What the Hague Convention on Child Abduction Actually Says

The 1980 Hague Convention on Child Abduction is the treaty that governs how a wrongfully taken child comes home. This page gives the treaty's own words, the official commentary courts treat as binding, the four places every real case is fought, and the one gap almost every parent misses — Article 38. Every quotation below is verified against the official HCCH text; nothing is paraphrased and presented as a quote.

Reviewed by the SafeReturn Alliance Research Team · sources cited throughout · see how we verify · last verified against official HCCH sources 2026-07-15

The core promise, in the treaty's own words

The Hague Convention on Child Abduction's objects, stated in Article 1, are "to secure the prompt return of children wrongfully removed to or retained in any Contracting State" and "to ensure that rights of custody and of access under the law of one Contracting State are effectively respected in the other Contracting States." Its Preamble is worth reading closely, because of what it does not say: it speaks of "the interests of children" — plural and generic — not "the best interests of the child." Courts rely on exactly that distinction when holding that a return hearing is not a full welfare inquiry.

Wrongfulness itself is defined narrowly in Article 3: a removal or retention is wrongful only where it breaches "rights of custody" under the law of the child's habitual residence, and those rights "were actually exercised, either jointly or alone, or would have been so exercised but for the removal or retention." Article 5 then supplies the Convention's own definitions — and nine words inside it turn out to matter more than almost anything else in the treaty: "rights of custody" include, "in particular, the right to determine the child's place of residence."

The six weeks is not a deadline

This is the most consistently misreported provision in the whole Convention. Article 11 requires that "the judicial or administrative authorities of Contracting States shall act expeditiously," and that if no decision has been reached within six weeks, the applicant "shall have the right to request a statement of the reasons for the delay." That is the entire mechanism. It is a duty to explain, not a deadline — and it is the structural reason the global average is now 207 days, not forty-two (see The Six-Week Promise for the full data picture, and the underlying statistics).

What the exceptions actually say

Article 12 requires the authority to "order the return of the child forthwith" where less than a year has passed since the wrongful removal, and even after a year "unless it is demonstrated that the child is now settled in its new environment." Article 13 then sets out the grounds a court may weigh against return: that the person opposing return "was not actually exercising the custody rights… or had consented to or subsequently acquiesced" (13(a)); that "there is a grave risk that his or her return would expose the child to physical or psychological harm or otherwise place the child in an intolerable situation" (13(b)); and, in a free-standing paragraph that is not a lettered sub-clause despite common shorthand, that the child "objects to being returned and has attained an age and degree of maturity at which it is appropriate to take account of its views." See verified examples of how these exceptions actually played out in our case database.

Two provisions complete the frame. Article 16 bars the country the child was taken to from deciding custody on the merits until a return decision is made. Article 19 states plainly that a return decision "shall not be taken to be a determination on the merits of any custody issue." Together they encode the Convention's whole bargain: a return court decides where the custody question will be answered — never what the answer is.

Pérez-Vera: the commentary courts treat as binding

When the Convention was adopted in 1980, the Hague Conference commissioned an Explanatory Report from Elisa Pérez-Vera. It is not itself law, but courts from Washington to Strasbourg cite it as the authoritative aid to interpretation.

"[The Article 13 exceptions] must be applied only so far as they go and no further," and are "to be interpreted in a restrictive fashion."Pérez-Vera Explanatory Report, para. 34 — the single most-quoted line in the field, also quoted verbatim by the European Court of Human Rights.

The Report also explains, deliberately, what the Convention does not define: "habitual residence" — the treaty's entire jurisdictional gateway — is treated as "a question of pure fact, differing in that respect from domicile." That omission was not an oversight; it is why courts today apply a fact-driven test rather than a rigid legal formula.

The four places every real case is fought

1. Habitual residence — the undefined gateway

For years courts split between a parental-intention approach and a child-centred, fact-driven approach. That split has largely closed — toward fact. In Monasky v. Taglieri (2020), the US Supreme Court held that habitual residence "depends on the totality of the circumstances specific to the case," and that no actual agreement between the parents is required — a rule that matters directly in abuse cases, since an agreement requirement "would enable a parent, by withholding agreement, unilaterally to block any finding of habitual residence for an infant." English and EU courts reached a similar destination: the test is "the place which reflects some degree of integration by the child in a social and family environment," and it is now well settled that a child does not lose one habitual residence without normally acquiring another. Read the full case analysis: Monasky v. Taglieri.

2. Rights of custody — and the ne exeat expansion

The return remedy is triggered only by breach of a right of custody, never by breach of a mere right of access — a deliberate drafting choice. The hardest question this raises: does a parent's veto over a child's relocation (a "ne exeat" right), without day-to-day care, count as a right of custody? In Abbott v. Abbott (2010), the US Supreme Court held 6–3 that it does, reasoning directly from Article 5's "right to determine the child's place of residence." English courts draw a finer line: an actual veto is a right of custody, but a mere right to ask a court to object is not — and a narrow category of "inchoate" custody rights can protect a long-term carer, such as a grandmother, who was never a legal guardian. Read the full case analysis: Abbott v. Abbott.

