# **Chapter 47: Deep Strike**

  Days two and three followed the same rhythm.

  The Oirats came at dawn, probed through the morning, absorbed casualties at a three-to-one ratio, and withdrew before dark. Not because they were broken — their discipline held impressively through each engagement — but because steppe cavalry doctrine understood the mathematics of siege: you didn't waste your best fighters in unsustainable attrition. You probed until you found the seam, then you committed.

  Togrul was looking for the seam.

  Wei spent both evenings reviewing the day's engagement reports and adjusting. The heavy bolt allocation he'd ordered after day one proved correct — the Oirat heavy cavalry's armor was absorbing the standard weight at range, and the heavier gauge added meaningful penetration at two hundred paces. He also noticed, reading Captain Ren's after-action from Fort Huailai, that the Oirats had changed their harassment pattern on day two: longer duration suppression before the assault, more systematic rather than random, designed to degrade the rotation rhythm rather than simply disrupt it.

  They were adapting. Learning from each assault what the next one needed to look like.

  That was the problem with fighting a competent enemy — competence compounded. Every failed assault made the next one more informed. Wei was winning the exchange ratio but losing the information war: the Oirats were purchasing intelligence with each probe, and at some point that intelligence would produce an approach that didn't run into a wall.

  He said this to Zhang on the evening of day three.

  Zhang was reading the same reports. "So what changes?"

  "Nothing in our defense. If we adjust our response pattern to deny them information, we also degrade our own response effectiveness. The doctrine holds." Wei traced the map with one finger. "What changes is the timeline. They need four or five days of probing to find Qingshan's approach. We know that's coming. Our job is to make sure the network covering Qingshan is fully activated before they commit to it."

  "The mobile reserve."

  "Captain Liang's cavalry. Two hundred riders. They've been staged west of the main valley since day one — far enough back that the Oirat screening force can't see them, close enough to move on either the Huailai approach or the Qingshan secondary approach in under two hours." Wei looked at Zhang. "When Togrul pivots east, Liang moves. Not to engage cavalry in open terrain — that's a losing proposition. To hit the supply train."

  Zhang thought through it. "Togrul will have screened his supply train."

  "Yes. But his screening force has been oriented north and west — toward our main line. When he pivots east, his screen pivots with him. The supply train will be exposed on its western flank for a window of six to eight hours while he repositions."

  "That's a tight window."

  "Which is why Liang's orders are already written." Wei folded the map. "He knows where to go. He knows when. He knows to burn and withdraw, not to hold ground. The only variable is Togrul's timing."

  ---

  Day four arrived with different sounds.

  The Oirat forces were moving, but not toward Huailai. The reports came in from the eastern observation posts — Zhang's unauthorized towers proving their value, as Wei had privately noted they would — in quick succession. Cavalry columns turning east. Supply wagons repositioning. The characteristic dust cloud of a force changing axis of advance.

  "Enemy bypassing Fort Huailai," Zhao reported from the signal desk. "Moving toward the secondary valley. Eastern approach."

  "Alert Fort Qingshan. They're the new target." Wei was already marking the map. "Signal Liang. Execute supply interdiction."

  Qingshan had been anticipating this — Wei had briefed its commander, Captain Su, three days ago on the probability of an eastern pivot. Su was a steady officer, experienced enough to recognize that being the secondary target in a defensive network didn't mean being the weak point; it meant being the bait. The distinction was important for morale.

  The Oirat repositioning took most of the morning.

  Wei watched it through the observation reports, tracking the pivot with the particular attention of someone reading a battle the way you read a complex document — not for individual words but for the shape of the argument. The pivot was clean, professionally executed. No confusion, no delay, the kind of large-scale tactical adjustment that required months of training and a command structure that actually communicated.

  Togrul was running a professional force. Wei had known this since the initial engagement reports, but watching it move at scale made the knowledge concrete in a way that reports didn't fully convey.

  He found himself respecting it, which he'd learned years ago was the appropriate response to a competent enemy. Not fear. Not dismissal. Respect and attention.

