Meridian, home

Research / MER/26/D20

Research Paper

Norm under Strain: The Cluster Munitions Ban after Vientiane

MER/26/D20 · Meridian Research Team ·

Norm under Strain: The Cluster Munitions Ban after Vientiane

The Third Review Conference of the Convention on Cluster Munitions, held in Vientiane from 14 to 18 September 2026, adopted a Review Document on the Lausanne Action Plan period, a Political Declaration, and the Vientiane Capital Action Plan for 2027-2031. Hosted by the most heavily affected State Party, the meeting reaffirmed a prohibition that has delivered complete destruction of declared stockpiles among obligated States Parties, while confronting the first withdrawal from a multilateral humanitarian disarmament treaty that bans an entire class of weapons, slower universalisation, and continued use and transfers by States not party. This paper separates treaty obligations from action-plan commitments, reads the Vientiane package against the Convention's articles, and sets tests for whether the new five-year roadmap can protect the norm without mistaking conference adoption for compliance in conflict.

Key findings and scope

  • The Convention entered into force on 1 August 2010. As of the States Parties listing consulted on 6 November 2026, there are 112 States Parties and 12 Signatories [1]. The Presidency's draft Review Document for the Conference states that 124 States have committed (112 Parties and 12 Signatories) [2, paragraph 21].
  • According to that Review Document, all 41 States Parties with stockpile-destruction obligations have completed destruction and declared compliance with Article 3, including States that finished after the Lausanne cycle began. States Parties have destroyed nearly 1.5 million cluster munitions and 179 million submunitions [2, paragraphs 39-40].
  • Lithuania's withdrawal took effect during the review period under Article 20, described in the Review Document as the first-ever withdrawal from a multilateral humanitarian disarmament treaty prohibiting an entire class of weapons [2, paragraphs 23-24].
  • The Lao People's Democratic Republic, as Presidency, reported that the Conference brought together more than 500 delegates from 91 countries, including 74 States Parties, 17 observer States and 27 international organisations, and adopted the Review Document, Political Declaration and Vientiane Action Plan containing 43 action lines for 2027-2031 [3; 4].
  • Timor-Leste signed a declaration of commitment to accede, according to the host Presidency's closing account. That is a political commitment to join, not yet accession completed on the treaty depositary record consulted for this paper [3].

Documents opened for this paper include the Convention framework as presented by the implementing community, the Presidency's July 2026 draft Review Document and draft Vientiane Capital Action Plan, the Implementation Support Unit's 2025 annual report, the Holy See's interpretative declaration upon adoption of the Final Report, and the host Presidency's official closing statement. Draft conference papers are used for the substance later confirmed as adopted by the Presidency and by the Holy See's interpretative declaration. Where a figure appears only in a draft and has not been independently re-tabulated from Article 7 reports after the Conference, it is attributed to that draft. No claim is made about the legality of any particular strike or transfer outside the Convention's membership.

1. What the Convention actually prohibits and requires

The Convention on Cluster Munitions was adopted in Dublin on 30 May 2008 and opened for signature in Oslo on 3 December 2008. It became binding international law on 1 August 2010 [1]. Its core is a comprehensive ban: States Parties undertake never under any circumstances to use, develop, produce, otherwise acquire, stockpile, retain or transfer cluster munitions, and not to assist, encourage or induce anyone to engage in those activities. Positive obligations cover stockpile destruction, clearance and destruction of cluster munition remnants, risk education, victim assistance, and international cooperation and assistance [2, paragraphs 1-2; 5].

That package is stricter than a use-only ban. Stockpile destruction and clearance deadlines, transparency reporting under Article 7, and victim-assistance duties make the Convention a humanitarian-disarmament instrument with measurable implementation tracks. Action plans adopted at Review Conferences are not themselves legally binding. The Vientiane Capital Action Plan states expressly that while not legally binding, its actions are intended to support effective, transparent and result-oriented implementation, with Article 7 reports remaining the primary monitoring means [4, paragraphs 4-5]. Confusing the Action Plan with the treaty is a common analytical error. The Plan organises political attention; the Convention creates obligations.

