanted: aerospace fighters fighting aerospace fighters high enough to exploit their strengths while conventional fighters remained lower, fast enough for their own mission and protected by geometry rather than fantasy.

The conventional strike section reached weapons release.

Exercise Control credited the strike.

The conventional pilots had not defeated the aerospace force.

They had accomplished the mission anyway.

That mattered more.

The third exercise produced the first simulated conventional-fighter kill against an aerospace machine.

No one treated it as proof of equality.

That would have insulted everyone involved.

A conventional section used coastal terrain and a surface controller to remain below the aggressor's best sensor picture. Another section appeared first and broke away under pressure. The aerospace pilot followed long enough to convert altitude into speed and speed into confidence.

The hidden section emerged from a different valley.

Two attack solutions appeared almost together.

The first missed.

The second did not.

Exercise Control called the aerospace fighter killed.

The conventional pilots cheered on the radio.

Mercer stopped that during the debrief.

"What did you prove?"

The section leader had learned from the previous week.

"That four of us with ground control and a prepared geometry can make one of them have a bad afternoon."

Havel nodded.

"Better."

The pilot looked toward him.

"Not that we are equal."

"You are not."

The answer was immediate.

No one in the room flinched.

Havel continued.

"That is not an insult. My fighter can go to orbit. Yours cannot. Mine can exploit a larger performance envelope. Yours costs less, requires less infrastructure, can be produced on worlds that cannot support mine and can put a competent pilot over a battlefield where otherwise there might be nothing."

He pointed toward the kill geometry.

"Outmatched does not mean helpless."

The sentence stayed.

More importantly, the pilots believed it because no one had lied to them first.

The improvement that followed was not dramatic enough for recruiting posters.

Conventional pilots stopped measuring themselves by whether they could beat aerospace pilots in symmetrical fights.

They became better at choosing asymmetrical ones.

They learned terrain masking.

Ground-controlled intercept.

Short exposure windows.

Mutual support.

Breaking contact before an aerospace opponent could turn superior performance into a second attack.

They learned to use numbers without pretending numbers erased quality.

Two sections could cover one another.

One could bait while another positioned.

A damaged conventional fighter could leave the fight while another took its place without forcing the entire defense plan to revolve around one expensive machine.

The aerospace pilots learned too.

They stopped describing conventional fighters as targets with wings.

A slow or lower-performance threat still occupied sensors, fuel, weapons and attention.

A conventional strike fighter ignored because it was not glamorous could still destroy a radar site, artillery position, bridge, fuel dump or exposed headquarters.

A conventional interceptor operating beneath a larger sensor network could still force an aerospace pilot to maneuver at an inconvenient moment.

And a conventional pilot who knew the local terrain could be extremely difficult to surprise in the part of the atmosphere that pilot flew every day.

The mixed air group changed its language.

No one called the conventional wings lesser in official planning.

They were lower-performance aircraft in some regimes.

That was a technical description.

It was not a mission description.

Short Fields

The unexpected transfer went ashore.

Carrier operations had been justified partly because the Federated Suns needed to train people to operate fighters from ships.

By summer, engineers realized the carrier was also teaching them how little runway they actually needed if they were willing to organize the ground correctly.

A carrier flight deck was an absurdly constrained airfield.

It had limited length.

Limited parking.

No spare room for indecision.

Fuel and weapons moved through tightly controlled routes.

Aircraft launched at precisely calculated weights.

Recovery equipment had to work every time.

Crash crews had to reach any point quickly.

Maintenance had to know exactly which aircraft could fly and which merely looked flyable.

Someone at Pensacola asked an impolite question.

Why did an expeditionary airfield need to be much larger?

The first answer was that land was not a ship.

The second answer was that this did not actually answer the question.

The resulting trials used an auxiliary naval field near the coast.

Engineers marked off progressively shorter usable runway lengths and treated everything outside the marked section as damaged, mined or unavailable.

They brought portable arresting equipment.

Mobile lighting.

Crash vehicles.

Fuel bladders and pumping modules.

Ordnance carts.

Maintenance shelters.

Communications equipment.

Then they applied carrier deck-control procedures to the ground.

Yellow-shirt directors became marshaling teams.

Deck spotting became ramp spotting.

Launch queues became runway queues.

Aircraft that did not need deep maintenance stayed near the launch point.

Aircraft that did moved out of the cycle.

Nobody parked a healthy fighter in the one location every returning aircraft needed merely because the squadron commander liked symmetry.

The first short-field cycle was ugly.

Good.

A conventional fighter aborted because its launch weight and the available distance did not agree.

Another recovered into the portable arresting system after a brake warning.

A third taxied into the wrong servicing lane and blocked two aircraft until a chief threatened to have the pilot assigned permanently to ground handling.

The second cycle was better.

By the fourth, the field was launching useful combat loads from a runway section that would previously have been classified as marginal for sustained fighter operations.

No one claimed the carrier had invented short-field aviation.

It had forced the Navy to become disciplined enough at constrained operations that old short-field ideas became easier to use.

The conventional squadrons benefited first.

That made sense.

Many planetary militias already operated from imperfect fields. Outback worlds maintained strips that would never resemble New Avalon. Regional defense plans assumed runways could be damaged during the opening hours of an attack.

Carrier-derived ground procedures gave those units a better answer than waiting for engineers to restore a full-length runway.

Operate from what remained.

Repair while flying.

Move the servicing footprint if the enemy found it.

Keep aircraft dispersed until the launch cycle required concentration.

The aerospace community became interested for a different reason.

Aerospace fighters did not suddenly become conventional aircraft.

They did not need to.

But planetary aerospace operations still required fuel, weapons, maintenance, rescue, communications and places where people could safely put machines between sorties.

If a full aerospace base was damaged or predictable, a smaller expeditionary site could keep part of the wing operating.

The Black Navy officers recognized the implication quickly.

A fighter that could operate from more places was harder to kill on the ground.

A carrier that was the only place its fighters could recover was a vulnerability.

A carrier whose fighters could divert to several austere fields possessed options.

Options accumulated.

Havel visited the short-field trial on the third week and found Mercer standing beside a conventional fighter while ground crews rearmed it under camouflage netting.

He looked at the shortened strip.

"You turned a carrier deck into a road base."

Mercer shook her head.

"We turned carrier habits into airfield habits."

"That sounds less interesting."

"It is also more accurate."

They watched another conventional fighter launch.

It used nearly all the available distance.

No one watching pretended the aircraft possessed aerospace performance.

It climbed anyway.

Havel followed it with his eyes.

"How many sorties today?"

"Thirty-eight by noon."

"From this?"

Mercer looked around at the tents, fuel modules, portable arresting gear and maintenance shelters.

"This is an airfield."

"This is a collection of trucks that happens to have pavement."

"That is what an expeditionary airfield is before the engineers become offended."

Havel laughed.

Then an aerospace fighter came across the coast at low altitude and entered the pattern.

It had been operating as high cover for the conventional strike cycle.

The controller placed it behind two conventional recoveries rather than giving it automatic priority.

Havel noticed.

"Your controller just made an aerospace fighter wait."

"Its fuel state is good."

"The conventional aircraft are less valuable."

Mercer looked at him.

Havel corrected himself before she could.

"Wrong word."

"Yes."

He checked the recovery board.

The first conventional fighter had battle-damage simulation.

The second was near its fuel floor.

The aerospace fighter was healthy.

Havel nodded.

"Priority by condition, not prestige."

"Now you are learning Blue Navy."

"I thought I was learning airfield operations."

"Same disease."

He smiled.

"Different environment."

The mixed-cycle trials produced another change nobody had formally ordered.

Conventional and aerospace squadrons began borrowing tactics from one another.

Aerospace pilots taught energy management at regimes conventional pilots could actually use rather than lecturing them about performance they did not possess.

Conventional pilots taught low-level navigation, weather reading and terrain exploitation to aerospace pilots who had spent much of their careers thinking of the atmosphere as the inconvenient part between surface and orbit.

Carrier controllers built common recovery language.

Maintenance teams standardized handoff reports.

Ordnance crews learned which procedures truly had to differ and which had remained different only because separate communities had written separate manuals.

The distinction between conventional and aerospace did not disappear.

It became useful instead of hierarchical.

The Black Navy still owned black space.

The Blue Navy still understood an ocean in ways no Sentinel crew needed to.

Aerospace fighters remained the higher-performance, more flexible combat machines.

Conventional fighters remained cheaper, more numerous and easier for many worlds to operate.

The lesson was not that one could replace the other.

It was that a commander who possessed both had more choices than a commander who insisted only one counted.

By autumn, mixed air plans stopped listing conventional fighters as the force used until aerospace arrived.

Sometimes the conventional wing was the primary atmospheric force while aerospace remained high cover.

Sometimes aerospace fighters made the first interception and conventional aircraft protected recovery lanes or struck ground targets.

Sometimes the conventional fighters provided the visible threat while aerospace aircraft waited for the enemy to react.

Sometimes the aerospace wing left the atmosphere entirely and the conventional squadrons simply continued the planetary mission without them.

That was interoperability.

Not pretending the aircraft were the same.

Building plans that became stronger because they were not.

The larger work happened beyond Essex.

Rapier destroyers and Longsword cruisers joined the exercises because a carrier without escorts only taught one ship how to be busy.

The Buckler program added the missing mobility.

A Buckler-class deployable light carrier could be transported between worlds aboard specially modified Mammoth DropShips, then placed into a planetary ocean and paired with Rapier and Longsword escorts.

The concept imposed discipline on the design.

Every request for another feature met the same question.

Does it still fit inside the interstellar transportation envelope?

If the answer became no, the requirement had to justify destroying the reason the ship existed.

Most did not.

The Buckler was not a smaller Essex.

It was a carrier package that could exist on a world where no carrier group had existed the month before.

A modified Mammoth arrived.

Another brought support.

The carrier deployed.

Escorts deployed.

Aircraft, ammunition, spares, maintenance equipment and shore support followed.

The Navy could establish a useful blue-water aviation presence on an ocean world without first building the shipyard that would have been required to create one locally.

That made the Buckler as much a logistics experiment as a naval one.

The first major 3010 fleet problems deliberately mixed the types.

Essex represented the large permanent carrier.

A Buckler group represented expeditionary reinforcement.

Rapiers screened.

Longswords carried heavier command and surface-combat burdens.

The question was not whether any one ship could defeat an enemy.

The question was whether the group could keep the carrier useful while accomplishing the mission.

Fleet Problem One kept the escorts close.

Everyone felt protected.

The aggressor found everyone at once.

Fleet Problem Two spread the screen.

A Rapier became isolated and was killed in simulation before support could reach it.

Fleet Problem Three pushed pickets farther outward.

The pickets saw the threat early.

Communications and command delay meant Essex did not understand what they had seen until too late.

Fleet Problem Four went beautifully.

The carrier air group destroyed the first major raid.

Pilots celebrated.

Controllers celebrated.

The aggressor staff launched the second raid.

Half the combat air patrol was rearming.

Too many fighters were low on fuel.

The reserve was too small.

Exercise Control announced simulated missile hits against Essex.

The AAR was unpleasant.

A squadron commander defended the first interception.

"We destroyed the raid."

The task-force commander nodded.

"Yes."

"We achieved favorable exchange."

"Yes."

"We lost almost nothing."

"Yes."

The squadron commander waited.

The admiral changed the display.

Second raid.

Essex.

Simulated damage.

"Congratulations."

No one smiled.

"The second one sank the carrier."

The lesson traveled quickly.

A carrier task force did not exist to win the first fight.

The carrier did not exist to prove it could defend itself.

An escort did not exist to remain physically close enough to reassure an admiral.

A fighter committed to one mission could not perform another.

The screen existed to provide information and time.

The carrier's air wing was not infinite merely because a hundred aircraft looked impressive on paper.

A damaged carrier that could not recover aircraft might be mission-killed even if it still floated.

And the carrier commander could not personally control every fighter, destroyer, cruiser, scout, tanker and rescue asset.

Intent had to move downward.

Information had to move upward.

The fleet had discovered the same problem David was fighting inside the B-3000.

Different environment.

Same disease.

Too much information.

Too much temptation to control.

The Navy wrote its answer differently.

THE CARRIER IS THE CENTER OF THE FORCE. IT IS NOT THE WHOLE FORCE.

​

David read the exercise summary.

Then sent it to the B-3000 team.

No explanation.

They understood.

The B-3000 - The Human Filter

By May, thirty days had become eighty-two.

Kara sent David a message containing only:

NINETY.

​

He did not answer.

The B-3000 prototype architecture now separated information into layers.

Not secrecy.

Responsibility.

A company commander received what the company needed.

Battalion received what crossed company boundaries.

Regiment received what crossed battalion boundaries.

RCT headquarters received exceptions, patterns and decisions requiring RCT resources.

A damaged ammunition truck did not need the commanding general.

A battalion whose ammunition distribution was collapsing did.

A broken bridge did not.

Three broken bridges changing the operational axis did.

The hardest part was not deciding what information mattered.

It was allowing lower headquarters to handle what did not.

David sat beside an engineer during one simulation.

A battalion fuel problem appeared.

The system marked it amber.

David opened the detail.

Then stopped.

The battalion logistics cell was already working it.

He closed the window.

The engineer noticed.

"You could have watched."

"I know."

"Why didn't you?"

David looked at him.

"Because they were doing their job."

The engineer smiled.

The next alert went red.

Entire regiment fuel forecast compromised.

David opened it.

That was his job.

The B-3000 became less a larger command computer than an architecture for deciding where attention belonged.

That was why the project took longer than thirty days.

The field architecture also forced the AFFS to confront a different problem.

Faster information could create faster micromanagement.

One exercise deliberately gave a brigade commander near-perfect awareness of three subordinate battalions.

He began issuing corrections at company level within fourteen minutes.

The first orders were good.

That made the mistake harder to see.

A company adjusted its route exactly as directed.

A second shifted reserve posture.

A third changed an artillery request.

All three actions improved the immediate local picture.

Then the brigade commander discovered he had spent twenty minutes commanding companies while the opposing force changed the brigade problem.

The exercise ended badly.

At the AAR, he defended the orders.

