From March 23 to May 12, 2008, US Army and Iraqi security forces engaged in an intense urban battle in and around Baghdad’s Shiite neighborhood of Sadr City. The battle forced units that had been mainly focused on counterinsurgency operations to convert overnight into combined-arms teams of armor and mechanized infantry and engage a large enemy element in a high-intensity fight in a city containing around two million noncombatants. The adaptions these units made and the results of this short but pivotal battle provide important lessons for future combat in dense urban environments. ...
Under what was called the Baghdad Security Plan, one of the surge brigades was assigned to Thawra District, although no US forces would be based in Sadr City itself. The unit implemented an aggressive plan to reduce overall levels of violence that consisted of moving into small outposts within the neighborhoods; using concrete barriers to construct checkpoints, safe neighborhoods, and safe markets that reduced the enemy’s ability to transport resources and to conduct mass-casualty attacks with VBIEDs and other weapons; targeting key insurgent leaders and groups; and improving ISF capabilities.
The role and effectiveness of concrete in reducing violence across Baghdad cannot be understated.
My memory is that war opponents hated those walls. But the walls worked. It was like medieval siege lines built by forces laying siege while under fire from the besieged, to squeeze and isolate an enemy fortress. We should keep this wall-building effort in mind when planning to fight in mega-cities.
I've said we shouldn't think of taking an entire city as the default mission. We should determine what parts of the city we need to take to further achieving the objective of the campaign:
[For] a conventional war where the campaign continues against the enemy army in the field, "controlling" a captured enemy city in the short run only requires controlling the militarily significant features of a city rather than pacifying the entire city. ...
[Why] would you waste effort pacifying a city when the enemy army in the field still fights and limited "military" control is sufficient to carry on that central campaign?
The recounting of the Sadr City battle--essentially assaulting an enemy-held city and not an urban counterinsurgency mission--also includes the pressure by the Iraqi government not to use firepower too much to carry out the missions. This alone argues against securing the entire city if we don't have to. And then there is the media reaction to civilians killed as "collateral damage" when fighting enemy forces. Why provide more opportunities for that source of condemnation if we don't need to?
If we only need to take part of the city as I suggest, walling off the parts of the city we don't need to take would be a useful tactic (and not the only one, given city undergrounds, sewers, and subways that would need to be blocked) for force protection and for containing the remaining enemy, like Japanese troops bypassed to wither on the vine in the World War II island-hopping campaign.
And yes, sometimes we will have to take an entire city. So my desire to avoid doing it doesn't mean we shouldn't know how to effectively take a city even though I'd prefer to take just the key parts.
Do read all of the article. It is quite good.
As an aside, why wasn't this approach used in the battle to liberate Mosul from ISIL? I certainly don't remember any mentions of wall-building during the long battle.
UPDATE: Casualty claims inflated by advocacy groups will harm any military effort in urban areas filled with civilians. So my view is minimize our efforts in cities to what is needed for the campaign.