Convention on Cluster Munitions commitmentsStates Parties112Signatories12
Figure 1. States Parties and Signatories to the Convention on Cluster Munitions as listed on the Convention’s States Parties page consulted on 6 November 2026. Source: Convention on Cluster Munitions, States Parties [1]. Chart drawn by Meridian from the published figures.

2. Stockpile destruction: a completed track with residual vigilance

The Review Document records that States Parties have destroyed nearly 1.5 million cluster munitions and 179 million submunitions, and that as of mid-2026 all 41 States Parties with stockpile-destruction obligations had completed destruction and declared Article 3 compliance. Fifteen of those completions occurred after the Lausanne Action Plan period began. The most recent completions demonstrate that the destruction track continued to deliver during the review cycle [2, paragraphs 39-40]. The Implementation Support Unit's 2025 annual report likewise treats declared stockpile destruction by States Parties as completed, shifting emphasis toward sustaining Article 3 compliance, retention limits for training and development under Article 3.6, and transparency [6].

Completion of declared stockpile destruction is a major humanitarian-disarmament result. It is not the end of Article 3 work. Previously unknown stockpiles have been discovered and destroyed, including by the United Kingdom as cited in the Review Document. Retention for permitted training and detection development must remain limited and transparent. Transfer provisions under Article 3(7) have been used for destruction assistance. These are maintenance tasks, not unfinished primary destruction [2, paragraphs 40-42; 6].

Table 1. Implementation tracks and the evidence that supports them

Track Evidence in opened documents Limits of that evidence
Membership 112 States Parties and 12 Signatories on the Convention listing; Review Document cites 124 committed States Listing pages and draft review tallies can diverge by one; depositary status is authoritative for legal membership
Stockpiles All 41 obligated States Parties declared Article 3 completion; ~1.5 million munitions and 179 million submunitions destroyed Figures are declared and reported totals, not an independent physical audit by Meridian
Clearance Extension requests continue, including Chad's Article 4 request considered at the Conference Clearance completion is uneven; deadlines extended are not failures by definition but require scrutiny
Norm strain Lithuania withdrawal under Article 20; continued use/transfer/production by States not party recorded as concern Political condemnation in an action plan is not adjudication of specific incidents
Forward plan Vientiane Capital Action Plan, 43 actions, 2027-2031 Not legally binding; success depends on national budgets and reporting

Sources: [1]; [2]; [3]; [4]; [7]. Missing values are marked as limits, not as zero.

3. Universalisation after the first withdrawal

The Review Document is candid about universalisation. Progress continued, but accession slowed. Lithuania's withdrawal is described as taking effect pursuant to Article 20 and as the first withdrawal from a multilateral humanitarian disarmament treaty prohibiting an entire class of weapons. States Parties expressed deep regret, reiterated readiness for dialogue, and encouraged Lithuania to rejoin [2, paragraphs 21-24]. The Vientiane draft Action Plan likewise acknowledges recent challenges to the norms established by the Convention, including the withdrawal of a State Party and new transfers and use of cluster munitions, and states that Parties condemn use by any actor under any circumstances while recalling their obligation to discourage use, development, production, stockpiling and transfer by States not party [4, paragraph 4].

Meridian's reading is that withdrawal is legally available under the Convention's own terms and politically costly for the norm. Treating withdrawal as legally impossible would misstate Article 20. Treating it as normatively trivial would misstate the Review Document's own language. The practical question for 2027-2031 is whether universalisation work concentrates on Signatories and receptive States not party, including Timor-Leste's declared commitment to accede [3], while maintaining a principled response to use by non-parties without pretending the Convention binds them.

Voting patterns on the annual United Nations General Assembly resolution, regional workshops and outreach by the Implementation Support Unit remain the measurable universalisation tools identified in the Implementation Support Unit report and Review Document [2; 6]. They are influence tools, not substitutes for ratification.

4. Clearance, risk education and victim assistance in the most affected host State

Hosting the Review Conference in Vientiane was symbolic and substantive. Lao PDR remains among the most heavily contaminated countries. The Presidency's closing account emphasises unexploded ordnance as a generational national priority and presents the Conference theme as accelerating progress toward a world free from the threat of cluster munitions [3]. The Conference considered Chad's Article 4 deadline extension request, with analysis and a draft decision circulated among official documents [7; 8]. Extension requests are a designed feature of the clearance regime, not automatic evidence of bad faith. They are, however, a stress indicator when contamination, funding and capacity are misaligned.