"They were correct."

David nodded.

"Mostly."

"Then what was the error?"

"You were correct at the wrong echelon."

The brigadier frowned.

David pointed toward the map.

"You improved three companies and lost the brigade."

That line entered the course material.

Controlled aggression required commanders at every level to understand two directions at once.

What the higher commander intended.

What the subordinate was entitled to decide.

A command system that erased either relationship made the army worse.

The best version of the B-3000 would not make orders unnecessary.

It would make unnecessary orders more obvious.

The breakthrough came from a sergeant, which annoyed exactly no one who had been paying attention for the previous twenty years.

The development team had spent a week trying to define the perfect automated priority rules for RCT-level reporting. Every version worked until the exercise staff created a situation the rules had not anticipated.

A green maintenance report became strategically important because it belonged to the only recovery company near a collapsing axis.

A red communications fault became almost irrelevant because the battalion had already shifted to a redundant net.

An amber medical report mattered enormously because the evacuation route crossed terrain the enemy was about to isolate.

The computer could classify conditions.

It could not always classify meaning.

Master Sergeant Inez Corbett, assigned from a headquarters signals section, listened to the engineers argue for twenty minutes.

Then she asked, "Why are you trying to make the machine be the staff?"

The room stopped.

One engineer said, "We're trying to prioritize information."

"So do that."

"That is what we're doing."

Corbett pointed toward the display.

"No. You're trying to decide what the commander should care about. Let his staff do that. Make the machine help us sort."

David leaned back.

"Human filter."

Corbett nodded.

"Sir, if you automate judgment, the first unusual war will make you stupid faster."

The phrase was impolite enough to survive.

The next architecture added staff cells between raw exception reporting and senior-command display. Logistics saw logistics patterns before the commander did. Medical staff saw casualty and evacuation problems. Engineers saw mobility. Intelligence saw collection gaps. Operations saw the combined effect.

The commanding general received fewer alerts.

They were better alerts.

In the next exercise, the system generated more than six hundred subordinate exceptions in four hours.

The commanding general saw twenty-eight.

He acted on eleven.

Afterward the engineers asked whether that meant the other five hundred seventy-two reports had been unnecessary.

Corbett answered before David could.

"No. It means someone else did their job."

That became another line in the training material.

The B-3000 was not becoming less capable because it hid detail from the commanding general.

It was becoming more capable because the organization around it learned where detail belonged.

David recognized the connection to IRON MIRROR immediately.

The exercise had taught him that seeing Jasmine's BattleMechs did not entitle him to command them.

The B-3000 taught the same lesson electronically.

Visibility was not ownership.

Luthien - A Different Question

The first working B-250 and B-500 systems did not reach the Draconis Combine through one dramatic intelligence coup.

Pieces arrived.

A damaged system recovered from a raid.

Components purchased through intermediaries.

Technical descriptions assembled from reports.

Enough eventually existed for the ISF to place a functioning lower-tier Davion command package in front of engineers who had been told not to admire it.

They did not.

They dissected it.

The architecture was hierarchical.

That made sense.

Company.

Battalion.

Regiment.

Information moved upward.

Orders moved down.

One engineer studying the targeting-data interface frowned.

"Why does this have to climb?"

The intelligence officer beside him looked over.

"What?"

"A target report."

He pointed at the data path.

"If this lance sees a target and that lance can shoot it, why should the useful part wait for battalion to understand the whole report?"

"Because command decides."

"Command decides what to do. That does not mean every useful piece of targeting information must climb the same ladder."

The officer stared at the schematic.

The engineer continued.

"Davion is trying to make headquarters understand more."

"Yes."

"Perhaps we should make warriors share faster."

The comment did not become a weapon system in 3010.

It became a question.

Questions lasted.

Sian - The Visible Answer

The Capellan Confederation obtained less complete B-series material and more contradictory intelligence.

That produced a different reaction.

Several planners concentrated on the hardware.

Command vehicles.

Computers.

Displays.

Communications.

If Davion advantage came from machines, machines could be copied.

Maskirovka analysts were less certain.

The reports from Kittery, New Aragon, Cylene and the expanding militia training system kept describing behavior that hardware did not explain by itself.

Units withdrew when they were supposed to.

Militia officers used readiness figures as arguments for missions rather than excuses to avoid them.

Technicians returned older equipment to reliable service.

Staffs filtered information instead of forwarding everything upward.

A young analyst summarized the problem badly and therefore usefully.

"They are making people act like the manuals mean it."

His supervisor looked at him.

"Rewrite that."

"Yes, sir."

The final version said:

DAVION IMPROVEMENTS APPEAR INCREASINGLY INSTITUTIONAL RATHER THAN PLATFORM-SPECIFIC.

​

It was less memorable.

It was also more dangerous.

ComStar - Reconstruction

ROM's June review began with a deliberately primitive question.

What existed in 3010 that did not exist, or did not function at comparable scale, in 2989?

The answer required more files than expected.

Schools.

Instructor pipelines.

Technical certification.

Strategic Refit Centers.

Militia standardization.

Industrial automation.

Expanded ammunition production.

Restored BattleMech lines.

New vehicle lines.

Expanded transport.

Harbor and recharge infrastructure.

Mobile maintenance.

Combat Training Centers.

Professional OPFOR.

Naval aviation schools.

Sentinels.

Blue Water carrier groups.

Mercenary administration.

Command systems.

NCO development.

Regional training.

March Militia rotations.

Marine formations.

Dragonlords.

Borderers.

The analyst stopped.

Then added:

Institutional capacity to create additional institutional capacity.

That phrase caused trouble.

It was too vague.

It was also correct.

The report went upward.

A senior ROM officer asked:

"Who is directing this?"

The analyst gave several names.

Ian Davion.

Hanse Davion.

David Davion.

Field Marshal Hall.

Industrial leaders.

Academy heads.

The officer frowned.

"No."

The analyst stopped.

"What do you mean, no?"

"Who replaced Andrew?"

The analyst looked at the timeline.

Andrew's death.

2999.

Acceleration afterward.

Programs continued.

Some expanded.

Others mutated into things Andrew had never planned.

"No one."

The officer stared at him.

That answer went higher.

Third Quarter

Point Barrow - The Second Dragonlords

The 2nd Dragonlords arrived at Point Barrow expecting training.

They received humiliation.

The distinction mattered.

Training was something a unit could imagine itself completing. Humiliation arrived when the unit discovered that the things it believed about itself were not yet true.

The RCT had reasons for confidence. It was fully equipped. Its personnel books were healthy. Its BattleMech formations contained good pilots. Its armor regiments had crews who knew their vehicles. Infantry battalions could march, shoot, dig, ride, and complain with professional enthusiasm. The artillery knew how to put shells where observers asked. The aerospace regiment could fly. The engineers could build and destroy. The logisticians had spent two years proving that Lindsay was not merely a parking place for a large formation.

None of those statements meant the 2nd Dragonlords could yet do all of those things together under pressure.

Point Barrow existed to make that problem impossible to hide.

The Combat Training Center had two permanent Professional OPFOR commands. They did not possess identical tables of organization because their job was not to be identical. One was fast, reconnaissance-heavy, and fond of making the training unit chase problems that were not important. The other carried more armor, infantry, engineers and artillery and preferred to make a formation commit to an answer before changing the question.

Between them, they had spent years fighting almost every kind of AFFS formation that rotated through the ranges.

They knew how Guards units reacted to attacks on prestige targets.

They knew which militia headquarters became cautious when communications failed.

They knew how young RCT staffs overused aerospace reconnaissance when they stopped trusting ground scouts.

They knew how commanders who had read too many exercise reports tried to avoid the last unit's mistake and walked directly into the next one.

The 2nd Dragonlords were simply the newest problem.

Their commanding general met the two OPFOR commanders during the opening conference.

Neither gave him useful comfort.

"How hard are you planning to hit us?" he asked.

The senior aggressor commander considered the question.

"As hard as you let us."

The second commander added, "And sometimes harder, because Exercise Control can give us ammunition you did not know we had."

The Dragonlord general looked at Hall, who had come to Point Barrow for the opening week.

"That seems unfair."

Hall nodded.

"Yes."

The general waited.

Hall did not add anything.

"That is the lesson?"

"Part of it."

"And the rest?"

Hall looked toward the range map.

"War is usually unfair before it becomes lethal."

The First Beating

The first exercise had a simple name because Training Command believed elaborate names encouraged officers to look for elaborate tricks.

ROAD MARCH ONE.

Advance seventy-two kilometers.

Seize two river crossings.

Establish a lodgment around Objective Green.

Hold for twelve hours.

The RCT commander read the order and smiled faintly.

"They are trying to insult us."

His chief of staff shook his head.

"Sir, after the opening brief, I think they are trying to make us insult ourselves."

That turned out to be accurate.

The Dragonlords moved fast.

Too fast.

The lead BattleMech regiment crossed the start line before sunrise and made excellent time through the first forty kilometers. Its scouts found the obvious OPFOR screen, pushed it aside, and reported the road network largely clear.

The BattleMechs kept moving.

Armor followed.

Infantry followed the armor.

Artillery displaced.

The headquarters saw green icons moving in the right direction and liked the picture.

The support columns saw something else.

One heavy recovery company was still waiting for a bridge classification on the western route.

A fuel battalion had lost twenty-seven minutes because two movement tables assigned priority to different convoys at the same junction.

One artillery battalion had moved its guns but not all of its ammunition trucks because a regimental staff assumed the brigade logistics net had confirmed the route.

The brigade logistics net assumed the regiment had.

Nobody had lied.

That was the problem.

The first OPFOR raid arrived at 0937.

It ignored the BattleMechs.

Six fast aggressor machines appeared on a ridge far behind the lead regiment, fired simulation weapons into the fuel distribution point, and vanished before the nearest Dragonlord company could reverse direction.

At 0951, an OPFOR aerospace flight made a low-level pass against a maintenance park that had not displaced because its assigned road remained occupied by infantry transports.

At 1008, engineers reported that the eastern crossing could not support two of the heavy transporters the movement plan assumed would use it.

At 1012, an armor brigade requested permission to change routes.

At 1013, the same brigade reported contact on its new route.

At 1018, the lead BattleMech regiment asked where its artillery was.

At 1020, the artillery asked where its ammunition was.

At 1023, the RCT commander stopped smiling.

The OPFOR did not destroy the Dragonlords.

It separated them.

That was worse.

The lead regiment seized the first crossing anyway. It did so with professional violence and enough mass that the aggressor screen did not pretend a delaying company could stop an RCT head-on.

Then the Dragonlords discovered that possession of a bridge did not mean the entire formation could reach it.

Exercise Control declared one secondary road cratered by simulated air attack.

Another became blocked when an OPFOR engineer team dropped a span over a drainage cut.

The RCT had alternate routes.

Those routes had not been rehearsed with the actual traffic load now trying to use them.

The road network became a parking problem with artillery.

The OPFOR returned after dark.

Not in strength.

It did not need strength.

A reconnaissance platoon cut a communications relay.

An infantry patrol reached a recovery staging area and marked three disabled BattleMechs as destroyed before anyone challenged them.

A pair of fast tanks fired on a convoy, forced it to disperse, and left.

An OPFOR artillery battery fired one simulated mission against a field kitchen because the exercise controller had noticed too many soldiers standing in the same place under visible lights.

The kitchen staff were furious.

Training Command considered that healthy.

At 0314, the RCT headquarters received three reports that Objective Green had been attacked.

One was correct.

One was five minutes old.

One was false.

The staff treated all three as equally urgent.

By dawn the Dragonlords still held the first bridge.

They did not hold the second.

They had not established the required lodgment.

Their lead formations remained tactically dangerous.

Their campaign had failed.

The AAR began two hours later.

Nobody had slept enough.

That was deliberate too.

The senior observer asked the commanding general one question.

"What beat you?"

The general looked at the map.

There were plenty of answers available.

OPFOR.

Road restrictions.

Exercise-control injects.

Unfamiliar terrain.

A bridge with a load rating nobody liked.

He rejected all of them.

"We outran ourselves."

The observer nodded.

"Good."

One of the BattleMech colonels looked irritated.

"We won every direct engagement."

The observer turned toward him.

"Then you have identified something you already knew you could do."

The colonel stopped.

The observer pointed at the failed lodgment.

"We are here for the rest."

The Dragonlord command sergeant major summarized the first week less politely in the enlisted dining facility.

"They kicked our buttons clean off."

A young MechWarrior objected.

"We destroyed more of them than they destroyed of us."

The sergeant major looked at him.

"Son, if you lose the mission while winning the kill count, the enemy gets to embroider the buttons."

The story spread faster than the official AAR.

The Second Beating

The next major exercise slowed the Dragonlords down.

They hated it.

Exercise Control gave them one primary supply route, two secondary routes of uncertain capacity, a damaged bridge, and a civilian evacuation column moving opposite military traffic.

The lead BattleMech regiment waited for engineers.

Armor waited for recovery clearance.

Artillery displaced early enough to remain in range.

Logistics officers placed movement-control teams at junctions before the roads became arguments.

The RCT reached its first objective four hours later than the schedule from ROAD MARCH ONE would have permitted.

Then the OPFOR counterattacked.

This time the Dragonlords had ammunition.

The guns were in range.

Recovery vehicles could reach damaged machines.

The reserve knew where it was supposed to go.

The counterattack failed.

The RCT commander stood over the final map and said what everyone was thinking.

"We were slower."

The observer asked, "Were you?"

The general looked at the intact supply route.

At the seized objective.

At the ammunition state.

At the recovery teams already moving simulated casualties.

"No."

"Better."

Then the observer changed the display.

The second objective remained enemy-held.

"You were also late."

The satisfaction disappeared.

Point Barrow did not let lessons remain comfortable long enough to become slogans.

The third exercise punished the correction.

The Dragonlords, determined not to outrun support again, became cautious.

Too cautious.

A reconnaissance report identified an apparent OPFOR concentration near a ridgeline guarding the direct route.

The RCT paused.

Artillery registered.

Aerospace scouts checked.

Engineers looked for bypass routes.

Everything was sensible.

The concentration was mostly decoys.