  Fort Qingshan came into contact at midday.

  The engagement was different from Huailai — the fortification was smaller, lower profile, with one hundred fifty defenders against what the reports estimated as six hundred cavalry in the first wave. The fields of fire were narrower, which meant the Oirats' dispersal tactic was more effective; they could spread wide enough that the fort's arc of fire couldn't cover the full assault front.

  Captain Su held the first wave. The second wave found a penetration point on the northern wall — a section where the parapet height was lower than the eastern wall, a design compromise that Wei had accepted when resources didn't allow full construction. The Oirats had found it.

  Wei watched the signal flags from Qingshan change.

  If you stumble upon this narrative on Amazon, be aware that it has been stolen from Royal Road. Please report it.

  "Su is withdrawing northern wall to secondary position," Zhao reported.

  "Eastern wall?"

  "Holding. He's trading the northern section for the killzone."

  "Good. Let him work."

  The battle for Qingshan ran for three hours — longer than the Huailai engagement, messier, closer quarters, the kind of fighting that doctrine helped with but couldn't fully govern. Su made two decisions during those three hours that Wei hadn't anticipated and both were correct: pulling the northern wall troops earlier than the standard protocol specified when he recognized the secondary position covered the penetration better, and redirecting his remaining artillery to the approach route rather than the penetration point once the cavalry was already inside, denying reinforcement to the leading element.

  Neither decision was in the doctrine. Both emerged from understanding what the doctrine was trying to accomplish and applying that understanding to a situation the doctrine hadn't specifically addressed.

  Wei read the engagement reports that evening with the particular satisfaction of seeing something he'd built work the way it was supposed to work — not because his people had followed his instructions, but because they'd understood his intentions well enough to improvise correctly when the instructions ran out.

  ---

  The supply raid executed at dawn on day five.

  Wei wasn't there for it — Captain Liang operated independently once the signal to move was given, and command post micromanagement of a cavalry raid sixty *li* from the main position was how raids failed. He'd given Liang everything he needed: the objective, the timing window, the extraction route, the authority to abort if the situation changed materially. The execution was Liang's.

  The reports came back in pieces through the morning.

  Two hundred cavalry reached the Oirat supply train's western flank before the repositioned screening force covered it. The window Wei had estimated at six to eight hours had been closer to five — Togrul had been more thorough than anticipated in moving his screen — but five hours was enough. Liang's force burned thirty supply wagons, scattered the draft animals from twelve more, killed the supply unit's senior officer in the initial contact, and was withdrawing before the cavalry response could intercept.

  Casualties: eleven wounded, three dead.

  Thirty wagons burned.

  The Oirat offensive stalled within hours of the supply raid.

  Not from the raid alone — thirty wagons wasn't enough to collapse a force of four thousand cavalry's logistics. But the uncertainty it created was. Togrul's supply train had just been struck from a direction his screening force hadn't covered, which meant either his intelligence about Ming cavalry positions was wrong or there was an element of the Ming force he hadn't located. Either possibility required him to stop committing forces forward until he understood the threat to his rear.

  A cavalry force that couldn't trust its supply chain didn't advance. That was basic steppe doctrine.

  The probes continued on day five and day six, but lighter — reconnaissance rather than assault, testing rather than committing. The Oirats were trying to locate Liang's cavalry before resuming the offensive. Liang had already withdrawn to a secondary staging position and was running silent.

  On the evening of day six, the withdrawal began.

  Zhang brought the report from the eastern observation post. "They're pulling back. Estimate thirty *li* north. Main force consolidating."

  Wei looked at the cumulative map. Six days. Four thousand cavalry committed, the largest Oirat offensive since the initial siege at the beginning of his frontier command.

  "Final casualty estimate?"

  "Enemy: six hundred killed or seriously wounded, confirmed. Probably higher — their battlefield clearance was faster than we could track." Zhang set down the report. "Our losses: one hundred eighty across all positions."