The draft Action Plan's cross-cutting actions emphasise national ownership, evidence-based implementation, predictable resources, and inclusive approaches that take account of gender, age and disability [4, section II]. The Review Document finds uneven victim participation in decision-making: nine States Parties reported inclusive strategies involving victims or affected communities, while only one included victims in official delegations during the measured Lausanne-period snapshot [2, paragraph 11]. That gap matters for a treaty that places victim assistance among its core positive obligations. An action plan that repeats inclusion language without changing delegation practice will not close it.

Environmental integration is a newer emphasis. The Review Document notes IMAS 07.13 (2024) and a 2025 technical note supporting environmental protection in mine-action planning, with national examples of integration [2, paragraphs 15-16]. The Vientiane package's attention to environmental and climate considerations, highlighted by the Presidency, extends that trend [3; 4]. These are process improvements. They do not themselves clear land.

5. Finance, the Implementation Support Unit and institutional housekeeping

Financial sustainability is described as mixed. The Review Document records an average of 31 States Parties paying assessed contributions annually in the measured period, and notes ongoing attention to Implementation Support Unit work plans and budgets [2, paragraph 16; 9]. The Unit's 2027-2031 revised work plan and budget were before the Conference [9]. International cooperation and assistance remained relatively stable, with an average of 23 States Parties providing support annually and about nine providing multi-year assistance [2, paragraph 10]. Predictability, not only annual totals, is the binding constraint for clearance and victim-assistance planning.

Institutional questions on the future programme of meetings, the Coordination Committee and sponsorship guidelines were also on the table through Presidency working papers [8]. These are not distractions from humanitarian outcomes. An underfunded Implementation Support Unit and irregular meeting cycle would weaken the transparency system on which the Action Plan depends.

6. The strongest objections

Objection 1: the ban is eroding because major users remain outside

The strongest sceptical case is that a Convention without the largest producers and users cannot stop the weapons that dominate current casualty narratives, and that withdrawal plus non-party use shows the norm failing. The Review Document itself records deep concern about increased civilian casualties from continued use and about transfers and production by States not party [2, paragraphs 31-32; 4, paragraph 4]. Meridian's assessment is that the objection correctly identifies the Convention's external limit. It understates the internal accomplishment of destroying declared stockpiles among Parties and the stigma effects that can still shape some non-party behaviour. A treaty can be both incomplete in coverage and effective among its Parties. Those are different tests.

Objection 2: action plans multiply while land remains contaminated

A second objection is that five-year plans create an illusion of progress through indicator language while extension requests accumulate. The Lausanne-period figures show institutional improvements (national strategies, workplans, standards) alongside persistent clearance burdens [2, paragraphs 7-14]. The objection holds where funding is short-term and reporting is treated as success. It fails where Article 4 extensions are paired with transparent, costed workplans and survivor-centred assistance. Vientiane's emphasis on domestic resource mobilisation alongside international support is the right diagnostic; delivery will show whether it is more than diagnostic [3; 4].

Objection 3: condemning all use is empty without enforcement

The Action Plan's condemnation of use by any actor is politically important and legally limited. The Convention's compliance provisions and meetings machinery are not a criminal court. States Parties can investigate, assist, publicise and, in extreme cases, pursue other international legal avenues. They cannot, through the Action Plan alone, impose the ban on non-parties. Honest communication requires keeping that limit visible.

Objection 4: hosting by the most affected State risks symbolism without resources

Choosing Vientiane raised the political profile of contamination and survivor experience. Symbolism without a financed clearance and victim-assistance surge would be extractive. The Presidency's own framing, tying the Action Plan to predictable financing and affected-community participation, acknowledges that risk [3]. Observers should track Lao clearance and victim-assistance finance in Article 7 reporting after 2026 as a concrete test of whether hosting changed resource flows.