While the Dragonlords carefully prepared to defeat it, the second OPFOR command crossed thirty kilometers of bad ground on the RCT's eastern flank and seized the actual road junction that controlled the next phase.

The Dragonlords did not lose many simulated units.

They lost six hours.

The observer wrote one sentence on the AAR board.

CAUTION IS NOT THE SAME THING AS CONTROL.

The RCT commander stared at it.

"You people enjoy this."

The observer answered honestly.

"A little."

By the end of the first month, the 2nd Dragonlords had failed three major exercises in three different ways.

Too fast.

Too slow.

Too careful.

Morale did not collapse.

It became irritated.

That was useful.

People stopped defending the rating they wanted and started attacking the problems they actually had.

Red Night

The night exercise was where the RCT first scared itself.

Training Command called it RED NIGHT because Training Command occasionally surrendered to dramatic names.

The scenario assumed a degraded communications environment and an enemy capable of electronic deception.

The Dragonlords knew that before crossing the line of departure.

Knowing did not make it easy.

The first hour went well.

Too well.

That should have warned them.

Then three things happened within nine minutes.

One battalion lost its primary data link.

An OPFOR unit began transmitting enough false traffic to make the network uncertain whether a friendly armor company had crossed a phase line.

A brigade headquarters displaced after simulated counterbattery fire and discovered its alternate antenna site had line-of-sight problems nobody had identified during daylight reconnaissance.

The B-series systems kept displaying information.

That did not mean the information was good.

The RCT staff began calling subordinate headquarters for confirmation.

The calls multiplied.

The network filled.

One BattleMech battalion stopped moving because its commander could not confirm whether artillery had cleared a planned route.

An infantry regiment continued moving because its commander had the same uncertainty and interpreted the RCT intent differently.

For fourteen minutes, two formations with the same mission solved the same uncertainty in opposite ways.

The OPFOR found the seam.

It hit the infantry first.

The BattleMech battalion heard the contact.

Its commander requested permission to reinforce.

The RCT commander looked at the display.

The request was tactically reasonable.

The larger mission was not threatened.

He denied it.

The battalion commander swore after closing the circuit.

Then obeyed.

The infantry regiment fought alone for twenty-three minutes and withdrew according to plan.

The OPFOR followed.

That was the mistake the Dragonlords had been waiting for.

The BattleMech battalion struck the aggressor flank as it crossed the next phase line.

Artillery arrived thirty seconds later.

The attack was not a masterpiece.

It was coordinated.

For the first time since arriving at Point Barrow, one of the OPFOR commanders had to commit his own reserve to recover an exercise situation rather than simply exploit a Dragonlord error.

The Dragonlord command post noticed.

No one celebrated.

They were too busy.

Later, at the AAR, the aggressor commander said, "Tonight you made us work."

The RCT commander smiled.

"Did we win?"

"No."

The smile vanished.

"You still lost the northern objective."

"I was beginning to like you."

"That would interfere with training."

RED NIGHT remained a failure.

It was the first failure that felt different.

The Dragonlords had stopped merely surviving the OPFOR's decisions.

They had begun forcing decisions in return.

Controlled Aggression

By August the Dragonlords had become slower.

Then they became faster.

Not in movement.

In decision.

The change began below RCT headquarters.

Battalion commanders stopped reporting problems they had already solved.

Regimental staffs learned the difference between informing higher headquarters and asking permission.

Artillery commanders repositioned before the request for fire arrived because they understood the next likely phase of the maneuver.

Engineers began moving bridging and route-clearance equipment toward probable requirements rather than waiting for a crater to become an emergency.

Logistics officers became professionally unpleasant about routes that could not sustain the tonnage assigned to them.

The B-series systems helped.

They also exposed people who expected a screen to make decisions for them.

One brigadier called RCT headquarters during a mobile-defense exercise.

"Request permission to shift Second Battalion south."

The commanding general checked the situation display.

"Why?"

"Enemy armor is attempting to turn the flank."

"Does shifting violate my intent?"

"No."

"Any restriction?"

"No."

"Then why are you calling me?"

Silence.

"Understood."

The battalion shifted.

The armor attack failed.

The general wrote three words in his notebook afterward.

CONTROLLED AGGRESSION WORKS.

The command sergeant major saw it.

"Sir, you may want to add a fourth."

"Which?"

"Sometimes."

The general laughed and added it.

The phrase spread anyway.

Not initiative without limits.

Not blind obedience.

Act inside intent.

Understand restrictions.

Know logistics.

Move.

The next phase of training became deliberately ugly.

The OPFOR stopped presenting itself as a coherent enemy force.

It split.

Recombined.

Used decoys.

Threatened bridges it did not care about.

Raided maintenance while displaying BattleMechs somewhere else.

Fed false reports into exercise channels.

Moved infantry through terrain the Dragonlord BattleMech staff initially considered irrelevant.

The RCT's first instinct was to spread.

The aggressors punished that immediately.

A BattleMech battalion sent to investigate a supposed breakthrough discovered a reinforced OPFOR group waiting for it.

Aerospace reconnaissance corrected the picture before the trap closed completely.

The battalion withdrew under pressure.

At the AAR, an observer asked why the RCT commander had not reinforced it and crushed the aggressor concentration.

The general pointed toward the actual objective.

"Because they wanted me to."

That answer earned a checkmark.

The following week, an aggressor force attacked an ammunition distribution point.

The Dragonlords reinforced.

A staff major asked why this attack deserved forces when the previous one had not.

The commanding general pointed toward the ammunition forecast.

"Because losing this changes tomorrow."

A BattleMech battalion moved.

An armor battalion followed.

Aerospace support arrived first.

The attackers withdrew.

The Dragonlords did not pursue.

A younger MechWarrior complained.

"We had them."

His battalion commander answered.

"No. We had the depot."

The phrase moved through the RCT.

So did something else.

Confidence.

Not the confidence they had brought from Lindsay.

That confidence had been based on possessing good units.

The new version came from knowing what those units could do together.

The Month They Stopped Keeping Score

The second month at Point Barrow produced fewer dramatic failures and more exhausting ones.

That was harder on morale.

A spectacular mistake could be hated.

A hundred small mistakes had to be fixed.

The RCT began a seventy-two-hour field problem with eighty-nine percent of its major equipment available.

That number fell before the OPFOR fired a simulated shot.

A coolant pump failed on a BattleMech that had passed inspection the night before.

Two tracked carriers threw road wheels on the same stretch of broken ground.

An artillery tractor developed a transmission fault while towing a gun uphill.

A conventional fighter was grounded for an instrument discrepancy that the pilot could not feel and maintenance could not immediately reproduce.

The old instinct was to call those things bad luck.

Point Barrow called them Tuesday.

The RCT commander began asking a different question at morning briefings.

"What is unavailable before the enemy touches us?"

The first answer embarrassed everyone.

The second was better.

By the fourth week, maintenance officers were briefing projected availability rather than yesterday's percentage.

Recovery officers began placing assets according to where machines were likely to fail, not where the original plan had drawn a neat rear area.

Medical planners did the same with casualty collection points.

The headquarters started treating fatigue as a resource state.

That irritated combat officers more than fuel accounting ever had.

One BattleMech battalion commander protested when Exercise Control ordered his unit to stand down four lances whose pilots had exceeded the exercise fatigue threshold.

"The machines are fine."

The medical officer answered.

"The machines are not the pilots."

"They can fight."

"Yes."

"Then why ground them?"

The RCT commander looked at the sleep log.

"Because I need them tomorrow too."

The battalion commander disliked the answer enough to remember it.

The OPFOR attacked six hours later.

The rested reserve lances were the ones that stopped the penetration.

No one mentioned the argument afterward.

They did not need to.

Another exercise forced the RCT to conduct a withdrawal while maintaining contact.

The Dragonlords hated withdrawals.

Most soldiers did.

The word carried too much history and too much fear of what other people might think.

Point Barrow removed the emotion by making withdrawal the mission.

Three defensive lines.

Two river crossings.

One simulated civilian population to move through the corridor.

The OPFOR would win if it fixed enough Dragonlord combat power in place to prevent the final displacement.

The first line went well.

The second nearly became a disaster when one BattleMech company stayed six minutes too long trying to finish an aggressor armor battalion it had already hurt badly.

Six minutes allowed OPFOR artillery to register.

The company lost four machines in simulation before it disengaged.

Its commander arrived at the AAR angry.

"We could have destroyed them."

The observer nodded.

"Yes."

"Then why is that wrong?"

The observer pointed to the clock.

"Because they destroyed your time."

The company commander stared.

The observer continued.

"The enemy does not have to kill you if he can make you stay."

That sentence reached the entire RCT by evening.

The next withdrawal was cleaner.

Not graceful.

Clean enough.

A damaged tank was abandoned after the crew transferred its classified equipment and climbed onto another vehicle.

A MechWarrior whose machine had lost a leg argued to remain and provide covering fire.

His company commander refused.

"You are not a bunker."

"Sir, I can still shoot."

"And we can recover the machine tomorrow if you are not captured inside it tonight."

The pilot ejected from the cockpit and left with the recovery team.

The OPFOR marked the BattleMech captured.

The RCT commander approved the decision anyway.

Machines were expensive.

People were harder to replace.

Point Barrow was turning doctrine into reflex one irritating decision at a time.

Holding Their Own

The measure of progress at Point Barrow was not that the Dragonlords began winning every exercise.

They did not.

Professional OPFOR existed specifically to prevent that kind of comforting fiction.

The measure was that the defeats became expensive for the aggressors too.

During a late-August delaying exercise, the fast OPFOR command attempted the same kind of rear-area raid that had wrecked the Dragonlords during their first week.

This time the fuel site moved before the raiders arrived.

The visible trucks were decoys.

The real distribution point sat twelve kilometers away beneath camouflage, protected by infantry and a reserve armor company that had been assigned without RCT headquarters directing every position.

The OPFOR hit the decoys.

Then discovered it had been seen crossing the approach route.

Dragonlord aerospace fighters came in low.

The aggressor force scattered.

Two companies escaped.

One did not.

Exercise Control marked it destroyed.

The OPFOR commander arrived at the AAR smiling.

"That was rude."

The Dragonlord logistics colonel answered, "Thank you."

Three days later, the heavy aggressor command broke one of the RCT's armor regiments in a meeting engagement and forced the Dragonlords out of a river valley.

The RCT lost the exercise.

Nobody panicked.

The armor regiment withdrew in order.

Recovery assets moved before being called.

Artillery covered the disengagement.

The reserve established the next defensive line.

The RCT commander refused two requests to counterattack immediately.

At the AAR the observer asked why.

"Because we were angry."

The observer waited.

"Angry units make expensive decisions."

Another checkmark.

The next exercise ended in what Exercise Control called a draw and both sides called unfinished.

The Dragonlords seized one objective.

The OPFOR retained another.

Neither could dislodge the other before the time limit without committing reserves needed elsewhere.

The aggressor commander looked at the clock.

"I dislike draws."

The Dragonlord general nodded.

"So do I."

"Good."

That exchange mattered more than a victory ceremony would have.

The Dragonlords were no longer arriving at AARs desperate to prove they had really won.

They wanted to know what would break first if the exercise continued another twelve hours.

Fuel?

Ammunition?

Pilots?

Tracks?

Communications?

Bridges?

That was the beginning of campaign thinking.

The Opponents Adjust

Professional OPFOR did not stand still while the Dragonlords improved.

That was another reason the certification meant something.

The aggressor staffs kept their own AARs.

After the first month, those reports mostly described Dragonlord mistakes.

By the third, they described Dragonlord habits that had become dangerous.

The fast OPFOR command stopped assuming rear-area targets would remain where reconnaissance first found them.

The heavy command began holding a second reserve because Dragonlord counterattacks were becoming less predictable.

Aggressor pilots reported that Dragonlord aerospace patrols no longer chased every contact.

Engineer teams found withdrawal routes covered more often.

Artillery observers had to move faster after firing because counterbattery response times were shortening.

The senior aggressor commander brought one report to the RCT commander before the final certification exercise.

"You should read this."

The Dragonlord general looked suspicious.

"Is this another trick?"

"Probably. Read it anyway."

The report contained no praise.

It contained revised OPFOR procedures.

DO NOT ASSUME DRAGONLORD REAR AREAS ARE STATIC.

EXPECT SUBORDINATE COMMANDERS TO ACT WITHOUT RCT CONFIRMATION.

AVOID EXTENDED PURSUIT AFTER LIMITED SUCCESS.

DRAGONLORD ARTILLERY REPOSITIONING FASTER THAN INITIAL BASELINE.

EXPECT LOGISTICS DECOYS.

The general read the last line twice.

"You changed doctrine because of us."

"Exercise procedure."

"That sounds less satisfying."

"It should. We are not writing your citation."

The general smiled anyway.

That was the first time Point Barrow showed him progress through the enemy's paperwork.

The lesson stayed with him.

A competent opponent did not continue making the same mistake simply because it had worked before.

Every success purchased adaptation.

The Dragonlords would need to keep learning after certification because the people shooting at them would too.

Six Days

The final certification exercise lasted six days.

No single day looked decisive.

That was intentional.

Both Point Barrow OPFOR commands participated.

For the first forty-eight hours they kicked the Dragonlords again.

A feint pulled one reconnaissance battalion too far west.

An artillery position was found because a crew used the same displacement route twice.

A medical collection point became overwhelmed when Exercise Control imposed casualties from two different axes at once.

The RCT lost a bridge on the first night and discovered the replacement crossing site had worse banks than the engineer reconnaissance suggested.

The old Dragonlords might have treated those events as a chain of emergencies.

The Point Barrow Dragonlords treated them as conditions.

The reconnaissance battalion returned without being ordered to save face.

The artillery regiment shifted firing responsibility between battalions while the exposed battery moved.

Medical requested transport from an infantry regiment whose trucks were temporarily underused.

The engineers opened the second-best crossing because the best crossing would take too long.

Nothing looked elegant.

The campaign continued.

On the third day, the OPFOR attempted to break the RCT by attacking three things at once.

A communications node.

An ammunition point.

A BattleMech regiment headquarters.

