The Biological Weapons Convention creates no body to inspect or verify compliance, and the working group set up in 2022 to strengthen it can report only by consensus

What is happening?

The Biological Weapons Convention bans the development, production and stockpiling of biological weapons. It opened for signature on 1972-04-10 and entered into force on 1975-03-26. The page of the UN Office for Disarmament Affairs, as viewed on 2026-10-10, lists 189 states parties and 4 signatories.

The treaty gives its parties two ways to respond to a doubt about compliance. Under Article V they consult and cooperate with one another. Under Article VI a party that finds a breach may lodge a complaint with the UN Security Council, with evidence, and all parties undertake to cooperate with any investigation the Council initiates. Article VII adds assistance to a party endangered by a violation. The full text read for this dossier contains no article that creates an independent body to inspect, investigate or verify compliance. An investigation path therefore exists, but it runs through the Security Council and not through an organ of the treaty itself.

Negotiations on a verification protocol ended without agreement in 2001, and secondary sources describe roughly two decades in which verification was not discussed inside the treaty framework. In 2022 the Ninth Review Conference set up a working group on strengthening the Convention, with compliance and verification as one of seven topics. That decision requires the group to work by consensus and to submit a report adopted by consensus to the next review conference. As of the revised draft final report dated 2026-02-07, parts of the text were still marked as only provisionally approved and bracketed language remained.

Whose problem is this?

The parties who would have to create a verification organ are the same parties whose conduct it would examine, and the rule under which they decide lets the text move only when none of them objects.

Where does this problem end?

The boundary is drawn around structure rather than conduct. The dossier records that the treaty was built without a checking body and that the forum meant to strengthen it can only act when every party agrees.

What is the state now, and what should it be?

Now

What it should be

The only target set by the parties themselves is procedural. The Ninth Review Conference decided that the working group would submit a report adopted by consensus, with its recommendations, to the Tenth Review Conference or a special conference. Press reports place the Tenth Review Conference in 2027 and say that conference will decide which recommendations take effect.

No document adopted by the parties, among those opened here, sets a target for compliance verification itself — no date by which a verification body should exist, and no description of what it should do. This dossier does not fill that gap with the positions of researchers or advocates.

How big is it?

The scale here is the size of the institution, not a count of affected people. The treaty binds 189 states parties, according to the UNODA page as viewed on 2026-10-10.

The body that supports those parties between meetings is small. A nonprofit science and security journal reported in March 2024 that the Implementation Support Unit had 4 staff and a 2023 budget of about 2.1 million US dollars, and set those figures beside more than 32 million dollars for the Organisation for the Prohibition of Chemical Weapons in 2022 and more than 166 million dollars for the International Atomic Energy Agency in 2022. Those comparison figures come from the same secondary report and carry its dates; no current staffing or budget figure was found in an opened primary document.

The working group itself was given fifteen meeting days a year from 2023 through 2026, sixty days in all, to cover seven topics of which compliance and verification is only one.

An affected population is not derived. The people who would be harmed by an undetected programme cannot be counted from these sources, and the treaty covers every state party at once.

Under what conditions does it arise?

1. The treaty was written with consultation and complaint, not inspection. Article V lets parties ask each other questions and Article VI lets a party take a complaint to the Security Council. Neither creates a standing body with its own authority to look, so whether any check happens depends on a party choosing to raise a concern and on a body outside the treaty choosing to act on it.

2. The attempt to add verification failed once and was not resumed for a long time. The protocol negotiations ended without agreement in 2001. Sources give different explanations for that outcome and this dossier does not attribute it to any party.

3. The forum that could recommend a remedy needs every party to agree. The working group set up in 2022 must reach consensus, and its final report must be adopted by consensus. One objection on any section can hold the whole report, and the draft of 2026-02-07 still carried bracketed language.

4. Verification competes for time inside a broad agenda. The working group covers seven topics, of which compliance and verification is one. A nonprofit arms-control journal reported in December 2024 that about ten meeting days of the schedule were assigned to compliance and verification.

5. The secretariat is administrative and small. The Implementation Support Unit handles administrative support, universalisation and the exchange of confidence-building measures, and its reported size was four staff in 2023.

