5 Armies Spent 3 Years Failing to Hold the Border. 40 SAS Sealed It in 2 Weeks.

Five armies spent three years on the border before the SAS arrived. The five were not a random collection of forces thrown at the problem without coordination. They were allied armies operating under a shared command structure with shared intelligence, shared logistics, and a shared assessment of what needed to happen.

 The border they were trying to hold was not particularly long. 43 km from the northern ridge down to the river crossing at the southern end. And the terrain, while difficult, was terrain that professional armies had operated in successfully in other contexts. Three of the five countries represented had recent operational experience in comparable environments.

 Two of them had equipment specifically designed for mountain border control. All five had significantly more than 40 men. After 3 years, the border was not held. Infiltration rates had declined in certain sectors at certain periods and individual units had achieved local successes that appeared in their afteraction reports as genuine progress.

But the border as a whole was not sealed and the pattern of failure. The cycles of apparent progress followed by resumed infiltration as soon as units rotated out had been consistent enough for long enough that the command structure overseeing the operation had begun to explore the possibility that the problem was structural rather than operational.

The SAS assessment team that arrived 6 weeks before the regiment’s deployment did not produce a written analysis of what the five armies had done wrong. That was not what they had been sent to produce. They produced an assessment of the terrain, the crossing points, the patterns of movement, and the specific locations where infiltration was occurring most persistently despite the presence of conventional forces.

 The assessment was four pages long. What it described without framing it as a critique of any of the five armies was that the crossing points being used were not the crossing points being watched. That the infiltrators had adapted to the presence of the conventional forces by identifying and using routes that the conventional forces had assessed as too difficult for regular use.

 The routes were too difficult for regular use. They were not too difficult for the people using them. The SAS assessment team recommended a deployment of 40 men. The command structure asked whether the team was certain about the number. The team confirmed the number and noted that adding more men would not change the result and might make it worse.

 The 40 SAS soldiers who deployed to the border arrived without the equipment profile that any of the five Allied armies had used. They did not bring the vehicle fleet, the communications infrastructure, the surveillance systems, or the logistical support chain that conventional border security operations of this scale typically require.

 They brought personal kit, weapons, sufficient rations for extended operations without resupply, and a set of maps of the terrain that differed from the standard military maps in a specific way. They showed roots that the standard maps classified as impassible. The routes were impassible for a unit moving at conventional military speed with conventional military loads.

 They were not impassible for small groups moving slowly and quietly with minimal weight. The SAS had obtained the alternative mapping from a combination of local sources and aerial photography interpreted by men who had operated in similar terrain before. They spent the first three days of the deployment walking the roots before they established any positions.

 And they walked them without weapons in their hands because walking with a weapon in hand is a different movement pattern from walking without one. And they wanted to understand how the routes felt to the people who used them. The positioning decision that followed the terrain walk was not discussed with the Allied command structure before implementation.

 This was not an oversight. The SAS commander on the ground had reviewed the three years of operational records from the five armies and had identified what he considered the central operational problem. That every positioning decision made by the Allied forces had been made from a map and had been communicated through a command chain which meant that by the time a position was occupied the crossing network had had time to observe the occupation and adapt to it.

 The SAS positioned at night in terrain sectors that the Allied Assessment had categorized as secondary priority without transmitting the specific grid references up the command chain before moving. The Allied command structure was informed of the positioning after it had been established. There was a formal complaint from one of the national contingent commanders which was noted and not acted upon by the joint headquarters.

 The SAS positions were not adjusted. The terrain walk had taken 14 hours. It covered the border sector that the intelligence picture had identified as the primary movement corridor and covered it in a way that produced a ground level understanding of the corridor’s actual structure, its specific features, its vulnerabilities, its relationship to the terrain on either side of it that the intelligence picture had not contained.

 The positioning decision that followed was the product of that additional understanding, not of the intelligence picture alone, but of what 14 hours on the ground had added to it, and in several specific respects subtracted from it. The first week produced no visible result at the level of the overall border.

 The infiltration numbers monitored by the Allied intelligence apparatus did not change significantly in the first 7 days. the Allied command structure, which had not been enthusiastic about the SAS deployment, not because they doubted the regiment’s capability, but because 40 men was a difficult number to explain in a political environment where 3 years of inadequate progress had created significant pressure for demonstrable action, began asking questions through the liaison channel that the SAS commander answered briefly and without

elaboration. He reported that the deployment was progressing as expected. He did not specify what he had expected or why. The command structure received this with the discomfort of organizations that prefer legible progress metrics and the discomfort was noted in the liaison records. It did not change the operational approach.