  Three to one. The ratio had held through six days and four distinct engagement types — frontal assault, secondary approach, close-quarters fortification fighting, and a deep supply interdiction. The doctrine had been tested in every dimension Wei had designed it for.

  "Liang's cavalry?"

  "Fourteen total. Three dead, eleven wounded. All wounded returned to duty."

  Wei thought about that number. Fourteen casualties to eliminate thirty wagons and halt four thousand cavalry's offensive momentum. The arithmetic was stark enough to be almost uncomfortable.

  "Togrul made mistakes?" Zhang asked.

  "Two significant ones." Wei traced the map. "He committed to the Huailai assault too long before pivoting east — gave us three full days of engagement data and let us finalize Liang's positioning. A faster pivot on day two would have caught the supply interdiction window before Liang was fully staged." He moved his finger. "Second mistake was the screen reorientation. When he pivoted east, he moved his entire screening force to cover the new axis of advance. Standard doctrine — you screen the direction you're moving. But it left the supply train's western flank open for exactly the window I'd estimated."

  "Would you have made those mistakes?"

  Wei considered honestly. "The first one, probably not — I've fought defensive networks enough to know how fast they generate intelligence, and I'd have pivoted faster. The second one..." He paused. "Maybe. Standard doctrine says you screen the direction of advance. The deviation Liang exploited is only visible if you specifically account for a mobile cavalry reserve staged outside the main network. Togrul's intelligence didn't locate Liang's position, so the screen gap was a logical choice with the information he had."

  "So he did everything right with what he knew."

  "Yes. That's what makes this defensible. We didn't win because he made avoidable errors. We won because we had information he didn't." Wei closed the map. "He'll be back with better intelligence. That's what the thirty-*li* consolidation is for."

  Zhang was quiet for a moment. "He's not done."

  "No. This was a probe at operational scale — testing whether the fortification network was as strong as his intelligence suggested. Now he knows the answer." Wei looked north through the command post window. The dust of the Oirat withdrawal was visible at the horizon as a faint smear, barely distinguishable from weather. "He'll restructure. Better intelligence coverage. Different approach geometry. Possibly a larger force."

  "How long?"

  "Months. Maybe less if he moves supply stockpiles efficiently." Wei turned from the window. "We use that time. Reinforce Qingshan's northern wall. Adjust the artillery position at Huailai for the heavy bolt allocation. Develop a second mobile reserve — Liang's success will have told Togrul there's cavalry in the network; he'll screen against it next time."

  Zhang started writing.

  "And send a full engagement report to General Fang," Wei added. "Not a summary — full documentation. Every engagement, every decision, every casualty. The counter-offensive planning in the capital needs to understand what four thousand Oirat cavalry looks like in prepared defensive terrain before they start calculating how twenty thousand Ming troops will perform in open steppe."

  "You think they'll listen?"

  "I think the Emperor will. Minister Hu won't." Wei sat. "But the Emperor is the one whose decision matters. And the Emperor asked me for honest numbers." He picked up the first of the garrison after-action reports. "Make sure the numbers are in there."

  Zhang went to draft the dispatch.

  Wei began reading.

  The casualty ledger was on the corner of the desk, closed. He'd update it tonight — one hundred eighty names in total across six days, the ones from day one already entered, five more evenings to complete. He'd developed the practice of entering them at the end of each engagement day rather than letting them accumulate, because accumulation made the individual names disappear into aggregate.

  Each name was a specific person who had died executing a specific decision that Wei had made.

  Some of those decisions had been right. Some had probably been improvable in retrospect. All of them had been made with the information available at the time.

  He didn't believe in spending too long on the retrospective — it was a trap, the kind that paralyzed rather than improved. But he believed in looking directly at the cost of each decision long enough to carry it accurately, because commanders who stopped looking at the cost started making decisions as if the cost didn't matter.

  He opened the ledger.

  One hundred eighty.

  He began writing.

  ---

  **End of Chapter 47**