What lessons should militaries learn from Afghanistan?

Afghanistan is a reminder that strategy is older than our kit. Two enduring lessons:

1) Political end state drives military feasibility. Clausewitz isn’t optional. If the desired end state is a centralized, corruption-resistant state with monopoly of force, but the society and incentives don’t support it, the army ends up doing nation-building by rifle—historically a poor trade. The British learned versions of this in the 19th century; the Soviets relearned it in 1979–89; we had our own iteration.

2) “Tactical success” can be strategically irrelevant. You can win engagements all day and still lose the war if the opponent’s center of gravity is time, legitimacy, and local power networks. Galula/Nagl COIN frameworks highlight legitimacy, but Afghanistan shows legitimacy can’t be imported or sustained by foreign presence indefinitely.

And a doctrinal point: militaries should retain the ability to do COIN-like operations without allowing them to hollow out conventional deterrence. The Cold War-era balance between irregular warfare capability and high-intensity readiness is worth revisiting—especially given peer threats.

From the equipment side, Afghanistan screams “mission drives loadout, not the other way around.” A few lessons:

- Weight discipline matters more than people admit. Mountains + heat + long dismounted patrols punished over-armor and overloaded packs. If your doctrine assumes constant vehicle support, Afghanistan shows what happens when terrain and IED threats break that assumption.

- Optics and comms were force multipliers. Good glass (LPVOs, ACOGs, rangefinders), reliable encrypted radios, and power management often mattered more than the newest caliber debate.

- Boots, hydration, and sustainment gear are not glamorous but decide performance. Blisters and heat injuries wreck units faster than forum arguments.

- Standardization + modularity: units kept “solving problems locally” with pouches, plate carriers, and aftermarket fixes. That’s a signal procurement should build adaptable systems, not lock everyone into a single rigid configuration.

Also, maintainers deserve more credit—dust and fine grit chew up everything. Reliability in harsh environments should be a top procurement KPI.

Big lesson for me: you can train a force to shoot and move faster than you can train a force to want the same outcome.

In advisor roles you’d see dudes who were brave and competent at the small-unit level, and then you’d hit the wall of pay issues, absenteeism, local politics, and “who do I answer to?” If leadership and accountability aren’t real, tactics don’t stick.

Another lesson is basics win: maintenance, logistics, and admin are combat power. You can hand over good equipment, but if the system can’t fuel it, fix it, and track it, it becomes dead weight.

And culturally: building trust takes years, but it can break fast. Constant rotation cycles, changing priorities, and shifting ROE made it hard to build continuity. If militaries do advisory missions again, they need longer-term teams, better language/cultural capability, and a realistic definition of “success.”

The lesson is brutally simple: stop pretending you can “engineer” legitimacy with PowerPoints and contracts.

Everyone loves blaming one tactical thing—equipment, a single leader, a single decision. No. If your partner government is viewed as predatory, the insurgency has an endless recruiting pipeline. You can’t out-train that.

And here’s the part people dodge: we optimized to reduce our casualties and maximize raid/strike efficiency, then acted surprised when the partner force didn’t develop independent initiative. If your model is “we will always have ISR, air, medevac, and contractors,” you’re not building an army—you’re building a dependent.

So the real takeaway is honesty: define what you will do, what you won’t do, and what you’re willing to pay in time and blood. Anything else is theater.

Afghanistan showed both the power and the limits of ISR dominance.

- Persistent drones changed tactical tempo: pattern-of-life, route surveillance, overwatch, quick target development. That saved lives.

- But ISR doesn’t equal understanding. You can watch a village for weeks and still miss the human networks, coercion, and informal governance that drive behavior.

- The other lesson is democratization: commercial drones and cheap sensors are now everywhere. Future “Afghanistans” won’t be one side with total aerial awareness and the other side blind. Expect counter-UAS to be as standard as counter-IED was.

Looking forward: fuse UAV feeds with better human intelligence validation and tighter decision loops, but don’t build strategy around the assumption that tech will reveal intent. It rarely does.

Afghanistan is a case study in how armored doctrine gets bent by terrain, politics, and IEDs.

- Heavy armor isn’t useless, but it’s not a magic key in mountains and narrow roads. Mobility corridors get predictable, and predictable routes get mined.

- MRAPs were a lifesaver against blasts, but they also changed how units moved—bigger vehicles, different driving skill sets, and logistical burden. They’re not “one size fits all.”

- Partner forces often got platforms without the maintenance ecosystem. An IFV or even a simple armored truck is only as good as spare parts, trained mechanics, recovery capability, and fuel. That system is harder than buying vehicles.

Lesson for future procurement: balance protected mobility with maintainability and local supportability. If it can’t be sustained without contractors, plan accordingly or don’t field it.

Even though Afghanistan is landlocked, a naval lens highlights a few strategic mechanics:

- Access and basing are everything. When overflight rights and regional politics shift, your sustainment and strike options change overnight. Militaries should plan for “contested access” even in counterinsurgency-style missions.

- Sea-based logistics didn’t solve the “last thousand miles.” You can deliver mountains of materiel to ports, but the real friction is inland distribution, security, and corruption leakage.

- Power projection isn’t just carriers and missiles; it’s diplomatic relationships that enable them. Afghanistan underlines how military plans fail if the regional alignment isn’t stable.

Takeaway: integrate logistics and diplomacy into operational design from day one, not as a support annex.