Six weeks earlier, the Dragonlords would have responded to all three because all three sounded important.

This time the regimental headquarters displaced under its own plan.

Signals rerouted the communications traffic.

The reserve went to the ammunition point.

The OPFOR broke contact before being trapped.

The Dragonlords did not chase.

On the fourth day, the heavy aggressor command committed most of an armor regiment against what it believed was a tired Dragonlord infantry sector.

The infantry was tired.

The sector was real.

The weakness was deliberate.

The Dragonlord RCT had finally learned to show the enemy something true without showing everything.

The OPFOR armor went in.

Infantry held long enough.

Artillery massed.

A Dragonlord armor regiment struck the flank.

The aggressor commander disengaged before the trap closed completely.

That annoyed the Dragonlords.

Good opponents were allowed to notice traps too.

Day Five was mostly movement.

That exhausted everyone more than a spectacular battle would have.

The RCT shifted its logistics base twice.

One BattleMech regiment operated below seventy percent simulated strength.

Aerospace availability fell because maintenance rules grounded aircraft that had accumulated too many sorties.

The commanding general accepted the reduced air cover instead of bullying maintenance into inventing readiness.

One brigade executive officer took command after his commander became a simulated casualty.

He made one bad decision.

Then corrected it.

No one at RCT headquarters replaced him merely because they could see the mistake.

On the sixth morning, the fast OPFOR command found what looked like the final opening.

A gap between the Dragonlord forward security screen and the logistics area.

The aggressors went through it.

They discovered an engineer battalion, an infantry reserve, two armor companies and a minefield simulation waiting on the route they had chosen.

Not a perfect ambush.

A bounded one.

The Dragonlords were not trying to destroy the OPFOR.

They were trying to force it away from the supply corridor while the main body completed the movement to the final objective.

The aggressor commander saw the geometry.

He withdrew.

The RCT completed the movement thirty-eight minutes before the exercise ended.

No cheering followed.

People were too tired.

The senior observer found the commanding general sitting on the step of a command vehicle with a cup of coffee that had surrendered hours earlier.

"How do you think you did?"

The general stared at the cup.

"Ask me after the board."

"Good answer."

"I hate when you say that."

The certification board met the next morning.

No one from the RCT attended.

That was deliberate.

The board reviewed training reports.

Command decisions.

Casualty handling.

Logistics.

Staff integration.

Subordinate initiative.

Failures.

Corrections.

The board did not ask whether the 2nd Dragonlords had become elite.

They had not.

It did not ask whether the OPFOR could still beat them.

It could.

The question was whether the RCT could now enter a campaign as a coherent Regular formation and continue functioning after the first plan broke.

The final line was short.

2ND DRAGONLORDS RCT: REGULAR.

The commanding general received the certification six minutes after his chief of staff.

He read it twice.

Full reinforced RCT.

Regular.

For the first time, both statements meant what he wanted them to mean.

He called his chief of staff.

"We passed."

"I know."

"How?"

"The board transmitted to headquarters first."

The general stared at the communicator.

"You knew before I did?"

"Six minutes."

"I hate you."

"Regularly, sir."

The general laughed.

Then looked through the command-post window toward the ranges that had spent months proving him wrong.

The Point Barrow OPFOR had kicked his RCT's buttons off repeatedly.

The Dragonlords had learned how to sew them back on while moving.

That was probably close enough to Regular.

The 2nd Dragonlords returned to Lindsay stronger than when it had left, although very little of its equipment had changed.

That was the point.

Sirdar - Ninety Minutes

The Capellan raiding force expected militia.

It found militia.

That was the problem.

The Sirdar Capellan March Militia had rotated formations through Combat Training Centers often enough that the habits were becoming ordinary.

The attacking BattleMech company was not incompetent.

Its commander had studied Davion reactions.

He expected defenders to cling to fixed positions until threatened with encirclement.

He expected militia scouts to pursue when given the chance.

He expected armor to disperse against BattleMechs rather than mass.

Those expectations had once been reasonable.

Captain Elin Varek received her mission shortly after dawn.

Delay.

Ninety minutes.

Quarry Junction.

Rallin River route.

Main Sirdar elements were moving.

Her force was not required to destroy the enemy.

That fact appeared twice in the order.

Someone in the planning staff had learned.

Varek had twelve BattleMechs.

Three Dervishes and one militia Kintaro in the fire-support and ambush element.

Two Clints and two Valkyries in reconnaissance.

Two Swordsmen and two Javelins in the line element.

The supporting force mattered more.

Tiger IIs.

Pegasus.

J. Edgars.

Hunter LRM carriers.

Infantry.

Engineers.

Forward observers.

Ammunition vehicles positioned farther back than Varek would have preferred and exactly where logistics insisted they belonged.

The Kintaro looked wrong to anyone familiar with the standard machine.

Its militia rebuild had traded away the token long-range missile rack for five jump jets and additional armor.

The change did not make it a general-purpose BattleMech.

It made it better at one unpleasant job.

Hide.

Move through terrain the enemy did not expect.

Jump into the kill zone.

Spend short-range firepower and heat.

Leave.

The Dervishes owned the approach.

The Kintaro owned the moment someone thought the approach had ended.

The Capellans entered the engagement expecting the BattleMechs to be the battle.

The first contact came from a Clint.

It fired.

Moved.

Reported.

A Valkyrie maintained contact from farther back.

The Capellan commander pushed reconnaissance forward.

The Clint withdrew.

No duel.

No attempt to prove anything.

The Capellan scouts advanced.

Dervish missiles arrived.

Not enough to destroy them.

Enough to make them deploy.

The clock began spending itself.

The enemy commander adapted quickly.

He tightened the formation.

Stopped chasing.

Used his faster machines to screen the advance rather than hunt kills.

Varek watched the change.

"Good."

Her executive officer looked at her.

"Good?"

"He's not stupid."

"That is generally worse."

"Yes."

The next attack came from the armor.

Pegasus and J. Edgars crossed the enemy's flank at speed under LRM cover.

They fired.

Not long.

Not heroically.

Then left.

The Capellan formation turned weapons toward them.

The hovercraft were already gone.

That bought four minutes.

Then the Tiger IIs fired.

The Capellan commander had seen the tanks on reconnaissance imagery.

Fifty-ton tracked machines.

Internal-combustion engines.

AC/10.

​

SRM-4.

​

Nothing about them demanded fear.

That was why they worked.

The first platoon had four tanks.

All four fired at the same Vindicator.

Not all hit.

Two did.

The Vindicator remained standing.

Its pilot reacted.

That was enough.

Second platoon fired on a Shadow Hawk.

Third shifted toward a second Vindicator.

Twelve tanks did not fight twelve independent duels.

They created three unfair ones.

The Capellan company stopped.

Weapons turned toward the tanks.

The Tigers reversed behind prepared cover.

Hunters fired from farther back.

Dervish missiles joined them.

The Capellan commander ordered a push before the tanks could establish a second firing line.

Correct.

Varek had expected it.

The Tigers were already moving.

Slowly.

They were not hovercraft.

No one asked them to pretend.

The first position existed to fire.

The second existed to survive the answer.

The Capellans advanced into the gap.

The Kintaro jumped.

Five jump jets carried it out from concealment onto the flank of an enemy medium.

Missiles and lasers struck at close range.

The Kintaro did not remain.

Its pilot spent heat.

Jumped again.

Disappeared behind broken terrain.

The Capellan commander stopped trying to crush individual ambushes.

He drove toward the junction.

That was also correct.

Varek changed the plan.

"Stop trying to turn him."

Her executive officer looked over.

"Then what?"

"Make him pay for going straight."

The Clints widened.

The Valkyries shifted to observation.

The Swordsmen stopped moving laterally and established direct-fire positions near the next choke point.

Dervishes changed target priorities from lead machines to the units trying to protect the flanks.

The Capellan formation remained coherent.

That made the battle harder.

The militia did not become magically superior because doctrine improved.

A Vindicator caught one Javelin crossing between positions and hit it hard enough to destroy the right arm and damage the hip.

The pilot stayed upright.

Varek ordered him out.

"I can fight."

"You can leave."

"Captain -"

"Recovery route three. Now."

The Javelin limped away.

A Hunter LRM carrier died when a Capellan light found its firing position before the relocation order completed.

The crew did not all get out.

The first Tiger II lost a track to direct fire.

Its commander asked whether recovery could reach it.

The answer came back no.

The crew fired twice more, removed encryption modules, pulled the ammunition feed safeties and abandoned the tank.

The machine continued burning behind them.

Doctrine did not make war clean.

It made some deaths avoidable.

Not all.

The militia infantry waited at Quarry Junction.

Not in one glorious defensive line.

In positions built to make the final approach expensive.

Engineers had prepared routes.

Some were open.

Some looked open.

One was neither.

The Capellan BattleMechs reached the outer junction at eighty-six minutes.

Varek's company had lost two machines to combat damage severe enough to remove them from the fight and one Hunter platoon had been badly hurt.

Several Tiger IIs were damaged.

One burned.

The enemy company remained dangerous.

The main Sirdar force began appearing on the operational plot.

Eighty-nine minutes.

Varek received the withdrawal code.

One of her Javelin pilots asked the inevitable question.

"We can hold."

Varek looked at the map.

"Yes."

A pause.

"Then -"

"That wasn't the mission."

The order went out.

Withdraw by echelon.

Hunters moved first.

Infantry followed prepared routes.

Damaged vehicles received recovery priority based on time.

One Tiger II that could not move was stripped of sensitive equipment and abandoned.

The crew left with another vehicle.

The Dervishes fired the last planned covering missions.

Clints remained in contact until the Sirdar relief force acquired the enemy directly.

The Capellan commander reached Quarry Junction at ninety-two minutes.

Then saw the approaching militia formations and regular attachments moving into prepared sectors beyond it.

He had not been defeated.

Neither had Varek.

That was the point.

At the AAR someone told her:

"You didn't beat them."

Varek nodded.

"No."

She looked at the mission clock.

"We weren't supposed to."

Training Command received the combat report.

Hall read it.

Wrote one line.

Sent it to every CTC.

THIS IS WHAT WE MEANT.

​

The Tank Nobody Respected

The Tiger II report traveled farther than Hall intended.

Capellan intelligence received portions first.

Then more.

The initial assessment was dismissive.

The tank itself justified some of that.

Medium weight.

Tracked.

ICE.

​

AC/10.

​

SRM-4.

​

Easy to maintain.

No advanced technology.

No miracle armor.

No extraordinary speed.

The revised assessment was more cautious.

The problem was not the tank.

The problem was how many places the tank appeared.

Planetary militias.

Conventional Divisions.

March Militias in useful numbers, though rarely to the exclusion of heavier or more specialized armor.

Training grounds.

Depot inventories.

Replacement allocations.

The tank was becoming ordinary.

That word mattered.

A rare excellent machine influenced battles.

An ordinary adequate machine influenced wars.

A Capellan analyst wrote:

ASSESSMENT SHOULD NOT CONCENTRATE ON SINGLE-VEHICLE THREAT. DAVION FORCE EMPLOYMENT EMPHASIZES PLATOON AND COMPANY FIRE CONCENTRATION.

The report continued.

Standard ammunition.

Standard maintenance.

ICE support requirements compatible with less-developed worlds.

Growing crew familiarity.

Increasing parts availability.

Another analyst read it.

Then wrote in the margin:

THAT'S THEIR ORDINARY MEDIUM TANK NOW.

​

The phrase was not intended for senior review.

It reached senior review.

It remained.

The question changed.

Not how dangerous was one Tiger II?

How many Tiger IIs could the Federated Suns place between an attacker and the objective?

How many trained crews?

How many ammunition trucks?

How many mechanics?

How many replacement tracks?

How many companies capable of concentrating fire without each tank deciding to fight its own private war?

Those were less comfortable questions.

What Kept Failing

The captured Centurion should have been unremarkable.

That was why the technician became worried.

The machine had been rebuilt.

Not transformed.

Its armor was conventional.

Engine conventional.

Weapons conventional.

The Luxor heavy autocannon was still a Luxor.

But the feed components were newer.

Harnesses had been replaced.

Sensors improved.

Access points changed.

Seals carried recent manufacture dates.

Several connectors matched standards found on other captured Davion equipment.

The technician opened another panel.

Then another.

He called his supervisor.

"What?"

"You need to see this."

The supervisor arrived expecting Lostech.

He found wiring.

His disappointment lasted five minutes.

Then he understood.

"They kept the gun."

"Yes."

"What did they change?"

The technician looked inside the machine.

"Everything around the part that kept failing."

That report joined others.

Shadow Hawks.

Javelins.

Clints.

Swordsmen.

Vehicles.

Different factories.

Different worlds.

Same fittings appearing where standardization allowed.

Same maintenance logic.

Same preference for replacing mundane failure points instead of waiting for a complete new machine.

The foreign intelligence problem changed.

For years the Federated Suns had been measured by production.

How many new BattleMechs?

How many tanks?

How many fighters?

The refurbishment network made those questions incomplete.

A thirty-year-old machine receiving new wiring, new seals, rebuilt actuators, modern diagnostics and a standardized communications package still looked thirty years old in an Order of Battle.

It did not behave thirty years old in a maintenance bay.

A century-old machine rebuilt to Category A reserve standard could emerge from storage and surprise an enemy analyst who had already mentally written it off.

The AFFS was not replacing the old army.

It was making the old army less old a piece at a time.

The oldest UrbanMech in the new service-life program eventually entered the same intelligence stream.

The external plate said 2698.

The rebuild record said 3010.

One analyst wrote:

DAVION IS RETURNING OBSOLETE EQUIPMENT TO SERVICE.

​

His supervisor changed it.

DAVION IS ASSIGNING REFURBISHED EQUIPMENT ACCORDING TO MISSION SUITABILITY RATHER THAN PRESTIGE.

​

The second sentence was less comforting.

The Other Houses

By October, the other Great Houses were no longer asking whether the Federated Suns was stronger.

They were asking why.

The Draconis Combine understood the military portion first.

Not because it admired Davion reform.

Because professional soldiers recognized professional problems.

A DCMS staff study began with the Dragonlords.

