Convention on Certain Conventional Weapons

• 1. Introduction

  ›

  The Convention on Certain Conventional Weapons was concluded in Geneva on the 10th of October 1980, born from a question that arms negotiators had long avoided: what happens when the weapon itself is the problem, not just how it is used? Not nuclear warheads. Not chemical agents. Ordinary battlefield hardware, the kind armies have relied on for generations. Landmines buried in fields. Laser systems capable of destroying a soldier's eyesight in an instant. Bomblets that scatter across farmland and wait, sometimes for decades, to detonate under a child's foot.

  The treaty's full name runs to more than a dozen words. The core idea, though, is compact: some weapons cause suffering so disproportionate, or strike victims so indiscriminately, that they deserve their own rules. What exactly those rules cover, and how hard they bite, turns out to be one of the harder questions in modern international law. The story of the CCW is a story of what governments are willing to prohibit, what they insist on keeping, and where the line keeps moving.

• 2. Five Protocols, Five Weapons

  ›

  Protocol I reached the clearest consensus first. It prohibits weapons whose primary effect is to injure through fragments that cannot be detected in the human body by X-rays. The logic is medical as much as moral: invisible fragments are difficult to remove, and the suffering they extend serves no military purpose a visible fragment would not serve equally well. The protocol is careful, though, not to prohibit plastic components in weapons design more broadly; the ban applies only when injuring by undetectable fragments is the primary intent.

  Protocol III addresses incendiary weapons, those designed to set fire to objects or cause burn injuries through flame, heat, or chemical reaction. Striking civilian populations with such weapons is prohibited in all circumstances. Air-delivered incendiary weapons face an additional restriction: they may not be used against military targets located within concentrations of civilians. The protocol carves out an explicit exception for munitions like smoke shells, where any incendiary effect is secondary rather than the designed purpose.

  Protocol IV broke new ground when it was adopted in Vienna on the 13th of October 1995. It was the first treaty provision to prohibit a weapon not yet in widespread battlefield use. Laser systems specifically designed to cause permanent blindness are banned; the parties also agree not to transfer such weapons to any state or non-state entity. Systems where blinding is an incidental rather than intended effect remain legal, though parties must take all feasible precautions to avoid those side effects.

  The fifth and newest protocol, adopted in Geneva on the 28th of November 2003 and entering into force in 2006, addresses explosive remnants of war, the unexploded ordnance left behind after fighting ends. Cluster munitions scatter bomblets across wide areas; many fail to detonate on impact. Protocol V places the burden of clearance on the parties that used those weapons, and requires them to share information about where and how explosive weapons were deployed. Each party is responsible for the territory under its control once a conflict ends.

• 3. Landmines and the Ottawa Split

  ›

  Protocol II on mines, booby traps, and other devices has a history of its own. The original 1980 text regulated rather than banned landmines, and the convention's scope at that point covered only international armed conflicts. A 1996 amendment, which entered into force on the 3rd of December 1998, changed both of those things. It extended the protocol's reach to internal conflicts and tightened the rules considerably: non-detectable anti-personnel mines were prohibited, as was their transfer; non-self-destructing and non-self-deactivating mines could no longer be used outside fenced, monitored, and marked areas; and parties were required to remove mines and booby traps once a conflict ends.

  None of that added up to a total ban. The failure to reach consensus on eliminating anti-personnel landmines entirely pushed negotiations outside the CCW framework. The result was the Ottawa Treaty, concluded separately and holding states to a complete prohibition on anti-personnel mines. The CCW's Protocol II remains on the books alongside it, binding states that have not signed Ottawa and governing a category of weapons the CCW was never able to fully close off.

  Decades later, divisions persist on a related front. China and Russia have opposed restrictions on anti-vehicle mines, including a requirement that such mines self-deactivate, blocking consensus on that issue as recently as 2017.

• 4. Scope, Enforcement, and Limits

  ›

  When the convention was first adopted in 1980, its protocols applied only to international armed conflicts, the classic state-versus-state scenario. Two amendment conferences, held in 1996 and 2001, extended coverage to non-international conflicts as well, capturing the civil wars and internal insurgencies that had come to dominate armed violence in the intervening years. Some provisions reach even further, continuing to apply after active hostilities end; Protocols II and V both impose obligations that run into the post-conflict period, covering the ongoing danger from mines and unexploded ordnance.

  The convention's structural weakness is equally plain. The CCW has no verification mechanism and no formal process for resolving compliance concerns. A state party that decides to abandon the treaty cannot simply walk away; it remains legally bound until one year after notifying the treaty depositary, the UN Secretary-General, of its intention to leave. But with no body empowered to investigate violations or impose penalties, that one-year delay is the convention's main enforcement lever. As of 2017, states had failed to reach consensus on adding a compliance mechanism to correct this gap. A minimum of two of the five protocols must be adopted for a state to be considered a party to the convention at all. As of the end of July 2024, there are 128 state parties, though several have adopted only a subset of the protocols.

• 5. Autonomous Weapons and the Open Questions

  ›

  States began discussing lethal autonomous weapon systems under the CCW framework in 2014, drawn by the same logic that had brought laser weapons and cluster munitions before the convention: new military technology raising questions about indiscriminate effects and unnecessary suffering. In 2016, the convention's states parties established an open-ended Group of Governmental Experts specifically to continue those discussions, covering international humanitarian law, accountability, possible prohibitions and regulations, and the degree of human control required over AI-enabled weapons.

  The talks have not produced a protocol. As of 2021, most of the major powers oppose an international ban on lethal autonomous weapons, an alignment that mirrors earlier resistance to stronger landmine rules and suggests the CCW's familiar pattern: discussion as an alternative to binding obligation. What distinguishes the autonomous weapons debate from earlier CCW negotiations is that the technology under discussion is still developing. Protocol IV on blinding lasers was adopted before such weapons had seen battlefield use. Whether states will move as early on autonomous systems, or wait until the harm is documented and the weapons are already deployed, remains the treaty's most consequential open question.