 The first week’s invisibility at the border level was not a failure of the SAS deployment. It was a necessary feature of an approach that was working on the border problem from below the surface rather than at the surface. The four allied nations monitoring the border during the SAS’s first week saw the same border they had been monitoring for three years, the same crossing patterns, the same movement signatures, the same operational picture.

 The picture would not change until the structure below it changed. The structure was changing. The first week was the week the structure changed. The change would not be visible at the surface for several days more, and it would not be fully measurable until after the deployment ended. The invisible first week was the deployment’s most important week.

 What happened in the second week was not a dramatic single event. It was a series of small events distributed across the 43 kilometers of the border that taken together produced a result the Allied monitoring systems registered as significant only when they were compiled at the end of the week. The infiltration network that had been using the border for 3 years had encountered for the first time in that period a consistent presence in the specific terrain sectors it considered safe.

 not at the obvious crossing points. Those had been covered and had adapted to being covered, which was why they were not the crossing points being used. In the terrain classified as impassible, where no allied unit had established sustained observation, the SAS had been present quietly for 8 days. The network had not known they were there.

 It had used routes it had used without incident for 3 years. The incidents that began to accumulate in the second week were small, individual, and spread across the terrain in a way that made the network’s adaptation strategy, identifying the covered points and routting around them, inapplicable because there were no uncovered points left to route to.

 By the end of the second week, the border sector reporting and assessment of sealed came from the Allied Intelligence Monitoring Network rather than from the SAS. The SAS filed a brief operational summary noting that the tasking had been completed. The summary was three paragraphs long. The Allied command structure, which had three years of operational records from five armies to compare it against, produced a response that was considerably longer.

 The response asked a series of questions that the SAS summary had not addressed. what force protection measures had been in place, what the resupply chain had looked like, what the communication protocols with local authorities had been, and what the SAS intended to do about the terrain sectors where no direct observation had been established.

The SAS response to these questions explained that the force protection had been the positioning itself, that resupply had been managed internally within the 40man group, that communication with local authorities had been conducted through means the summary did not detail, and that the terrain sectors with no direct observation had not required direct observation because denying the accessible sectors was sufficient.

 The Allied command structure received this response and produced another round of questions. The SAS did not answer the second round. The five National Army commanders who had spent three years on the border were briefed on the SAS outcome through the joint headquarters. The briefings were factual and uncontentious. No one in the briefing suggested that the five armies had been incompetent or that their failure to seal the border had been a consequence of inadequate capability.

The explanation offered was that the mission as originally configured providing a visible conventional military presence along a border corridor as a deterrent and interdiction measure was not matched to the problem which was an infiltration network using routes specifically selected for their unsuitability for conventional military presence.

 The SAS deployment had reconfigured the mission at the ground level in a way that matched the operational approach to the actual problem rather than to the institutional expectation of what border security should look like. This explanation was received without visible controversy in the briefings. It was somewhat more controversial in the subsequent discussions that did not appear in formal records.

 The terrain where the SAS had positioned during the two weeks of the deployment had a specific characteristic that none of the five Allied armies had exploited and the characteristic was not secret or obscure. It appeared in the terrain assessments that all five armies had produced during their operational periods.

 The characteristic was that the terrain classified as impassible was for a person who knew it and was willing to move through it slowly and without the equipment that made it impassible actually advantageous. It provided concealment that the more accessible terrain did not provide. It limited the directions from which an observer position could be approached.

 It produced a sound environment that made the approach of larger groups detectable at distances that did not apply in more open terrain. Each of the five armies had noted these characteristics in their terrain assessments and had then proceeded to set up positions in the accessible terrain because the accessible terrain was where the equipment could go and where the supply chain could reach and where the command structures communication protocols could be maintained.

 The SAS did not have that equipment and did not use those supply and communication protocols. This made the inaccessible terrain available to them in a way it had not been available to anyone else. The 40man group operated without rotations during the two-week period. This was a planning decision made before deployment and it shaped everything that followed.