The obvious conclusion was that House Davion had simply concentrated experienced personnel into new formations.

That conclusion survived less than a week.

The personnel records did not support it.

Veteran officers had moved.

Senior NCOs had moved.

Cadres had moved.

But no single existing brigade had been gutted to create the new RCTs.

The deeper investigators found schools.

Instructor pipelines.

Training rotations.

Reserve equipment.

Technical education.

The slow creation of enough experienced people that experience itself could be distributed instead of stolen from one unit to create another.

A DCMS colonel read the analysis twice.

"Where did the cadre come from?"

The intelligence officer pointed toward the annex.

"Everywhere."

"That is not an answer."

"It may be the answer."

The colonel did not like that.

He liked the next part less.

The 2nd Dragonlords had been held out of deliberate combat employment while Green despite being a full reinforced RCT.

To a culture that valued battlefield proof, the decision initially looked cautious.

Then the colonel saw Point Barrow.

The AFFS had spent most of a year correcting the formation rather than accepting its paper strength.

"They could have used them."

"Yes."

"They chose not to."

"When they had a choice."

The colonel sat back.

That was not softness.

That was confidence in depth.

The Combine began adapting.

More attention went toward training establishments.

Unit integration exercises received greater emphasis in selected formations.

The lower-tier B-series systems fed technical study.

Not every response worked.

Not every senior officer agreed.

That also was normal.

Competent enemies did not need to become copies to learn.

The Capellan Confederation reacted with greater urgency.

The border was closer to the most visible Davion improvements.

Tiger IIs appeared in reports.

Swordsmen.

Clints.

Refurbished Centurions.

Militia units that withdrew instead of dying helpfully in place.

The temptation was to answer visible equipment with visible equipment.

More production.

More units.

More command machines.

The Strategios could authorize all three.

The harder questions came afterward.

Who maintained them?

Who trained the staffs?

Who replaced instructors removed to create new formations?

Who certified a unit that looked good during parade season and failed when artillery, infantry and armor all needed the same road?

One Capellan planning meeting ended with a procurement officer accusing an intelligence analyst of defeatism.

The analyst responded by placing a Davion readiness report beside a Capellan one.

"They are not invincible."

"Then stop talking like they are."

"I am not. I am saying they have become better at discovering when they are weak before we do it for them."

That ended the argument more effectively than shouting would have.

The Lyran Commonwealth had a different problem.

Industry itself was not mysterious.

Tharkad understood factories.

It understood large production runs.

It understood that a Great House could mobilize enormous industrial capacity when politics, money and infrastructure aligned.

The Davion lesson therefore looked easier at first.

Build more.

Automate more.

Standardize more.

Then a Lyran logistics officer asked:

"Who maintains all of it?"

The room became less enthusiastic.

Factories could be expanded faster than experienced technicians could be created.

New formations could be equipped faster than staffs could learn to integrate them.

A carrier hull could be financed faster than a carrier crew could become experienced.

The Lyrans could copy machinery.

The people problem remained stubbornly biological.

The Free Worlds League found the situation irritating in a different way.

Almost every individual Davion reform existed somewhere in the League.

Excellent technical schools.

Advanced industry.

Professional military academies.

Strong regional formations.

Merchant networks.

Shipyards.

Industrial worlds.

Mercenary relationships.

One analyst summarized the frustration:

"Everything the Davions do exists somewhere in the League."

A senior official looked at him.

"Then why don't we get the same result?"

The analyst hesitated.

"Because theirs talks to itself."

That sentence offended nearly everyone.

It remained true.

The Taurian Concordat saw almost all of it through older fear.

More roads meant military movement.

Better militia meant secure rear areas.

Technical schools meant technicians for future invasion armies.

Blue Water carriers looked like tools for planetary coercion even though most could never leave the worlds where they were built.

Bucklers were worse precisely because they could.

A deployable light carrier arriving inside a modified Mammoth fit too neatly inside Taurian nightmare logic.

One defense official pointed at the transport profile.

"That is not a training ship."

A younger analyst answered carefully.

"No, sir. It is a carrier designed to be moved between worlds."

"Exactly."

The analyst knew better than to add that a carrier designed to defend an isolated ocean world and a carrier designed to support invasion could be the same ship depending upon who owned the coastline.

The Magistracy of Canopus saw the dual-use problem more clearly.

A stronger Federated Suns could be dangerous.

It could also make trade routes safer.

Schools produced military technicians and hospital technicians.

Transport infrastructure moved troops and medicine.

Waylands repaired combat vehicles and civilian machinery.

The Canopian conclusion remained cautious engagement.

Useful neighbors were still neighbors with armies.

The Outworlds Alliance watched education more than tanks.

Competence was attractive.

That remained its quiet concern.

The Sphere was beginning to understand that Davion strength was not one thing.

That did not mean anyone agreed what to do about it.

Autumn Rotations

The March Militia CTC schedule entered its second major rotation cycle in September.

By then the paperwork had become ordinary enough to hide the scale of the change.

A battalion from one command arrived as another left.

A staff exercise began while an armor regiment loaded for home.

Professional OPFOR moved between ranges.

Instructor teams rotated.

Ammunition allotments followed training plans rather than local political priority.

Transportation Command complained.

Training Command complained back.

That was healthy.

One regional commander asked whether the constant movement was consuming too much readiness.

Hall sent the question to the readiness board instead of answering personally.

The analysis came back irritatingly balanced.

Yes.

Training rotations temporarily reduced local strength.

Yes.

Movement caused maintenance wear.

Yes.

Ammunition used in training had to be replaced.

Yes.

Civilian transport occasionally had to be contracted because military lift was finite.

The conclusion still supported continuing.

The reason was simple.

A formation that never left its home district became excellent at defending the version of the district it already knew.

War rarely honored that version.

The rotations forced militia commands to practice leaving familiar roads, integrating unfamiliar attachments and operating under headquarters that did not already know every local commander by voice.

They also exposed weaknesses before those weaknesses became enemy intelligence.

One militia artillery regiment discovered that its ammunition-handling drills worked perfectly at its home depot and poorly with the pallet equipment used at another CTC.

A standard adapter package appeared three months later.

One infantry regiment discovered that half its vehicle radios could not accept the newest encryption module without field wiring changes.

The change became a depot kit.

A militia headquarters discovered that its casualty reporting system generated excellent medical data and terrible operational summaries.

The B-500 training team fixed the reporting format.

None of those discoveries deserved a medal.

Together they made mobilization less surprising.

That was the point.

The LCT planners watched carefully.

The future Borderers would depend upon the same regional infrastructure without becoming March Militia units themselves.

Training routes.

Ammunition packages.

JumpShip access.

CTC familiarity.

Regional intelligence.

The doctrine increasingly assumed that an LCT would not arrive in a region as a stranger.

It would know the roads before it needed to burn them.

It would know which bridges could carry heavy armor before the first retreat.

It would know which militia headquarters could receive attachments without collapsing under them.

The Borderers were not a substitute for local defense.

They were intended to make local defense harder to isolate.

The rotations also changed the culture of the units that returned home.

A militia battalion came back from a CTC in October and found its district under a real alert forty-eight hours later. A merchant JumpShip had appeared on an irregular vector with damaged identification hardware, and a nervous local controller had reported the possibility of a disguised raider.

It was not.

The merchant crew had suffered an electrical fire and spent most of the approach trying to keep life support and maneuvering control functioning at the same time.

The alert still became a useful test.

Six months earlier the battalion would have assembled at its principal armory and waited for orders.

This time its companies went directly to preassigned dispersal points. Engineers checked two bridges without being asked. The medical section opened a casualty collection point because a landing accident remained possible even after the contact began identifying correctly. The headquarters established a liaison link with civilian traffic control instead of demanding that civilian traffic disappear.

Nothing fired.

Nothing exploded.

The battalion stood down before midnight.

The commander sent Training Command a message the next morning.

CTC LESSONS VALIDATED. APOLOGIES FOR LACK OF ENEMY.

The reply arrived three hours later.

ENEMY NOT REQUIRED FOR READINESS.

The exchange became mildly famous.

More importantly, the battalion's post-alert maintenance report showed lower vehicle fault rates than its previous real mobilization. The difference was not dramatic. It was measurable.

Movement had become practice instead of disruption.

That was what the rotation schedule was buying.

It was also why the professional OPFOR commands began appearing on personnel preference sheets. Soldiers understood that the CTC system was no longer a temporary reform project. It was becoming one of the places careers could be built.

By autumn, some young officers were asking for OPFOR duty before anyone told them it would be good for promotion.

Hall considered that the healthier reason.

Fourth Quarter

October - The Fourth Leaves Novaya

The 4th Dragonlords finished its Novaya cycle without ceremony.

The ceremony came later.

The certification did not.

Training Command's final report was almost disappointingly plain.

RCT integration acceptable.

Command relationships validated.

Artillery and aerospace coordination improved.

Logistics met field standard.

Engineer support effective under movement.

Remaining deficiencies within normal unit-correction authority.

Combat certification approved.

The orange square became green.

No one changed the color because a politician asked.

It changed because the reasons for orange had been corrected.

The 4th received movement orders for Brockway.

The 5th Dragonlords began occupying the institutional space the 4th had vacated months earlier on New Avalon.

That did not mean the 5th inherited the 4th's readiness.

It inherited the pipeline.

People arrived.

Equipment arrived.

Cadre formed.

The RCT began learning its own mistakes.

Its activation date remained December.

Its Novaya certification would come later.

That distinction irritated public-affairs staff and pleased Hall.

"Activated is not ready," he said.

The phrase had become old enough that people stopped writing it down.

That pleased him more.

The 9th Deneb Light Cavalry completed its restoration on Northwind during the same quarter.

The ceremony was small by New Avalon standards and enormous by the standards of people who understood what the name meant.

The 10th Deneb Light Cavalry formally transferred the last items it had held in trust.

Not all were weapons.

Records.

A damaged regimental standard.

A memorial book.

A box containing insignia removed from machines that had never returned.

The 9th's new commander accepted each item.

When the colors were presented, one old Deneb veteran stood at the edge of the formation and cried openly.

No one looked away to spare him.

Some grief deserved witnesses.

The 10th prepared for Rowe.

The 9th prepared to become something alive rather than historical.

November - The Borderers

The 1st Avalon Borderers did not exist because David Davion needed a command.

That point mattered enough that Hall repeated it to three different offices.

The formation had been designed before anyone knew who would command it.

The same was true of the 2nd Avalon Borderers forming beside it.

What changed in November was not the requirement.

It was the source of people.

Twenty-Four Commands

The twelve Combat Training Centers maintained twenty-four Professional OPFOR commands between them.

Two per center.

They had begun as a training investment.

By 3010 they had accidentally become one of the deepest pools of combined-arms experience in the Federated Suns.

An OPFOR officer might spend one week commanding against a militia brigade, the next against a regular BattleMech regiment, and the next helping design an exercise for an RCT whose staff had already read every public doctrinal manual.

Aggressor troops learned unfamiliar equipment quickly because their jobs often required them to represent different opponents.

They learned to attach and detach units without emotional drama.

They learned common communications procedures because nothing destroyed an exercise faster than an aggressor formation that could not talk to itself.

They learned to receive a new battalion in the morning and make it useful by afternoon.

Most importantly, they learned that an organization was not integrated merely because everyone involved was competent.

Training Command had spent years using the OPFOR commands to teach that lesson to everyone else.

Personnel Command finally noticed that the OPFOR personnel had learned it too.

The first proposal was too aggressive.

Move whole companies from selected OPFOR commands into the new Borderers.

Hall rejected it before the briefing ended.

"No."

The personnel officer looked toward the force table.

"Marshal, these are volunteers."

"Good."

"They want field assignments."

"Also good."

"Then why no?"

Hall brought up the CTC readiness map.

"Because I am not hollowing the schools to make the field army look better."

That ended the first proposal.

The second proposal worked because it treated the OPFOR system as a pipeline instead of a warehouse.

No OPFOR command would provide more than one BattleMech lance to the Avalon program.

Twenty-four commands.

Twenty-four lances.

Ninety-six experienced MechWarriors.

The arithmetic caught Hanse first.

"That is two reinforced battalions."

Hall nodded.

"Twelve lances to the First Avalon. Twelve to the Second."

"One lance from every command?"

"Maximum. Some commands will send fewer if readiness requires it. But that is the ceiling."

David looked at the distribution.

"And they do not bring their OPFOR machines."

"Correct."

The CTCs kept their training equipment.

The volunteers transferred as personnel and experience.

The new Borderers received machines from production, refurbished Category A reserves and normal formation allocations.

That distinction prevented a new combat formation from being built by quietly breaking twenty-four training formations.

The same rule applied to the other arms, although the packets differed by specialty.

An armor platoon here.

A fire-direction cell there.

An engineer squad.

A signals section.

A maintenance chief with three experienced technicians.

A senior infantry NCO who had spent six years teaching mechanized battalions why roads mattered.

No CTC gave up enough people from one place to lose its identity.

All twelve gave something.

The vacancies did not remain empty.

That was the second half of the plan.

New graduates.

Recruits.

Experienced soldiers volunteering for OPFOR tours because field commands had begun treating aggressor service as valuable rather than as exile from a real unit.

The OPFOR positions became career assignments people competed for.

A company commander at one CTC complained that he was losing his best lance leader to the 1st Avalon Borderers.

Hall asked, "Did she volunteer?"

"Yes."

"Would you take her back after a field tour?"

The commander paused.

"Immediately."

"Then you are not losing her. You are lending the army a better officer than it had before you trained her."

The commander still complained.

He signed the release.

That was how the system grew without pretending people were infinite.

People Who Already Knew How to Be New

The first Avalon BattleMech battalion assembled faster than David expected.

Not because the personnel already knew one another.

They did not.

A lance from Broken Wheel had never worked with a lance from Point Barrow.

An aggressor officer from Kearny had habits that irritated a lance leader from the New Avalon CTC within the first hour.

One group used tighter radio discipline.

Another expected lance leaders to speak freely when contact changed.

Three commanders had different preferred spacing for road movement.

Everybody had an opinion about reconnaissance.

That was exactly what David wanted to see.