Aviation lessons are huge:

- Airpower is unmatched for mobility, CAS, MEDEVAC, and ISR, but it creates dependency if the partner force can’t generate sorties, maintain aircraft, or control airspace.

- Rotary-wing sustainment is hard. Dust, altitude, and supply chains punish helicopters. If your entire tactical concept relies on helos, you need robust maintenance, pilot pipelines, and spares—forever.

- Close air support worked tactically, but strategic effects were mixed. If the ground force’s confidence depends on guaranteed air, morale and initiative can collapse when that safety net goes away.

Lesson: design partner air forces around what they can actually maintain and staff (often simpler aircraft, fewer types), and make sure ground doctrine works without constant air intervention.

For people thinking “what should militaries learn,” I’d add a personnel angle:

- Advisor missions require a different talent mix than conventional units: patience, communication, cultural awareness, and integrity under ambiguity. Selecting and training for that should be deliberate, not an afterthought.

- Rotation and burnout are real. Long deployments and repeated cycles erode readiness and retention. If a force structure assumes endless deployments, it’s not sustainable.

- Skills that paid off: language ability, negotiation/de-escalation, basic trauma care, and disciplined reporting.

For newcomers considering joining: talk to a recruiter and also to people who actually did advisory or security cooperation work. It’s not just “combat”; it’s a lot of relationship-building and problem-solving.

SOF lessons from Afghanistan are often misunderstood because people see the highlight reel.

- Special operations are great at raiding, partnering, and enabling—but they can’t substitute for a coherent political strategy or for conventional force capacity.

- Overreliance becomes a trap: when SOF are the default tool for everything, you burn them out and distort the force. Direct action keeps expanding because it’s measurable, while harder governance/security work lags.

- Partnering works best when it’s long-term with consistent teams. Constant turnover resets trust.

SOF should be used with clear objectives and limits. Otherwise you get endless tactical success without durable security.

Afghanistan reinforced that environment is an enemy you can’t negotiate with.

- Heat, altitude, and dehydration quietly degrade decision-making. Units that baked acclimatization, water discipline, and foot care into routine performed better.

- Navigation and basic fieldcraft matter even with GPS. Batteries die, signals get jammed, and terrain can force detours. Being able to move confidently without tech is still a skill.

- Security in austere outposts: sanitation, waste management, and food/water storage aren’t glamorous, but they prevent mission-killing sickness and preserve readiness. (If anyone’s planning outdoor training, use qualified instructors and follow safety standards.)

Lesson: invest in “boring” fundamentals. They’re what keep you functional on day 60, not day 2.

The core lesson is alignment of incentives—inside the partner state and among external stakeholders.

- Corruption wasn’t just “bad behavior”; it was an alternative governance system with beneficiaries. If your aid and contracting flows reinforce that, you’re funding instability.

- Intelligence collection was often strong at the tactical level, but strategic forecasting struggled with political will, elite bargaining, and regional sanctuaries. Those are not purely military problems.

- Timelines broadcast intent. When withdrawal timelines become credible, actors hedge. That’s rational behavior, not betrayal.

Militaries should push policymakers for clearer assumptions: what does success look like without indefinite funding, contractors, and external air support? If it can’t stand on its own, call it what it is—a dependent security arrangement.

Afghanistan is a logistics and sustainment masterclass.

- The “last mile” was often the whole problem: rough roads, security threats, seasonal weather, and administrative leakage. Planning factors must reflect reality, not spreadsheet optimism.

- Infrastructure projects need lifecycle thinking. Building a facility is easy; staffing, powering, maintaining, and securing it for years is hard. Too many builds became liabilities.

- Contractor dependence is a strategic vulnerability. If key systems require contractors for maintenance, you need a plan for what happens when contractors can’t or won’t operate.

Lesson: treat sustainment as a decisive operation. If you can’t sustain it, you don’t truly own it.

I’m still learning, so sorry if this is basic, but is the main lesson that “COIN doesn’t work,” or is it more like “COIN only works under specific political conditions”?

Also, when people say the local forces were “trained wrong,” what does that mean in concrete terms? Like, were they trained to rely on air support too much, or was it leadership/corruption more than training?

Trying to understand what parts are military lessons vs. political ones.

If you model Afghanistan as a system, the failure mode looks like mismatched assumptions in the “support graph.”

The partner force had nodes that were externally provided: airlift/ISR, contractor maintenance, pay systems, certain intelligence capabilities, and sometimes even planning functions. Remove a few key nodes and the graph collapses quickly.

So the lesson for future force design is: build resilience and redundancy into partner capacity. Can they communicate if one network fails? Can they pay troops reliably? Can they sustain ops without foreign enablers for 30–60 days?

War games should stress-test advisory missions with shocks: loss of air support, supply interdiction, leadership turnover, and regional political shifts. If the plan fails those tests, it’s not a plan.

Afghanistan hints at where robotics could have helped—but also where it wouldn’t.

- Ground robots for route clearance, IED investigation, and perimeter reconnaissance reduce risk. That’s a practical, near-term win.

- Logistics robotics (small UGV mules, autonomous resupply) could cut down on convoy exposure and soldier load in rough terrain—if they’re rugged and maintainable.

- But robotics can’t fix legitimacy, leadership, or cohesion. It can only improve how you execute a strategy.

Lesson: invest in unmanned systems that reduce predictable risks (convoys, base security, ISR gaps), while staying realistic about sustainment and training. If a robot needs a contractor and perfect supply chains, it won’t survive real deployment conditions.