3. Grave risk — and the domestic-violence collision

This is the field's live front line. Pérez-Vera's "narrow construction" is the starting orthodoxy — but the UK Supreme Court, in Re E (2011), expressly declined to treat narrow construction as an extra legal test, holding instead that the words of Article 13(1)(b) mean what they say. Re E also established that "grave" qualifies the risk, not the harm — a sliding scale, not a fixed bar. On domestic violence, the HCCH's own Guide to Good Practice and the case law now agree that violence against the taking parent can create a grave risk to the child — while also making clear that allegations of violence are not, by themselves, sufficient; the decisive question is what will actually happen to the child if returned. The entire fight then moves to protective measures — can the risk be neutralised by undertakings or mirror orders?

"Voluntary undertakings are not easily enforceable, and therefore may not be effective…"HCCH, Guide to Good Practice under the 1980 Convention, Part VI — Article 13(1)(b) (2020)

In 2022, the US Supreme Court in Golan v. Saada went further, holding unanimously that a court is "not categorically required to examine all possible ameliorative measures" before denying return — grave risk and amelioration are separate questions.

4. The remaining exceptions

Two more limits matter in practice. Article 12's "settled" exception only opens after a year — and in Lozano v. Montoya Alvarez (2014), the US Supreme Court held there is no equitable extension of that year even where the taking parent concealed the child; the clock still runs. And Article 20 — the human-rights exception — sits above even Article 13: Pérez-Vera holds it is not enough to show return would be incompatible, "even manifestly incompatible," with the requested State's fundamental principles. In four decades it has almost never succeeded standing alone.

Strasbourg's near-miss

In Neulinger and Shuruk v. Switzerland (2010), the European Court of Human Rights used language suggesting an "in-depth examination" of the child's whole situation before any return — language that, read literally, would have collapsed the Convention's summary procedure into a full custody trial. Three years later, in X v. Latvia (2013), the Grand Chamber walked it back: the national court must genuinely take an arguable grave-risk allegation into account and give reasoned reasons — not re-try custody. Article 8 of the European Convention on Human Rights requires reasons, not a welfare trial. Read the full analyses: Neulinger and Shuruk v. Switzerland and X v. Latvia.

The structural mismatch

The Convention was drafted against a mental picture: a non-custodial parent snatches the child and flees. The HCCH's own statistics describe a different world. In the most recent global study, 75% of taking parents were mothers — the highest ever recorded — and 88% were the child's primary or joint carer. The instrument's core move — return the child, strip the taker of any advantage — was built to defeat a strategic abductor. It is now applied, overwhelmingly, to the parent who was already raising the child. That is not evidence the Convention has failed; children are still wrongfully taken and still need to come home. It is evidence that Article 13(1)(b) is carrying a structural load its drafters never designed it to carry — precisely why that single provision is where nearly all of today's litigation lives. See the country-by-country picture in our Accountability Index.

The gap almost every parent misses: Article 38

103 States are Contracting Parties to the Hague Convention on Child Abduction. That number reassures people, and it should not — because being a "Contracting Party" does not always mean the Convention is in force with any given other country.

An accession "will have effect only as regards the relations between the acceding State and such Contracting States as will have declared their acceptance of the accession."Article 38, 1980 Hague Convention

Two States can therefore both be Contracting Parties and still have no Convention relationship with each other at all. This is not theoretical: of the 103 parties, 70 joined by accession — and some of them have been accepted by only a handful of the other 102. Central Authorities do reject return applications on exactly this ground.

This is the single most actionable fact on this page. The right question is never "is X a Hague country?" — it is "is the Convention actually in force between my country and theirs, and was it already in force on the date my child was taken?" We built a definitive, per-country-pair answer, verified line-by-line against the official HCCH acceptance table: check whether the Convention is in force between your two countries →

Where this leaves a parent right now

Three things follow directly from everything above. First, speed matters more than any single legal argument — because the data shows time consistently works against return, contact your Central Authority and a qualified local lawyer immediately (see our first-24-hours guide). Second, a return order is about which country's courts hear the case — never about final custody — understanding that early prevents both false hope and false despair. Third, confirm the Convention actually applies to your specific pair of countries before relying on it — the Article 38 gap above is exactly why. For more on how the Hague Convention on Child Abduction works in practice, see our FAQ and about SafeReturn Alliance.

Frequently asked questions

Does Article 11 give a binding six-week deadline? No. It creates a duty to act expeditiously and, if six weeks pass, a right to ask for the reasons for the delay — not a binding deadline. That structural weakness is why the global average is now 207 days.

What is a right of custody, and does a travel-veto count? Article 5 defines rights of custody to include the right to determine the child's place of residence. In Abbott v. Abbott (2010) the US Supreme Court held that a ne exeat travel-veto right is therefore a right of custody, even without day-to-day care.

How narrow is the Article 13(1)(b) grave-risk exception? The Pérez-Vera Report says the exceptions must be interpreted restrictively. But the UK Supreme Court in Re E (2011) declined to treat narrow construction as an added legal test, and held that grave risk is a sliding scale, not a fixed bar.

Being a Contracting Party — does that mean the Convention is in force with my country? Not necessarily. Under Article 38, an accession only takes effect between the acceding State and States that have expressly accepted it. Two Contracting Parties can have no Convention relationship at all — use our per-country checker to confirm your specific pair.