The battalion's first planning session ran fourteen minutes past schedule because people kept producing sensible alternatives.

David finally stopped the discussion.

"You were selected because you think."

Several people looked pleased.

"Do not become attached to that feeling."

The pleasure disappeared.

"I still need one plan."

A lance commander raised a hand.

"Sir, which one?"

David looked at the battalion commander.

"His."

The battalion commander understood the gift and the burden at the same time.

He chose.

The others executed.

That was the first integration lesson.

The second came in the field.

OPFOR veterans were accustomed to exploiting hesitation.

Now there were too many of them trying to exploit the same hesitation.

Two lances maneuvered toward the same flank because both had recognized an opening.

A third tried to support them.

The center became thin.

The opposing force went through it.

The battalion lost the objective.

At the AAR David asked what happened.

One lance leader answered.

"We all saw the opportunity."

"Yes."

"We took initiative."

"Yes."

"And lost."

"Also yes."

David pointed toward the map.

"Controlled aggression requires the controlled part too."

Nobody forgot.

The infantry regiments integrated differently.

They had more new personnel and fewer veteran aggressor packets than the BattleMech battalion.

That made the OPFOR-trained NCOs disproportionately important.

They knew how to build rehearsals around attachments.

They knew to make a newly assigned engineer platoon brief its capabilities before the first mission instead of after someone discovered a vehicle could not cross the obstacle selected for it.

They knew how to tell a battalion commander that the route on the map did not exist on the ground.

They were occasionally insufferable about it.

The new soldiers learned quickly because the veterans were accustomed to explaining not merely what to do, but what mistake the procedure was designed to prevent.

Armor followed the same pattern.

The cavalry battalions learned to disappear without becoming lost.

The heavier battalions learned that direct-fire strength did not excuse sitting in one place until artillery found them.

Artillery learned the Borderer command expected displacement plans before the first round left the tube.

Signals learned that every headquarters would move.

Logistics learned that "light" still meant an appalling number of trucks.

By the end of November the 1st and 2nd Avalon Borderers had something most newly activated formations did not possess.

A large percentage of their experienced core had spent years being the unfamiliar attachment.

Now they were very difficult to surprise with unfamiliarity.

The Cost of Taking the Instructors

The price appeared immediately.

CTC commanders reported vacancies.

Not crises.

Vacancies.

Training Command treated the distinction seriously.

Replacement personnel were not simply thrown into aggressor units because the OPFOR happened to have open billets.

Some came from academy classes.

Some from units completing field tours.

Some were veteran NCOs who had discovered they liked teaching more than they liked pretending they did not.

Others were young officers specifically requesting aggressor duty because service there had begun appearing in promotion packets as evidence of adaptability.

A Personnel Command clerk noticed the trend and sent Hall a note.

OPFOR DUTY IS BECOMING DESIRABLE.

Hall wrote beneath it.

GOOD. DO NOT MAKE IT COMFORTABLE.

The clerk kept the note.

The twenty-four commands rebuilt their missing lances and small specialist packets over the following training cycles.

Not instantly.

Honestly.

For a few months some OPFOR units had more green personnel than their commanders preferred.

That became another training opportunity.

Experienced aggressors learned to teach replacements while continuing to train the rest of the AFFS.

The CTC system did not merely feed the field army.

The field army now fed experience back.

That was the institutional loop Andrew had wanted without ever having drawn this particular version of it.

The Volunteers

The transfer orders produced a reaction Training Command had not expected.

More people volunteered.

Not for the Avalon Borderers.

For the vacancies they left behind.

A young armor officer at the Kestrel CTC submitted for OPFOR duty two days after a veteran platoon leader received orders to the 2nd Avalon.

His commander asked why.

"Because everybody says they fight everyone."

"They do."

"Then I want to learn there before somebody shoots at me for real."

A signals NCO gave a different answer.

"Every unit I have supported says the OPFOR people are annoying."

The personnel clerk waited.

"I want to know why."

The clerk wrote recommended beside the application.

Some applicants wanted future promotion.

Some wanted variety.

Some were tired of garrison duty.

A few had spent years being beaten by Professional OPFOR and had developed the military equivalent of professional curiosity.

Training Command accepted none of those motives by themselves.

Applicants still had to meet standards.

A bad officer did not become useful because he wanted a fashionable assignment.

But the applicant pool changed.

That mattered.

The OPFOR commands could now select replacements rather than simply accept whoever Personnel had available.

A veteran aggressor captain complained that the new people were too ambitious.

His command sergeant major answered, "Good. We can cure that."

The captain looked at him.

"How?"

"Put them against the Guards."

The first replacement groups entered the CTC pipeline before the Avalon Borderers completed activation.

They watched experienced personnel pack for field assignments and understood that OPFOR service no longer meant disappearing from the real army.

It could be a route through it.

Hall disliked the phrase route because people immediately tried to turn routes into guarantees.

He issued guidance instead.

OPFOR DUTY MAY INFORM FUTURE ASSIGNMENT. IT DOES NOT ENTITLE ANY PERSON TO PREFERRED ASSIGNMENT.

A lieutenant read it and said, "So do well and maybe?"

His sergeant answered, "That is most of life, sir."

By December the twenty-four commands had replacement packets moving through them.

Some would take months to mature.

That was acceptable.

The CTCs had given the field army experienced people.

The field army had given the CTCs competition for the chance to become experienced.

For Hall, that exchange mattered almost as much as the two new LCTs.

A pipeline that emptied itself was consumption.

A pipeline that made people want to enter it was an institution.

Activation Problem

Training Command refused to let the new Borderers activate on paperwork alone.

The final validation was deliberately smaller than IRON MIRROR and more personal than Point Barrow.

Each LCT received seventy-two hours.

The 1st Avalon had to deploy from assembly areas, establish a screen, conduct a delaying action, displace its support base, and finish with enough combat power concentrated to strike a designated objective.

The 2nd Avalon received a different sequence built around route security, counter-raid operations and a rapid reinforcement of a threatened militia sector.

Neither exercise was designed to prove the doctrine.

The doctrine had already been tested.

The question was whether permanent people could make it ordinary.

The 1st Avalon discovered its first problem before crossing the start line.

The twelve OPFOR-derived BattleMech lances were too good at improvising around local friction.

That sounded like praise until three companies independently altered their loading order to solve three small problems and created one large one for the transport control officer.

David was not yet the formation commander, but Training Command had invited him to observe because the promotion board had not completed its work and everyone involved understood the likely direction without pretending it was decided.

He watched three lance leaders explain why each change had made sense.

All three were correct.

The movement plan was still broken.

The provisional LCT commander listened to them, then said, "You are all very experienced at being the cleverest person in somebody else's exercise."

No one spoke.

"Now learn to be part of my movement table."

The loading order returned to one plan.

The LCT crossed the line forty-one minutes late.

It recovered the time by the end of the first day because the same experienced personnel who had created the problem knew how to operate inside changed conditions once the command decision was clear.

That became the pattern.

The 1st Avalon made fewer novice mistakes than most new formations.

It made more expert mistakes.

People anticipated.

People improvised.

People occasionally improved a plan that did not need improvement.

The headquarters learned to give wide freedom where the mission benefited and narrow control where synchronization mattered.

The artillery fire plan was not a suggestion.

The withdrawal window was not a suggestion.

A scout route could change if contact changed.

A refueling schedule could not change merely because a lance commander disliked waiting eight minutes.

By the second day, the distinction was becoming visible.

An OPFOR-derived cavalry officer found an unexpected ford and reported it.

The battalion commander approved the route change immediately because it shortened movement without disrupting anyone else.

Three hours later, a BattleMech company requested permission to pursue a retreating exercise enemy beyond its phase line.

Denied.

The company stopped.

No argument.

The 1st Avalon completed the final strike with seventeen minutes remaining.

Not spectacular.

Enough.

The 2nd Avalon did better on time and worse on casualties.

Its headquarters moved beautifully.

One armor battalion did not.

The battalion became fixed in a fight it should have bypassed, and the LCT commander had to commit part of his reserve to extract it.

The mission still succeeded.

The AAR did not let success erase the cost.

The 2nd's commander looked at the casualty board.

"We passed."

The observer nodded.

"Yes."

"That sounded disappointed."

"You are fully activated. I am now allowed to expect you to be better next month."

The commander laughed.

"That seems unfair."

"You have been listening to the Dragonlords."

Both Avalon LCTs passed.

Training Command entered the results without adjectives.

1ST AVALON BORDERERS LCT - ACTIVATED. REGULAR.

2ND AVALON BORDERERS LCT - ACTIVATED. REGULAR.

The words looked almost too simple for the amount of infrastructure behind them.

Twenty-four OPFOR commands had contributed people without being gutted.

Twelve CTCs had accepted replacement burdens and trained into them.

Production and reserve depots had supplied machines without pretending stored equipment was already a unit.

Infantry schools had filled companies.

Artillery schools had produced crews.

Signals and logistics had done the work that photographs would not show.

Two formations appeared on the rolls.

The real achievement was that the system that produced them still existed afterward.

Two Borderers

The 1st and 2nd Avalon Borderers activated fully in December.

That did not mean perfect.

It meant their principal combat, support and command elements existed; their equipment was present at combat-usable levels; their logistics and communications had passed activation standards; and their headquarters could command the formations they actually possessed.

Both were assessed Regular at activation.

That was unusual for new formations.

Training Command documented why.

The experienced OPFOR core had shortened integration time.

The CTC volunteers were not individually magical.

They were simply accustomed to joining things.

They already understood that an attachment arrived with a commander, a maintenance problem, a communications requirement, a fuel requirement and a different idea of what the word "soon" meant.

New graduates and recruits still needed experience.

The veterans gave them a structure in which to gain it.

The two Borderers also developed different personalities almost immediately.

The 2nd Avalon favored sharper central planning and very fast staff work.

Its commanding general liked having three branches prepared before the first order was issued.

The 1st Avalon became noisier.

David had not yet taken permanent command, but the officers selected for it had already internalized too much of the CTC culture to remain quiet around a bad idea.

Meetings contained arguments.

They also ended with decisions.

Hall approved both.

"Same doctrine," he told Ian. "Different nervous systems."

Ian smiled.

"Good."

The 1st Bromhead Borderers remained cadre.

That status was printed openly.

Cadre.

Not provisional regiment.

Not forming combat team.

Not "nearly ready."

Cadre.

Experienced leaders, instructors, selected specialists and the beginning of an organization that would eventually deserve a larger name.

One political aide suggested that announcing three Borderer formations would sound stronger than announcing two and a cadre.

Ian looked at him.

"Would the third be stronger?"

"No, Highness."

"Then why would I want to sound stronger than I am?"

The aide did not answer.

The question was not rhetorical.

That was why it hurt.

Six Waylands

Wolf's Dragoons received their six Waylands before the end of November.

Joshua Wolf inspected all six personally.

David found him halfway inside an open maintenance compartment.

"Still think thirty-six was reasonable?"

Joshua backed out.

"Yes."

"Of course you do."

"Twelve would have been diplomatic."

"Six was diplomatic."

"Eight would have been friendly."

"You are leaving for another employer."

Joshua smiled.

"Exactly. You should want us to remember you fondly."

David looked at the six machines.

"Do you?"

Joshua's smile faded into something more serious.

"Yes."

That ended the bargaining.

The Heavy MHQ transferred separately.

Dragoon technicians had spent months learning its systems before the final handover.

No hidden lockouts.

No remote controls.

No clever intelligence package buried inside.

The Dragoons checked anyway.

Ian would have been disappointed if they had not.

The departure orders accumulated.

Families packed.

Schools transferred records.

Civilian employees decided whether to follow the regiment or remain on New Valencia.

The Ministry of Mercenary Relations processed more questions than combat operations ever generated.

Pets.

Household goods.

Medical records.

Educational credits.

Final pay.

Property claims.

Replacement parts purchased under old contract authority but delivered after expiration.

One junior official complained that none of it felt like military history.

His supervisor looked at the stack.

"That's because history usually starts after someone else finished the paperwork."

The six machines did not arrive naked.

Each came with a support package negotiated separately from the vehicle itself: initial spares, fabrication feedstock, diagnostic references, operator conversion material, and enough cross-training time that the Dragoons would not spend their first month treating the Waylands like expensive cargo.

Joshua noticed.

"You could have counted the machines and stopped."

David looked at him.

"Then you would own six problems with wheels."

One Dragoon technician corrected him from beneath an open side panel.

"Tracks."

David leaned down.

"Six problems with tracks."

"Better."

The training package mattered because the Dragoons would be moving into a different supply relationship. AFFS depots would no longer be the automatic answer when a Wayland needed something difficult. The machines therefore had to leave with enough documentation and enough trained people to remain Dragoon assets rather than permanent reminders of Davion support.

Kara insisted on one final condition that was not a contract condition at all.

"If you break one in a way we have not seen before, send us the report."

Joshua smiled.

"Free engineering data?"

"Mutual self-interest. We built the things. I would prefer to learn how they fail before one fails the same way under someone else's people."

That arrangement required no signature.

Technicians understood it immediately.

December - Five Flags and One Cadre

December did what December had begun doing in the Federated Suns.

It made planning visible.

The 5th Dragonlords activated.

The 9th Deneb Light Cavalry returned to the rolls.

The 1st Avalon Borderers activated.

The 2nd Avalon Borderers activated.

The 1st Bromhead Borderers became an official cadre.

None of those statements meant the same thing.

The 5th Dragonlords possessed its command structure, principal combat formations and enough equipment to exist as an RCT. It still owed Training Command the hard part. Novaya waited.

The 9th Deneb Light Cavalry had lineage, personnel, command and a restored identity, but still required the ordinary work that every re-formed regiment required before history stopped doing the carrying for it.

The Avalon Borderers were different.

Their principal formations were present.

Their staffs had passed activation exercises.

Their combined-arms nets worked.

Their experienced cores had spent years at CTCs learning how quickly a formation could fail when everyone assumed integration was somebody else's responsibility.

They were not Veteran formations.

They had not earned that word.

They were Regular, fully activated Light Combat Teams with an unusual amount of institutional experience packed into the people who taught everyone else how to fight unfamiliar opponents.