 Conventional border security operations rotate units through positions to maintain readiness and to manage the sustainment requirements that conventional equipment profiles create. The rotation cycle is also the most significant source of vulnerability in a position-based interdiction operation because the period immediately before and after a rotation is when the position’s characteristics, its exact location, its sight lines, its activity patterns are most observable to anyone watching from outside. A network that has had three

years to observe rotation patterns has had three years to identify the rotation windows and to schedule its movements accordingly. The SAS group did not rotate. The same 40 men were in the same positions throughout the two weeks. Their activity patterns were not those of a unit cycling through predetermined readiness states.

 They were the patterns of men who had settled into terrain for an extended period, which are quieter, more irregular, and harder to distinguish from the absence of any presence at all. The Allied Command Structures formal assessment of the SAS deployment produced 6 weeks after the border was declared sealed, was a careful document.

 It acknowledged the outcome, noted the unconventional approach, and made a series of recommendations about how elements of the SEAS methodology might be incorporated into standard allied border security protocols for future operations. The recommendations were reasonable and the document that contained them was well constructed.

 The SAS liaison officer who reviewed the document before it was finalized noted in a marginal comment that was retained in the working copy that most of the recommendations were about adding SAS style approaches to existing conventional frameworks rather than replacing those frameworks and that this was likely to produce limited results because the SAS approach worked in part because it was not accompanied by a conventional framework.

 The comment was discussed, noted as a valid observation, and did not change the recommendations. The document was finalized with the recommendations intact. The assessment was produced on schedule and distributed through the standard channels. It contained findings about the two-week deployment’s effectiveness relative to the three-year operation that were accurate and by any professional standard significant.

 The level at which the assessment was circulated indicated that receiving commands were expected to treat the findings as operationally relevant rather than as historical record. They were treated accordingly. The changes in subsequent planning that the findings produced were the evidence of that treatment and they were visible within the planning cycle that followed the assessment’s distribution.

 The border held for 14 months after the SAS deployment ended. It then began to show infiltration again at a lower rate than the pre-eployment period, but at a rate that indicated the network had resumed and had adapted to the closure. The Allied command structure requested another SAS deployment.

 The SAS returned with a different group of 40 men, used a different set of positions derived from a new terrain assessment, and sealed the border again in 11 days. The Allied command structure asked whether the positions from the previous deployment should be reused since they had been effective. The SAS team said no. The previous positions had been effective because the network had not known they were there.

 The network now knew they had been there. Going back to the same positions was in the SAS assessment precisely what the network would expect a second deployment to do. A second visit to the same positions would find the same terrain occupied by a different kind of attention. The SAS used positions the network had no way of having predicted.

 The border closed again in 11 days. What the five national armies had spent 3 years attempting to demonstrate that a border 43 km long in difficult terrain could be controlled by a sufficient number of professionally capable soldiers was not wrong as a proposition. It was wrong as an operational approach to this specific border at this specific time.

 The sufficient number of soldiers was not 5,000 or 20,000. The sufficient number was 40. And the quality of the difference between 40 that worked and 20,000 that did not was not a quality of individual soldier capability. It was a quality of approach. The willingness to go where the equipment could not go, to stay without rotating, to position in the terrain the problem was actually using rather than in the terrain a conventional military presence could comfortably occupy.

 The five armies had brought what they had. The SAS had brought what the problem required. The outcome followed from that difference with the inevitability of water finding a gradient. The demonstration had not been achieved. Not because the five armies had failed to achieve what they were attempting, but because what they were attempting was an inherently incomplete response to the problem as the SAS assessment had defined it.

 You cannot demonstrate that a border is closed by clearing it repeatedly. You can only demonstrate that it is closed by removing the infrastructure that allows it to be reopened after each clearing. The three years of clearing had been producing the first kind of demonstration. The two weeks had been producing the second.

 The two kinds of demonstration were different in kind, and the three years had not moved the problem toward the second kind. The operational record for the two-week deployment runs to 11 pages in the SAS regimenal archive. The allied command structures records of the same period run to several thousand pages. The difference is not a reflection of the completeness of the SAS records.

 It is a reflection of the fact that the SAS record describes what happened. And the Allied records describe, among other things, the process of trying to understand what happened and why, which requires significantly more pages. The understanding the Allied command structure eventually reached that the SAS had solved the problem by removing from their operational approach everything that had caused the problem to be unsolvable for the five armies was accurate.