What has been tried?

The draft recommendations, as far as the opened pages show, would set up two mechanisms, one to review developments in science and technology and one to promote international cooperation under Article X. Whether the final report will contain any compliance or verification arrangement had not been settled in any document opened here.

What was found?

Why is it still unsolved?

Institutional gap — the ban exists, but no organ was created to check that it is kept, and the forum that could propose one moves only when every party agrees.

A complaint under Article VI depends on one party deciding to lodge it and on the Security Council deciding to investigate. Consultation between parties depends on the parties being willing to ask and to answer. Neither path produces a routine, impartial look at what any party is doing.

The second layer is procedural. The working group mandated in 2022 can only report what all parties accept, and its final report has to be adopted as a whole. Agreement on most of a draft does not make any of it final, because a report that is mostly agreed is still not adopted. The opened draft of February 2026 shows the shape of this directly, with agreed passages sitting beside provisional and bracketed ones.

The third layer is capacity. The secretariat that supports the treaty is administrative and was reported at four people in 2023. Even a decision to create a verification function would have to build an institution almost from nothing, and the decision itself has not been made in any document opened here.

What observation would mean it is solved?

Candidates — (a) the working group adopts its final report by consensus and that report contains an arrangement on compliance or verification (b) a review conference decides to establish a treaty body with a mandate to investigate or verify compliance (c) such a body exists with staff, budget and procedures, and its activity is reported.

(a) alone is weaker than it looks. A consensus report can be adopted precisely because its compliance section was left general, and adoption by the working group does not by itself put anything into effect — press reports say that decision belongs to the next review conference.

(b) alone counts paper. A decision to establish a body can leave its powers, funding and start date for later meetings, each of which needs agreement again.

(c) is the observation that matters, but it has to be read with (b). A body that exists without authority to look anywhere a concern arises would be another consultation forum under a different name. The three have to be read together.

What is it connected to?

Fills with researchthe verification structures of the chemical and nuclear regimes, the mechanism of the UN Secretary-General for investigating alleged use, international health preparedness instruments, and the governance of dual-use life sciences research. Relation types and evidence grades for these links were not confirmed in this round.

What these sources do not say

• The official outcome of the ninth session in August 2026. The only account is a trade publication report. No outcome document from that session was opened, and the result of the December 2026 session does not yet exist.
• Whether the final report will include any compliance or verification arrangement. The opened pages of the 2026-02-07 draft describe two mechanisms, neither of them a verification body, and only the first six pages were read; the sources do not say what the remaining text or later drafts provide on compliance.
• The current staff and budget of the Implementation Support Unit. The four staff and 2.1 million dollars are 2023 figures from a secondary report, and no current figure appears in the sources.
• The span behind the ten meeting days. The working group was allotted fifteen days a year from 2023 through 2026, sixty in all. The secondary report that gives about ten days for compliance and verification does not, in the form opened here, make clear which years that count covers, and this dossier does not compute a share.
• Why the 2001 protocol negotiations failed. Sources attribute the outcome differently and none of them is a public determination by a treaty body. This dossier does not attribute it.
• Whether any state party has breached the treaty. No opened source offers an independently verified finding, which is the subject of this dossier rather than a gap in it. No allegation is carried here.
• The published positions of individual states parties. The working group, its chair and the support unit describe procedure. No document setting out the position of a particular party on verification was found among the sources for this dossier, including the party reported to have objected to contributions from non-state entities, which is not named here.
• How often Article VI has been used. None of the opened sources counts complaints lodged with the Security Council under the treaty.

See the evidence

Primary documents read. The treaty text, the two UNODA pages and the revised draft final report of 2026-02-07 were opened directly; of the draft, the first six pages were read. Everything about the 2025 outcome, the ninth session of August 2026, the size of the support unit and the history of the 2001 negotiations comes from secondary sources and is graded accordingly. Where secondary sources agree they are stated plainly; where only one source carries a fact, the dossier says so in the grade column. This is a Path A output (research-based definition), so observation_refs is empty and provenance_mode: press-derived.

People affected

Estimated range Not derivable

The reason and what is missing are listed under “What is missing” below