Military history
Predicting the Next War: The 2006 Argument and How It Aged
In 2006 the American officer corps was in the middle of an argument about what the next war would look like, and the argument was unusually public. Journal articles, conference papers and the emerging military blogs carried a debate that mattered because it was about where to put money and training time for the following two decades. The record of that argument is now old enough to be graded, which makes it more instructive than most predictions.
The two positions
The irregular camp argued that state-on-state conventional war was becoming rare and that the future belonged to insurgency, terrorism, militias and conflicts fought among populations instead of across front lines. On that reading the Army needed more infantry, more language and cultural capability, more advisers, more military police and civil affairs, and less heavy armour. Iraq and Afghanistan were the model, not the exception.
The conventional camp argued that this generalised from a short and unrepresentative period. States with real militaries still existed, were modernising, and would be the actual threat. Heavy formations, air superiority, integrated air defence and long-range fires were expensive to build and slow to rebuild, and an army that let those skills lapse would find them very hard to recover. Preparing only for the war you are in is how you lose the next one.
A third strand, less tidy, argued that the distinction was false: that future opponents would combine regular formations, irregular forces, long-range fires and information operations simultaneously. That position was labelled hybrid warfare and drew heavily on the Lebanon fighting of that period, in which a non-state force used prepared positions, guided anti-tank weapons and mass rocket fire in combination.
How it aged
The honest scoring is that each camp was right about something and wrong about the timing.
The irregular camp was right for about a decade. The conflicts the American military actually fought between 2006 and the mid-2010s were overwhelmingly the ones they described, and the counter-insurgency capability built in that period was needed and used.
The conventional camp was right about the return. From the mid-2010s the emphasis shifted back to peer competition, and the skills the irregular camp had been willing to trade away: combined-arms manoeuvre at scale, air defence, counter-battery fire, electronic warfare, large-formation logistics, turned out to be exactly the skills that had decayed and were slow to rebuild. That is the argument’s strongest single vindication.
The hybrid strand was right about the shape. Recent fighting has consistently combined the supposedly separate categories: regular formations and irregular units in the same battle space, cheap precision weapons alongside massed artillery, and information operations running throughout. The neat division the other two camps were arguing over has not held up.
What nobody in 2006 weighted correctly
Two developments changed the picture more than either side’s prediction.
Cheap unmanned aircraft. In 2006 unmanned systems meant a small number of expensive theatre-level airframes providing persistent surveillance over uncontested airspace. Nobody in the mainstream of the debate anticipated very large numbers of very cheap aircraft flown by the squads that need them, plus expendable loitering munitions in the same size class, and the consequence that concealment would become both much harder and much more important.
The reach of commercial technology. Satellite imagery, encrypted messaging, commercial drones, open-source geolocation and social media have all become inputs to targeting and to intelligence, in ways that a 2006 argument about force structure did not consider because most of it did not exist at scale.
The transferable lesson
The argument’s value now is not in who won it. It is in the failure mode both camps shared: each extrapolated from the most recent significant conflict, and each was confident about a timeline.
The pattern is old and it repeats. Doctrine is written by the officers who fought the last war and it encodes that war’s assumptions invisibly. Equipment programmes and training pipelines take longer to turn than strategic circumstances do, so a force is generally designed for the previous conflict and employed in the next. Recognising the mismatch takes years; correcting it takes more.
The practical response is not better forecasting, which has a poor record. It is retaining the capability to relearn — keeping doctrine, schools and formations flexible enough that a shift in the character of conflict is expensive rather than catastrophic. That conclusion was available in 2006 and was the least popular position in the debate, because it commits to no particular future and therefore justifies no particular budget.
Questions and answers
What was the substance of the disagreement?
Whether the armed forces should be built for large-scale conventional war against a peer, or for prolonged operations among populations. Both were serious positions with evidence behind them, and both had institutional interests attached: the platforms and formations each implied were different, and so was who got the budget.
Why can force structure not simply cover both?
Because the same money buys different things and the same soldiers are trained for different tasks. A force optimised for rapid conventional operations is not the force that holds terrain among a population for years, and building both means being worse at each. Every force structure is a bet, and the cost of being wrong is paid years after the decision.
How did the argument resolve?
It did not, and it has since reversed direction at least once: toward counterinsurgency after 2006, then back toward large-scale conventional competition. That oscillation is the most instructive part: the lesson usually drawn from each turn is to build for the war currently being fought, which is the same error in the opposite direction.