The Bromhead cadre had the easiest status to explain.

It was not ready.

Everyone knew.

That honesty was part of the strength.

Foreign intelligence saw five new entries.

The AFFS saw five different readiness problems.

That difference explained much of 3010.

Three Promotion Packets

Ian saw David's promotion recommendation only after the board had completed most of its work.

He had arranged that deliberately.

Not hidden it.

Not interfered with it.

Stayed out.

The packet began in Training Command.

Then Operations.

Logistics comments.

CTC reports.

B-3000 evaluation.

IRON MIRROR.

​

Unit assessments.

Professional endorsements.

Professional objections.

Ian read those carefully.

One officer described Colonel Davion as:

OPERATIONALLY INFURIATING BUT CONSISTENTLY FOCUSED ON MISSION OUTCOME OVER PERSONAL TACTICAL SUCCESS.

​

Ian smiled.

Hanse read over his shoulder.

"Who wrote that?"

"Heavy Guards."

"That is almost affectionate."

"For them."

Another endorsement came from an officer who openly disliked David.

That mattered more.

The board had debated.

Also important.

One member had questioned whether David's proximity to the ruling family created an appearance problem.

Another had replied that refusing promotion because of the surname would be no more professional than granting it because of the surname.

Ian stopped reading.

Hanse watched him.

"You knew."

"I knew the packet existed."

"That isn't what I meant."

Ian closed it.

"No."

Hanse smiled.

"You're proud."

"Yes."

"Because he's getting promoted."

Ian shook his head.

"Because I had nothing to do with it."

That silenced Hanse.

The final recommendation read:

BRIGADIER GENERAL, DAVID DAVION.

​

Assignment:

COMMANDING GENERAL, 1ST AVALON BORDERERS LIGHT COMBAT TEAM.

​

Ian reopened the packet.

Not because he doubted it.

Because he wanted to see how they had reached the answer.

The reasons were not romantic.

Operational planning.

Training work.

Command-system development.

Performance under independent exercise authority.

Ability to integrate combined arms.

Repeated evidence that he developed subordinates rather than collecting their decisions.

Readiness.

The promotion board cared about readiness.

Ian did too.

Jasmine's packet came through another chain.

Combat arms.

OPFOR performance.

Command evaluations.

Training results.

Independent field assessments.

COLONEL, JASMINE RASHID.

​

Kara's came through logistics and technical command.

Maintenance.

Production integration.

Strategic Refit Center work.

Training systems.

Fleet and ground support.

Industrial standardization.

LEFTENANT COLONEL, KARA O'SULLIVAN.

​

The packets did not arrive together.

No one had coordinated them.

That pleased Ian almost as much as David's.

Three careers.

Three chains.

Three answers.

The family connection existed.

So did the work.

Jasmine received her news from Hall.

He handed her the promotion order after an OPFOR debrief.

She read it once.

Then looked up.

"Colonel."

"Yes."

"I thought you were about to tell me the exercise design was too expensive."

"It is."

"Then this is confusing."

Hall smiled.

"You can be promoted and still spend too much ammunition."

Jasmine looked back at the order.

"Does David know?"

"Not from me."

"Good."

Hall raised an eyebrow.

"You want to tell him?"

"I want to see whether his packet arrived first so I can be insufferable in the correct direction."

Hall considered that.

"Professional development."

"Exactly."

Kara received hers in a maintenance meeting.

That seemed appropriate.

The message arrived while three engineers argued about whether a revised actuator inspection interval should be mandatory or conditional by duty cycle.

Kara read the order.

The engineers waited.

One finally asked, "Ma'am?"

Kara looked up.

"Apparently I am a Leftenant Colonel now."

The youngest engineer smiled.

"Congratulations."

"Thank you. Your inspection interval is still wrong."

The meeting continued.

Someone later asked whether she wanted a formal celebration.

Kara said yes.

She was not against celebrations.

She was against using them to stop work that remained wrong.

Ian also checked what the three packets did not contain.

No recommendation cited marriage.

No board member argued that promoting one required promoting the others.

No staff officer had tried to synchronize dates so the announcements would photograph well together.

That mattered because the easiest way to cheapen three earned careers would have been to turn them into one family event.

Jasmine's file contained almost nothing about David beyond exercises in which both names necessarily appeared.

Kara's contained even less.

Her strongest endorsements came from people who cared about maintenance hours, production tolerances, technical training and the uncomfortable habit she had of asking whether a new system could still be repaired by the people expected to use it.

One senior logistics officer wrote:

MAJOR O'SULLIVAN HAS AN UNHELPFUL TENDENCY TO DISCOVER THAT OUR MOST ELEGANT SOLUTIONS REQUIRE TECHNICIANS WE DO NOT POSSESS. THIS TENDENCY SHOULD BE ENCOURAGED.

Hanse read the sentence twice.

"Apparently operationally infuriating is contagious."

Ian smiled.

"Different disease."

Jasmine's objections were different. One evaluator considered her too willing to let subordinate commanders make mistakes during exercises instead of intervening early.

Hall's response was attached beneath it.

THAT IS WHAT THE EXERCISE IS FOR.

The board retained both comments.

Good promotion packets were not praise collections.

They were arguments preserved honestly enough that the approving authority could see what had actually been considered.

Ian signed each recommendation separately.

That was how he wanted the careers remembered.

Brigadier

David entered Ian's office expecting a discussion about the 1st Avalon Borderers.

That was technically what he received.

Hanse was there.

David immediately became suspicious.

"What did he do?"

Hanse looked offended.

"Why am I always the criminal?"

"Experience."

Ian pointed toward the chair.

"Sit."

David sat.

Ian pushed the packet across the desk.

David read the first page.

Stopped.

Read it again.

Hanse was smiling.

David looked at him.

"You knew."

"Recently."

David looked at Ian.

Ian said, "I stayed out."

David looked back at the paper.

The endorsements were visible.

Hall.

Operations.

Training.

Logistics.

Command Systems Development Group.

Several commanders.

One name made him blink.

"He doesn't like me."

Ian smiled.

"I noticed."

David read the comment.

"Operationally infuriating."

Hanse laughed.

"That may be my favorite."

David ignored him.

The rank line remained.

Brigadier General.

He looked toward Ian.

"You're giving me a brigade."

"Recommending."

Hanse rolled his eyes.

Ian continued.

"The distinction matters."

David knew.

The 1st Avalon Borderers was not a ceremonial command.

It was the first permanent independent field formation that would be his.

No exercise controller.

No senior commander waiting above him to catch the error.

No borrowed authority.

Ian watched the realization arrive.

"You've commanded pieces."

"Yes."

"Exercises."

"Yes."

"Staff."

"Yes."

"Operational groupings."

"Yes."

Ian tapped the order.

"This is yours."

David looked at the organization.

BattleMechs.

Armor.

Infantry.

Artillery.

Aerospace.

Engineers.

Logistics.

People.

Lots of people.

Hanse leaned against the desk.

"Try not to look terrified."

"I am deciding whether hitting you would affect the promotion."

"It might."

Ian almost smiled.

Then became serious.

"Not because your name is Davion."

David looked toward him.

"And not because you're my cousin."

"I know."

"I need you to hear it anyway."

David nodded.

Ian continued.

"Some people will say it."

"I know."

"They would say it if you remained a colonel for ten years."

"Yes."

"The people whose opinions matter know how this packet was built."

David looked at the endorsements again.

Ian said, "So do you."

Silence.

Then David asked:

"When do I take command?"

Hanse laughed.

Ian smiled.

"That is why you're getting it."

David stood.

Then stopped.

Ian stood too.

For a moment they were not First Prince and newly promoted general.

They were cousins who had inherited different pieces of Andrew's unfinished work.

Ian embraced him.

Briefly.

Firmly.

Then stepped back.

"You're allowed to say it now."

David looked confused.

"Say what?"

"I commanded."

David understood.

The smile came slowly.

"Yes."

Hanse immediately ruined the moment.

"Brigadier General Davion."

David looked at him.

"Don't."

"General."

"Hanse."

"Sir."

Ian laughed.

It was worth it.

The 1st Avalon Borderers

The command ceremony occurred three days later.

David hated the speech.

Not the words.

Having to give one.

The formation stood before him in blocks representing pieces that still had more work to do.

BattleMech crews.

Armor.

Infantry.

Artillery.

Aerospace representatives.

Engineers.

Signals.

Medical.

Logistics.

The first thing David noticed was how many people were not MechWarriors.

That pleased him.

He looked at the prepared speech.

Then folded it.

Hanse, watching from beside Ian, whispered, "There goes Public Affairs."

Ian did not answer.

David spoke without amplification for the first sentence, then remembered the formation was too large for that and used the microphone.

"You have been told we are a light combat team."

A pause.

"Some of you have also seen the equipment list and concluded that someone in Personnel has a sense of humor."

A few smiles.

"Light does not mean weak. It does not mean disposable. It does not mean we are the formation someone sends because an RCT is too valuable."

He looked across the ranks.

"Our job is to make time expensive."

That quieted them.

"If an enemy ignores us, we hurt him. If he chases us, we delay him. If he concentrates enough force to destroy us, we move. If the mission requires us to hold, we hold until holding no longer serves the mission."

He let the next words sit.

"Delay is not martyrdom."

Hall watched from the reviewing stand and did not smile.

He did not need to.

David continued.

"You will be judged by what the enemy fails to accomplish on schedule. You will also be judged by whether your people are alive to do it again."

He looked toward the logistics battalion.

"And before anyone gets excited about raiding enemy supply lines, remember that we also have one."

That got the laugh.

The speech ended shortly afterward.

Public Affairs complained that it was too short.

David considered that evidence of success.

The First Staff Meeting

The first staff meeting occurred before the ceremony flowers had been removed.

David preferred it that way.

The 1st Avalon headquarters still smelled new in the irritating sense that half the furniture had not yet acquired scratches and everyone was too aware of where they were supposed to sit.

The operations officer began with a prepared command brief.

David stopped him on the second slide.

"I have read the organization chart."

The officer hesitated.

"Yes, sir."

"Tell me what you think will break first."

Silence.

Then the logistics officer said, "Movement scheduling."

The signals officer said, "Cross-attachment communications."

The artillery commander said, "Fire-control handoffs when we split the LCT."

The senior infantry officer said, "People assuming light means we can move through roads that do not exist."

The chief maintenance officer said, "All of them are wrong. Recovery."

David looked around the table.

"Good."

The operations officer slowly closed the prepared briefing.

They spent two hours on failure instead.

That told David more about the command than the activation report had.

The OPFOR veterans were obvious without needing to identify themselves. They were the people most comfortable saying they did not know how a new attachment would behave until they exercised with it. Several had arrived from different CTCs with different habits and strong opinions about whose range produced the most irritating aggressors.

David prohibited that argument only after it consumed seventeen minutes.

"You may resume after we beat someone else."

A major from one of the former OPFOR commands asked the question David had expected eventually.

"Sir, how much freedom do you intend to give subordinate commanders?"

David looked at him.

"Enough to make me uncomfortable."

A few people smiled.

He continued.

"If you know my intent, your boundaries, your support state and the risks you are authorized to take, act. If you need something outside those limits, ask. If the situation changes so badly that the original intent becomes stupid, tell me before you heroically obey it."

The artillery commander raised a hand slightly.

"And if we make the wrong decision inside those limits?"

"Then I am still the commanding general."

That quieted the room.

David pointed toward the first exercise schedule.

"I will correct bad judgment. I will not make you borrow my judgment because you are afraid of being blamed."

The command sergeant major, who had said almost nothing, finally spoke.

"That will take practice, sir."

"Yes."

"For you too."

David looked at him.

Then nodded.

"Especially for me."

The first ninety-day plan changed that afternoon.

More exercises were designed with separated task groups instead of keeping the entire LCT within convenient command range. Communications would fail by design. Attachments would arrive late. Logistics packages would be switched between battalions. One battalion commander would periodically receive a mission with less information than headquarters possessed and be expected to proceed within intent rather than wait for perfect clarity.

David added one restriction of his own.

No exercise could be declared successful solely because the designated target was destroyed.

The force had to return.

Casualties had to be evacuated.

Disabled equipment had to be recovered or deliberately abandoned under the exercise conditions.

Ammunition expenditure had to remain supportable.

The next mission had to remain possible.

The operations officer read the additions.

"You are making the withdrawal part of the objective."

"It is part of the objective."

That lesson had been growing through the chapter of his life for years.

At Galtor it would become the difference between raiding and volunteering to be surrounded.

Ninety Days

David received the 1st Avalon Borderers operational workup order after the activation ceremony.

Ninety days.

Not because the formation was uncertified.

Because certification answered whether the LCT could function.

It did not answer whether David and the LCT could function together.

Hall had no intention of learning that under Kurita fire.

The first three months of 3011 were therefore blocked for field exercises.

No ceremonial tour.

No palace obligations beyond what could not be escaped.

No assumption that David's command of provisional OPFOR formations during IRON MIRROR made permanent command automatic.

The 1st Avalon would run movement exercises, live-fire work, night problems, strategic loading, unloading, delay missions, raids, withdrawals, casualty replacements and enough communications failures to make the signals battalion hate Training Command personally.

David read the schedule.

"Three months."

Hall nodded.

"You asked for a field command."

"I did."

"This is the field part."

David turned the page.

The prospective operational area appeared.

The Draconis March.

The Galtor Thumb.

The map did not contain an invasion arrow.

It contained defensive belts.

Robinson Chevalier RCTs.

March Militia positions.

Established Borderer commands.

Planetary defenses.

Those forces would hold the doors.

The 1st Avalon Borderers would be one of the things allowed to move through them.

David looked at Hall.

"Offensive raids."

"If you pass the workup."

"Hit supply points. Communications. Movement nodes. Forces that think the border protects them."

"And leave before the Combine can make you pay for staying."

David studied the routes.

The idea was not conquest.

It was pressure.

Borderers and Chevaliers would make the Davion side hard to enter.

The 1st Avalon would make the Kurita side uncomfortable to stand near.

A line from IRON MIRROR returned to him.

Make time expensive.