6A. Reading the Vientiane Capital Action Plan as a management instrument

The draft Action Plan's introduction links Convention implementation to the Sustainable Development Goals, the Convention on the Rights of Persons with Disabilities, and the women, youth and children agendas, while insisting that Article 7 transparency reports remain the primary monitoring tool [4, paragraphs 2-5]. That dual register matters. It invites mainstreaming without allowing thematic co-benefits to replace countable destruction, clearance and assistance results. Cross-cutting Action #1 and the following cross-cutting actions emphasise national ownership, institutional frameworks, coordination, evidence-based implementation, predictable resources and inclusive approaches [4, section II]. Thematic sections then organise universalisation, stockpile destruction and retention, clearance, risk education, victim assistance, international cooperation and national implementation into numbered actions through Action #43 [4].

Meridian's management reading is that 43 actions are useful only if presidencies and coordinators select a smaller set of annual priorities for Meetings of States Parties. An undifferentiated checklist of 43 equally weighted items will recreate the reporting burden problem familiar from other humanitarian-disarmament plans. The Review Document's own Lausanne-period snapshot already shows uneven victim inclusion and limited systematic integration into broader humanitarian and development frameworks [2, paragraphs 11-14]. The Vientiane Plan should be judged by whether those weak cells improve, not by whether States can quote action numbers in statements.

The Holy See's interpretative declaration upon adoption confirms that the Review Document, the Vientiane Capital Declaration and the Action Plan were adopted as a package, while recording a distinct interpretative position on gender terminology [10]. That confirmation is important for sourcing: July advance copies can be read as the substance of the adopted package, with the host Presidency's closing account supplying attendance and political framing [3; 10]. It does not make the Action Plan legally binding. States Parties that treat adoption as the end of implementation planning will have misread paragraph 5 of the Plan itself [4].

6B. Clearance pressure and the meaning of extension requests

Article 4 clearance deadlines are among the Convention's most visible compliance clocks. Chad's extension request, analysed for the Conference, illustrates the ordinary legal pathway when contamination, survey quality and funding misalign [13]. Extensions are not amnesties. They are contingent grants that should come with revised milestones, resource plans and transparency. A Conference that grants extensions without tightening expectations would weaken Article 4. A Conference that refuses extensions without regard to residual risk and capacity would punish honesty. The documents opened here show the machinery functioning; they do not, by themselves, prove that every extended deadline is on a credible path.

Risk education and survey quality are the quiet determinants of whether clearance statistics mean safer land. The Review Document and Action Plan both treat risk education as a distinct pillar rather than a public-relations annex [2; 4]. In highly contaminated rural economies, failure to sustain risk education after a surge of donor-funded clearance can erase gains. That is an implementation risk for Lao PDR and other affected Parties during 2027-2031, independent of conference drafting quality [3].

7. Tests for 2027-2031

  1. Membership: Do Signatories and States that issued accession commitments, including Timor-Leste if it proceeds, complete depositary formalities?
  2. Withdrawal diplomacy: Is there documented engagement aimed at return, without diluting Article 1 obligations for remaining Parties?
  3. Clearance: Are Article 4 extensions tied to measurable annual milestones and funded workplans?
  4. Victims: Does the number of States Parties including survivors in official decision-making rise above the Lausanne-period snapshot?
  5. Finance: Do multi-year assistance commitments and assessed-contribution payment rates improve relative to the Review Document baseline?

8. Conclusion

Vientiane did not repair the Convention's coverage gaps, and it could not. It did consolidate a completed stockpile-destruction story among obligated Parties, name the first withdrawal as a normative shock, and adopt a 43-action roadmap that is only as strong as national budgets and Article 7 honesty. The cluster-munitions ban remains a strict prohibition for those inside it and an incomplete stigma for those outside it. Whether 2027-2031 is remembered for accelerated clearance in places such as Lao PDR, or for further erosion around the edges of membership, will not be decided by the quality of conference drafting. It will be decided by whether the Vientiane actions are funded, reported and defended when the next hard case arrives.