 It was also as the people who reached it recognized a description of something that is easy to understand and difficult to replicate because the things removed were not incidental features of the Allied army’s approach. They were the approach. The 11 pages were dense in the way that SAS operational summaries are dense, specific, chronological, and without the narrative context that would have made them legible to a reader unfamiliar with the operation.

 Each of the 11 pages assumed the reader had access to the planning documents, the intelligence picture, and the terrain assessments that had preceded it. Without that context, the summary was a sequence of actions with times, locations, and outcomes. With that context, it was the most complete available account of how two weeks and 37 men had resolved a problem that three years and significantly larger resources had not.

The Joint Command’s own operational record, which ran to considerably more than 11 pages, was the context. The two documents together were the account. Neither was fully intelligible without the other. The afteraction review produced by the joint headquarters following the second deployment contained a recommendation that was not present in the review following the first that the allied nations should consider establishing a small permanent liaison capacity with the SAS regimenal structure to allow for faster deployment

of similar capabilities in future border security contexts. The recommendation was noted by four of the five national military staffs and tabled for further consideration. One national staff rejected the recommendation on the grounds that creating a dependency on external specialist capability was an inappropriate response to a border security challenge that should be manageable by organic national forces.

This position was formally recorded. In subsequent years, that nation’s border forces requested SAS technical advice on three separate occasions. The requests were handled through bilateral channels rather than through the joint headquarters, and the connection to the formal rejection of the liaison recommendation was not noted in any official document.

 The individual soldiers who had formed the 40-man group were drawn from different squadrons within the regiment, selected on the basis of specific terrain experience and on the basis of availability within the deployment cycle that the regiment was managing at the time. They were not the most experienced men in the regiment at border operations in abstract.

 They were the most appropriate combination of men for this specific border at this specific time, which is not the same thing. The distinction was important to the regimental planning staff who assembled the group and it was one of the things that the Allied command structure found difficult to translate into its own force generation models.

The Allied models worked from general capability categories, mountain infantry, border control specialist, surveillance operator and assigned personnel to tasks on the basis of those categories. The SAS group was assembled around a different question. who specifically has operated in terrain like this and can operate in it with this specific set of constraints for 2 weeks without resupply.

 The local population in the border area had been aware of the infiltration network for the 3 years of allied operations. This was known to the Allied intelligence structures and was documented in their assessments. The local population’s awareness of the network was not in the Allied assessment, something that could be addressed directly during the operational period, and the Five Army’s engagement with local communities had been limited to what the force protection requirements of a conventional military presence permitted. The SAS approach to the local

population was different and was not documented in the operational record in any form that could be reconstructed after the fact. What is documented is that the terrain information the SAS used, the route mapping that showed paths the standard military maps classified as impassible, was more detailed and more current than any aerial photography alone could have produced.

 The source of the additional detail is not specified in any available record. The awareness of the local population was not academic. They had been living with the infiltration network’s presence for the duration of the three-year operational period, and their knowledge of it was the product of daily proximity to its operations. The movement patterns, the economic relationships, the social structures the network had embedded in the local geography over a period long enough that it had become for many residents an unremarkable feature of the local

environment rather than an intrusion into it. The SAS presence during the two weeks added a disruption to that environment conducted at a pace and through methods the local population’s own information network could not track that the network itself could not reliably communicate about. The infiltration network that the SAS had sealed operated with a logic that the Allied armies had understood in general terms for most of the three years they had been on the border.

 The network was moving people and material across a 43 km line using a selection of routes that had been specifically chosen over time for their unsuitability for conventional interdiction. The routes had not been chosen randomly. They had been chosen through a process of observation. The network had watched the Allied forces for long enough to understand exactly which terrain the Allied forces could not effectively cover and had consolidated its activity into that terrain.

 This was not sophisticated planning in the sense of requiring unusual intelligence or unusual resources. It was patient observation followed by straightforward adaptation to what the observation revealed. The Allied forces had been observable for 3 years. The network had had three years to observe them. The SAS arrived and spent 3 days observing before they established any position, which was 3 days more than any Allied unit had spent on that activity before beginning to operate.

 The classification level of the SAS operational record for the two deployments means that specific tactical details of how the 40man group covered the border, where each element positioned, what communication protocols they used, how they managed movement between positions without being observed are not available in open sources and will not be available for a significant period.