Across a border, the same doctrine could make certainty expensive too.

Hall let him study the map for another minute.

Then he placed a second sheet beside it.

Rules of engagement.

David read them.

Military objectives.

Civilian infrastructure protected unless directly used for military purpose and no reasonable alternative existed.

Medical sites protected.

Surrender accepted.

Prisoners accounted for.

Enemy dead recovered where practicable and returned through recognized channels.

No false surrender.

No misuse of medical markings.

No promise of safe passage unless the commander intended to honor it.

David looked up.

"I know."

Hall nodded.

"I know you know."

"Then why give it to me?"

"Because next year you will cross the border with your name painted on the headquarters vehicles."

David understood.

Hall continued.

"The enemy will learn what David Davion means from what you do when you have the advantage."

That landed harder than the promotion had.

David looked back at the map.

"Then they learn the truth."

Hall studied him.

"Make sure it is a truth you can live with in 3014."

David looked at him sharply.

Hall did not explain the year.

He did not need prophecy to know reputations accumulated before the battles that made them famous.

The orders went into David's case.

The Galtor Thumb waited on the other side of New Year's Day.

The Galtor Problem

The operational concept behind the ninety-day workup was discussed two nights later in a smaller room.

Ian.

Hanse.

Hall.

David.

A Draconis March operations officer.

The map covered the Galtor Thumb and the approaches around it.

The operations officer began with the defensive picture.

Robinson Chevalier RCTs anchored the heavier response positions.

March Militia commands and established Borderer formations covered local routes, warning nets and the places Kurita raiders had learned to probe repeatedly.

None of those forces were passive.

They patrolled.

They counter-raided when appropriate.

They trained.

But their first obligation was to keep Davion worlds from becoming opportunities.

David looked at the 1st Avalon marker.

"So you do not want me sitting on a planet."

The operations officer shook his head.

"Not unless the situation changes."

Hanse leaned over the map.

"The point of strengthening the line was to stop every mobile formation from becoming a garrison."

That had been one of the quiet consequences of twenty years of militia and Borderer reform.

Local defense did more than protect local worlds.

It released regular formations to move.

The 1st Avalon was built to move.

Hall pointed across the border.

"Your 3011 problem is to make them react to you without giving them something worth trapping."

David studied likely targets.

Supply depots.

Transit points.

Communications facilities.

Repair sites.

Patrol concentrations.

Units preparing raids of their own.

He did not point at cities.

Ian noticed.

"No trophies," the First Prince said.

"No, Highness."

"No staying because the fight becomes interesting."

David glanced at Hall.

"I have been trained out of that."

Hall answered, "We have made progress."

Hanse smiled.

The operations officer changed the display to show likely Combine reaction times.

The geometry was the entire campaign.

Hit.

Force a response.

Leave.

Appear somewhere else before the enemy knew whether the first raid had been the main effort.

The 1st Avalon could not conquer the Thumb.

It could make the DCMS spend transport, fuel, patrol time and attention defending it.

The defensive line behind David mattered as much as the raiding force ahead of it.

If Kurita retaliated, Chevaliers, Borderers and March Militia would be waiting on prepared ground.

If Kurita stayed defensive, David kept the initiative.

If Kurita concentrated enough force to catch him, another sector became thinner.

Ian looked at the map for a long time.

"And if he gets greedy?"

David asked, "Kurita or me?"

"Yes."

That was why the ninety days came first.

The 1st Avalon would practice the hardest part of raiding before firing a real shot.

Leaving.

Not after defeat.

After success.

Hall wrote the training requirement into the workup order.

EVERY RAID ENDS WITH A WITHDRAWAL EXERCISE.

David read it.

"You really do know me."

"I know commanders."

The map remained on the table after the meeting ended.

The Galtor Thumb looked quiet.

That quiet was exactly why the assignment mattered. A border that looked stable could still consume initiative one raid at a time. David would spend the winter learning whether the 1st Avalon could cross it, strike, disengage, and return with its people, its discipline, and its word intact.

It would not stay that way.

A Reputation Before a Battle

David took the rules-of-engagement sheet home.

Kara found it beside the Galtor map.

"You are reading this like a maintenance manual."

"It is one."

"For what?"

David looked at the list again.

"Reputation."

Kara sat across from him.

He pointed to the line concerning safe passage.

"If I tell a Kurita commander he can withdraw wounded through a corridor, I have to mean it."

"Yes."

"Even if I could use the corridor to trap him later."

"Yes."

"If we accept surrender, we account for prisoners."

"Yes."

"If we say a hospital is protected, then some idiot with an artillery map does not get to decide the word protected means until inconvenient."

Kara watched him.

"You already believe all of that."

"That is different from making thousands of people believe I mean it."

She understood.

The first raid could establish a pattern.

So could the fifth.

The dangerous one would be the raid where keeping the pattern cost something.

David looked toward the Galtor Thumb.

"Kurita respects strength."

"So do most soldiers."

"They also remember dishonor."

Kara leaned back.

"Then do not give them any."

That sounded simple.

Most hard rules did before they became expensive.

David folded the sheet and put it into the same case as the operational orders.

He did not know Takashi Kurita would one day read reports carrying his name.

He did not need to.

The correct reason to keep a promise was that someone had trusted it.

Anything else was strategy afterward.

First Circuit

The December ROM report was shorter than the June reconstruction.

That made it more dangerous.

The analysts had stopped trying to identify the program responsible.

There were too many.

One section compared formation growth with support growth.

Another compared factory expansion with technical-school output.

Another tracked militia rotations through Combat Training Centers.

Another examined the B-series.

Another traced refurbished machines whose combat capability exceeded assumptions based on age.

Another tracked Blue Water carrier development beside Sentinel expansion.

Another followed mercenary retention and recruitment.

Another studied the fact that Davion programs increasingly produced second-order programs.

A Regional Training Battalion created instructors.

Those instructors created more trained personnel.

Standardized components shortened repair time.

Shorter repair time increased equipment availability.

Higher equipment availability increased training opportunity.

Training improved operational performance.

Operational reports changed design standards.

Design standards returned to factories.

The system fed itself.

Julian Tiepolo read the summary.

Then asked the question others had already asked.

"Who replaced Andrew?"

The ROM officer had prepared names.

Tiepolo saw the hesitation.

"No."

The officer stopped.

Tiepolo set the report down.

"Then stop giving me names."

Silence.

"What replaced him?"

The answer took longer.

Finally:

"Institutions."

Tiepolo looked back at the report.

That was harder to kill.

Harder to bribe.

Harder to discredit.

Harder to interrupt.

Not impossible.

Nothing human was impossible to damage.

But the problem had changed.

Assassinating a prince could disrupt a government.

It could not assassinate twelve Combat Training Centers.

It could not assassinate standardized wiring.

It could not assassinate every instructor.

Every technician.

Every depot.

Every rebuilt factory.

Every staff officer who had learned to mark a unit orange instead of lying about yellow.

ComStar had spent decades watching for recovery of technology.

The Federated Suns was recovering something less dramatic.

Competence.

That might be worse.

Tiepolo turned one page back.

"What can be copied?"

The ROM officer answered more confidently.

"Most of the individual programs."

"And the whole?"

"Not quickly."

"Why?"

"Because the whole is twenty-one years old."

Tiepolo closed the report.

That was the sentence he remembered.

The Federated Suns did not possess twenty-one years of technological superiority.

It possessed twenty-one years of accumulated habits.

Those were harder to steal in a crate.

Another adept offered the obvious response.

"Interrupt the linkages."

Tiepolo looked at him.

"Which ones?"

The adept opened the annex.

Transport first. Delay machine tools. Interfere with technical recruiting. Encourage regional rivalries over training allocations. Increase the cost of mercenary contracts. Exploit resentment when experienced personnel were transferred to instructor duty. Disrupt standardization by encouraging local procurement. Slow communications where possible without revealing intent.

Each proposal was plausible.

None was sufficient.

The ROM officer said so carefully.

"A transport disruption would hurt movement."

"For how long?"

"Until they reroute or accept delay."

"A factory?"

"They increasingly have multiple sources for mundane components."

"An instructor pipeline?"

"Useful target. Also difficult to strike broadly without becoming obvious."

Tiepolo looked at the page.

"A prince?"

No one answered immediately.

The answer was already in the report.

Andrew had died.

The system had continued.

That did not make leadership irrelevant. Ian mattered. Hanse mattered. David and Hall and hundreds of other people mattered. Remove enough competent people and any institution degraded.

But there was no longer one person whose disappearance returned the Federated Suns to 2989.

That was the part ROM disliked.

A centralized enemy offered leverage.

A distributed one offered work.

Tiepolo closed the intervention annex.

"Do not mistake difficulty for invulnerability."

"No, Primus."

"And do not mistake visibility for understanding. They are publishing half of this in procurement notices and training schedules because they do not think the dangerous part is secret."

The ROM officer nodded.

The dangerous part was repetition.

A technician teaching another technician.

An OPFOR lieutenant becoming a battalion commander.

A militia unit leaving its homeworld, failing on a range, returning, and teaching the next rotation what had hurt.

A carrier deck crew learning one procedure and exporting it to an airfield.

None of those required a hidden vault.

Tiepolo looked once more at the twenty-one-year timeline.

"Continue collection."

The officer waited for more.

There was none.

Sometimes the most alarming conclusion an intelligence service could reach was that the target did not possess a single secret worth stealing.

The Capacity Review

The final meeting of 3010 occurred without ceremony.

Ian wanted the 3011 production and formation plan.

He received six versions.

That annoyed him.

He asked why.

The answer was capacity.

The AFFS could raise more formations.

Probably.

The industrial system could equip them.

Probably.

The schools could provide officers.

Probably.

The NCO pipeline could stretch.

Probably.

Training Command could increase throughput.

Probably.

Transport could handle additional movement.

Probably.

Technical services could support another acceleration.

Probably.

Ian listened.

Then looked at Hall.

"Can we raise more next year?"

Hall answered.

"Probably."

"Can we train them properly?"

A pause.

"Probably."

Ian looked toward Logistics.

"Can we sustain them?"

Another pause.

"Probably."

Transport.

"Move them?"

"Probably."

Technical Command.

"Maintain them?"

"Probably."

Ian closed the folder.

"Then we're not ready."

No one argued.

Twenty-one years earlier, that answer would have sounded like weakness.

In 3010 it meant something different.

The Federated Suns had become strong enough to recognize the shape of overreach before stepping into it.

Hanse looked at the closed folder.

"So what do we do?"

Ian answered.

"Make probably become yes."

That produced the next year's work.

More instructors.

More technical throughput.

More transport margin.

More ammunition stock.

More maintenance depth.

Not another flag because someone wanted to end the meeting with one.

Before the meeting ended, Finance asked the question everyone else had politely avoided.

"How do we know when the answer becomes yes?"

Ian looked at the capacity board again.

Hall answered first.

"When adding one more formation does not require us to quietly lower the standard somewhere else."

Logistics added, "When its ammunition allocation is real rather than borrowed from an emergency reserve we still describe as available."

Training Command said, "When instructors can be assigned without turning the schools behind them into cadres."

Transport said, "When movement can be scheduled without assuming no other crisis occurs at the same time."

Technical Command said, "When maintenance supervisors exist before the machines arrive."

Ian listened to each answer.

"So readiness is margin."

David, newly aware of how much margin the 1st Avalon would consume simply by exercising hard for ninety days, nodded.

"Increasingly."

That was another difference from 2989.

Then, strength had often meant using everything because there was nothing behind it.

Now the AFFS was beginning to understand that a force with no margin was already spending tomorrow.

That did not mean the next year would be cautious in the ordinary sense. It meant growth would follow the bottleneck instead of pretending the bottleneck had disappeared. The objective was not fewer ambitions. It was enough depth that ambition did not have to borrow readiness from somewhere else.

Outside the palace, factories continued working.

Students studied.

Technicians rebuilt machines.

The 1st Avalon Borderers received a ninety-day operational workup schedule under its new commanding general, followed by a prospective Draconis March deployment if the formation met the standard.

The 2nd Avalon Borderers began its own field workup under a different commander, proving that common doctrine did not require identical solutions.

The 1st Bromhead cadre trained people who would eventually train more.

The 5th Dragonlords prepared for the pipeline that would tell it what its activation ceremony had not.

The 9th Deneb Light Cavalry wore its own colors again.

UrbanMechs waited in militia hangars beside infantry plans for battles everyone hoped never reached a city.

Tiger IIs rolled through conventional divisions and planetary militias in numbers no intelligence service had originally considered glamorous enough to fear.

Essex worked through another flight cycle while her task-force staff argued about screen geometry.

A Buckler deployment package disappeared inside modified Mammoths because a carrier that could move between worlds was useful only if the Navy could actually move it.

The B-3000 team argued about which alerts deserved a general's attention.

At Point Barrow, the training ranges emptied behind the 2nd Dragonlords as the RCT returned to Lindsay with the same name and a different meaning.

On Sirdar, crews recovered what could be recovered from Quarry Junction and wrote reports about what could not.

On Northwind, Highlanders came home without being immediately turned into someone else's solution.

On New Valencia, Wolf's Dragoons prepared to leave.

Nothing about the year resembled a decisive victory.

No capital had fallen.

No enemy army had been annihilated.

No miracle technology had transformed the balance of power.

The strongest new BattleMech in the Federated Suns was not the story.

Neither was the newest tank.

Neither was the carrier.

Neither was the command computer.

Those were things.

The shape beneath them was harder to see.

A militia that understood delay.

A commander who understood when not to pursue.

A technician who could replace the same connector on three different machines.

An instructor who told an Avalon Hussars company that four UrbanMechs had never been its real problem.

A carrier commander who learned that winning the first raid meant nothing if the second one reached the flight deck.

A logistics officer who could say no.

A readiness board willing to write orange.

A prince willing to believe it.

Andrew Davion had begun by asking whether the Federated Suns could keep what it already possessed alive.

By 3010, the answer was increasingly yes.

The next question was larger.

How much strength could the Federation create without destroying the systems that created it?

For the first time in generations, that was the problem.

And Ian Davion was wise enough to be afraid of the answer.