References

  1. Convention on Cluster Munitions, "States Parties" listing, https://www.clusterconvention.org/states-parties/ (accessed 6 November 2026).
  2. Presidency of the Third Review Conference, Draft Review Document on the implementation of the Lausanne Action Plan (2022-2026) and the status and operation of the Convention, CCM/CONF/2026/WP.1, 17 July 2026, https://www.clusterconvention.org/wp-content/uploads/2026/07/CCM-CONF-2026-WP.1-Advance-Copy.pdf (accessed 6 November 2026).
  3. Lao News Agency (KPL), "Lao PDR Concludes Successful Presidency of the Third Review Conference on Cluster Munitions", 19 September 2026, https://kpl.gov.la/En/detail.aspx?id=106546 (accessed 6 November 2026).
  4. Presidency of the Third Review Conference, Draft Vientiane Capital Action Plan on the implementation of the CCM, 2027-2031, CCM/CONF/2026/WP.2, 17 July 2026, https://www.clusterconvention.org/wp-content/uploads/2026/07/CCM-CONF-2026-WP.2-Advance-Copy.pdf (accessed 6 November 2026).
  5. Convention on Cluster Munitions, Convention text portal, https://www.clusterconvention.org/convention-text/ (accessed 6 November 2026).
  6. Convention on Cluster Munitions Implementation Support Unit, 2025 Annual Report, CCM/CONF/2026/6, https://www.clusterconvention.org/wp-content/uploads/2026/09/CCM-CONF-2026-6-CCM-ISU-2025-Annual-Report.pdf (accessed 6 November 2026).
  7. United Nations Office for Disarmament Affairs, "Convention on Cluster Munitions, Third Review Conference (2026)", Meetings Place, https://meetings.unoda.org/ccm-revcon/convention-on-cluster-munitions-third-review-conference-2026 (accessed 6 November 2026).
  8. Convention on Cluster Munitions Third Review Conference documents page (Presidency), https://www.clusterconvention.org/third-review-conference-3rc/ (accessed 6 November 2026).
  9. CCM Implementation Support Unit, 2027-2031 Revised Work Plan and Budget, CCM/CONF/2026/WP.14, https://www.clusterconvention.org/wp-content/uploads/2026/09/CCM-CONF-2026-WP-14-CCM-ISU-2027-2031-Revised-Work-Plan-and-Budget.pdf (accessed 6 November 2026).
  10. Holy See, "Interpretative Declaration of the Holy See upon adoption of the Final Report of the Third Review Conference of the Convention on Cluster Munitions", Vientiane, 18 September 2026, https://www.clusterconvention.org/wp-content/uploads/2026/10/Holy-See-Interpretative-Declaration-CCM-3RC.pdf (accessed 6 November 2026).
  11. Presidency, CCM Political Declaration draft for consultations, CCM/CONF/2026/WP.3, 17 July 2026, https://www.clusterconvention.org/wp-content/uploads/2026/07/CCM-CONF-2026-WP.3-Advance-Copy.pdf (accessed 6 November 2026).
  12. Lao News Agency (KPL), "Third Review Conference of the Convention on Cluster Munitions officially opens", 14 September 2026, https://kpl.gov.la/en/detail.aspx?id=106407 (accessed 6 November 2026).
  13. Analysis and Draft Decision of Chad's Deadline Extension Request under Article 4.1, https://www.clusterconvention.org/wp-content/uploads/2026/10/Analysis-and-Draft-Decision-of-Chads-Deadline-Extension-Request.pdf (accessed 6 November 2026).
  14. Concept note on the Third Review Conference process and documents, March 2026, https://www.clusterconvention.org/wp-content/uploads/2026/03/CN3RC-Concept-note-on-Third-Review-Conference-Process-and-Documents.pdf (accessed 6 November 2026).
  15. Convention on Cluster Munitions, Vientiane Action Plan portal page, https://www.clusterconvention.org/vientiane-action-plan/ (accessed 6 November 2026).
  16. Convention on Cluster Munitions, Lausanne Action Plan portal page, https://www.clusterconvention.org/lausanne-action-plan/ (accessed 6 November 2026).
  17. ICBL-CMC, event page for the Third Review Conference, https://icblcmc.org/events/third-review-conference-of-the-convention-on-cluster-munitions (accessed 6 November 2026; used for civil-society framing of hosting in Lao PDR, not for legal membership counts).
  18. Presidency working paper CCM/CONF/2026/WP.6, elements for consideration on the future programme of meetings, https://www.clusterconvention.org/wp-content/uploads/2026/09/CCM-CONF-2026-WP.6-Advance-Copy-1.pdf (accessed 6 November 2026).