 What is available is the outcome, the timeline and the structure of the allied response. And from these three elements, the operational approach can be inferred with reasonable accuracy by anyone who understands how the SAS typically operates in terrain interdiction tasks. The inference produces a picture of a group that disappeared into difficult terrain and became invisible to the network it was interdicting using the terrain’s own characteristics as its primary capability multiplier.

 This is not a novel concept. It is the operating principle that has characterized the regiment since the second world war. And it has characterized it because the environments where the principle works are the environments where conventional forces cannot follow. The political environment surrounding the border operation during the three years before the SAS deployment had been a significant factor in shaping how the five armies had operated.

 Each of the five nations had domestic audiences receiving information about the operation and each national contingent commander had been operating with an awareness of the domestic reporting. Operations that produced visible activity, patrols, checkpoints, equipment deployments were easier to report on in ways that satisfied domestic audiences than operations that produced no visible activity because the activity was concealed.

 The pressure toward visible operational activity was not unique to any one nation. It was a structural feature of politically observed military operations. The SAS deployment was not politically observed in the same way. The regiment’s operations are not reported on in a form that creates domestic audience expectations about what visible activity should look like.

 This allowed the 40 men to operate invisibly without generating political pressure to become visible, which was an operational advantage that could not be planned for, but that turned out to be decisive. The comparative cost of the twoe SAS deployment against the 3 years of Allied operations was not a calculation that appeared in any official document.

 It was discussed informally in the margins of the allied command structures review process and the figure that emerged from those discussions which was approximate and contested in its components was significant enough that several of the national contingent commanders asked that the informal discussion not be formalized in any document that would be subject to national parliamentary scrutiny.

 The cost comparison was in the terms that made it most uncomfortable straightforward. The SAS deployment had produced the outcome that three years of significantly more expensive operations had not produced in two weeks with 40 men using less equipment than a single allied company had brought to the border.

 The combination of those numbers did not survive well in formal settings. The comparative cost analysis that was not calculated formally was calculated informally by everyone in the joint command structure who was aware of both the three years and the two weeks. The informal calculation produced different numbers depending on the variables included.

 Whether you counted only the allied forces directly committed to the border operations or the broader infrastructure costs. Whether you counted the opposition’s operational capability as a cost to the allied side during the three years, whether you counted the opportunity costs of assets tied to the border problem rather than available for other operational requirements.

 The variables produced different numbers. The conclusion was the same regardless of which variables were included. The two weeks had done what the three years had not done. The margin, however, the calculation was performed was not small. The SAS commander who led the deployment was asked in a post-operational interview conducted by the Allied Joint Headquarters whether he had been confident the approach would work within 2 weeks.

 He said he had been confident that if the assessment was correct, the approach would work and that the twoe timeline had been set as a planning assumption rather than a guarantee. He said that if the assessment had been wrong about the route selection patterns, the two-week deployment would likely not have been sufficient and a different approach would have been required.

 He was asked whether he would have wanted more men if the assessment had been wrong. He said the number of men was not the variable he would have changed. He would have changed the positioning rather than the strength. The interviewer asked a follow-up question about what would have happened if additional national contingent forces had been available to support the SAS elements.

 The SAS commander said the additional forces would have been a problem to manage rather than an asset. He said this without apparent awareness that the response might be received as diplomatically suboptimal. The interviewer noted the response and moved to the next question. The national army that had produced the longest operational record during the three years of Allied border operations, the most sustained presence, the most detailed documentation, the highest number of personnel cycled through the theater, produced a formal analysis of

the SAS outcome that ran to 40 pages. The analysis was professionally competent and its factual content was accurate. Its conclusion, which came after 38 pages of careful examination of the operational data, was that the SAS had been more effective on the border than the Allied forces because the SAS had used a methodology that was not replicable at scale within conventional force structures.

 This conclusion was accurate. It was also, as the analysis inadvertently demonstrated by taking 40 pages to reach it, a conclusion that was easier to state than to derive from first principles, and the difficulty of deriving it from first principles was itself a finding about the Allied force structures that the analysis did not explicitly name.

 The longer record was a product of the nation’s geographical position relative to the border sector, which had made its forces the natural lead for the joint command’s clearing operations in the area and of its willingness to maintain that commitment over the full three-year period when other nations had rotated their contributions in and out.

 The commitment was real and professionally sustained. The outcome was the same as for the nations whose commitment had been shorter. The border remained open. The movement corridor continued to function and the operational picture three years after the joint command’s formation was not materially different from the picture at its formation.

 The regiment’s internal file on the operation is not available for examination. What is available from the joint command records that are no longer classified is the timeline. Assessment team deployed four weeks of preparation. 40 men on the ground. Border sealed in 14 days. Border declared sealed by the Allied Monitoring Network. SAS group extracted on day 15.

The timeline is clean. The Allied Command Structures records of the same period are not clean. They contain the assessments, the questions, the follow-up questions, the liaison communications, the political reporting, the force rotation schedules, and the supply chain records for the five army’s presence on the border across three years.

 The two records placed side by side are in some ways the most direct comparison available of how two different approaches to a military problem look when they are over. One of them is a timeline. The other is a history. The difference between a timeline and a history is a measure of how complicated the problem needed to be.

 The file is believed to exist in the form standard for SAS operational records. a brief summary document, an operational log with its entries, and the supporting intelligence and assessment materials that had informed the operations planning. The external access restriction is standard for operations of this classification and is not specific to this operation.

 What can be established from the external record, the joint commands documentation, the allied national records, the intelligence picture available in the relevant period is sufficient to reconstruct the operation structure and outcome, if not all of its internal planning detail. The question of whether the SAS approach to the border could have been applied three years earlier.

Whether the same 40 men or 40 different men using the same methodology could have sealed the border in two weeks in the first year of the Allied operation was raised informally in several of the post-operational discussions. No one who raised it produced a confident answer. The terrain had been the same three years earlier.

 The infiltration network had been operating for at least as long. The SAS methodology was not a new development. The obstacle, as best as anyone could reconstruct it, was not a shortage of the capability. It was a shortage of the conditions under which the capability could be deployed. The political tolerance for a 40-man operation when political pressure was demanding visible large-scale response.

The command structures willingness to allow a small group to operate without the oversight protocols that conventional force employment requires. and the institutional confidence to tell the domestic audiences that the appropriate response to a three-year border problem was 40 men and 2 weeks. Those conditions had taken 3 years to develop.

 They had developed because everything else had been tried first. The SAS group returned from the deployment and filed its operational summary. The men rotated back into the squadron duties from which they had been drawn. There was no ceremony and no formal recognition. The regiment does not hold ceremonies for operations of this scale.

 What occurred instead was the kind of informal acknowledgement that passes between soldiers who know what an operation involved and what it required. Brief and professional and not repeated because in the culture of the regiment the work is the acknowledgement and anything additional is considered unnecessary. The border had been sealed.

That was what they had gone to do. They had done it in two weeks. The analysis of what that meant and the hundreds of pages that would be written about it in the years that followed by people who had not been there was happening in buildings that were not their buildings and they were not present for it.

 The operational summary was filed with a directness characteristic of SAS documentation. What had been done? What had been achieved? What had been used? The summary did not include a comparison with the three years of allied operations that had preceded it because comparisons of that kind were not what operational summaries were for.

Operational summaries describe the operation. The comparison was implicit in the operation’s outcome and in the three years that had preceded it and in the simple fact that both were in the record. Anyone who read both parts of the record could produce the comparison. The SAS did not write summaries that produced comparisons for the reader.

They wrote summaries that stated outcomes. The outcome of the two weeks was in the summary. The three years were in the Allied record. The comparison lived in the gap between the two documents and the gap was 33 months. The 43 km border remained a monitored line in subsequent years with periodic SAS involvement when the monitoring indicated resumed infiltration at levels that exceeded what Allied forces could manage.

 Each involvement was shorter than the one before it. Not because the regiment was getting faster, but because the network was adapting more cautiously after each closure, which meant that each closure, once achieved, lasted longer. The infiltration that followed each SAS deployment was at a lower rate from a reduced network using routes that were less reliable than the routes that had been available 3 years before.

 The regiment had not solved the border problem permanently. No 40-man group operating for 2 weeks can solve a political and geographic problem permanently. What they had done was changed the terms on which the problem operated. The network that had spent three years adapting to five armies had now spent the same three years being periodically reminded that some of the terrain it considered safe was not safe.

And that reminder had changed how the network planned, which had changed what the problem was.