The People Inside the Agreement

A person steps off an aircraft. The agreement is already old news to the people who signed it. For the person arriving, it has just become the place where tomorrow will happen.

Where will I sleep? Can I call my family? What am I allowed to do? Who will explain this letter? If I am afraid to go somewhere, who will listen? How long will I be here?

The answers matter before anyone knows whether the arrangement will be called a diplomatic success.

We should be careful with the category third-country national. In this discussion, it means a person transferred from the United States to a country of which they are not a national. It does not tell us that person’s complete legal history, character, needs or intentions. It does not establish a criminal conviction. It does not tell us whether the person has sought protection or whether a return would be safe.

A label that helps governments organize a process should not become a substitute for learning what an individual case requires.

The same care belongs to local people. A citizen who asks who pays for a programme is not automatically hostile to migrants. A public servant who identifies a capacity problem is not automatically obstructive. People can care about newcomers and still expect the government to make a defensible commitment.

The task is to build a system in which compassion and responsibility can work together.

Arrivals and people present

The first operational distinction is simple enough to fit on a kitchen calendar.

How many people may arrive during a quarter? How many people may be present and require support at one time?

These are different numbers.

Imagine an arrangement admitting ten people in its first quarter. None leaves during that period, so ten remain. In the second quarter, ten more arrive and two leave. Eighteen remain. In the third, ten arrive and four leave. Twenty-four remain. In the fourth, ten arrive and eight leave. Twenty-six remain.

The intake never exceeds ten in a quarter. The number present keeps growing because departures are fewer than arrivals.

Scroll sideways to see all columns. Keyboard: focus the table and use the left and right arrow keys.

Quarter	Arrivals	Departures	Present at end
Q1	10	0	10
Q2	10	2	18
Q3	10	4	24
Q4	10	8	26

Figure 3. An invented stock-and-flow example. People present at the end of a period equal people present at the start, plus arrivals, minus departures. These are not Saint Lucia programme counts.

The arithmetic does not tell us that a particular programme will accumulate cases in this way. It tells us what must be tracked if we want to know whether it is doing so.

A responsible capacity limit should therefore consider the people currently present, their needs, unresolved decisions and the services available. The ability to process ten straightforward cases is not necessarily the ability to support ten complex ones. The same headcount can represent very different demands on staff and resources.

This is why a cap needs a definition. Does it limit proposed cases, accepted cases, arrivals, people supported at once or all of these in different ways? Does an unused allocation carry forward? Who can authorize an exception? What happens when an existing case takes longer than expected?

A number without these details can reassure more than it controls.

Temporary is a question about each person

A two-year intake arrangement does not necessarily mean a two-year stay for every person. Nor does a description such as transit tell us how quickly a final destination will become available.

A passport may be missing. A receiving country may need to verify nationality. A person may raise a protection claim. Travel may depend on documentation, health, airline requirements or another state’s decision. These are possible obstacles that a sound process must anticipate; they are not claims about the private circumstances of the people already transferred to Saint Lucia.

The government should be able to explain the pathways without disclosing personal files.

One person may have a lawful route to remain. Another may wish to return home. Another may require a protection assessment before any onward movement can occur. The system must be capable of responding to the case rather than forcing every case into the timetable used to sell the programme.

A missed target is not permission to disregard a right.

For the same reason, ending future intake is different from ending existing responsibility. A government may be able to stop accepting new cases while still owing duties to people already present. The termination plan must say who supports them, which funds remain available, how decisions continue and what remedies exist if a promised service disappears.

A programme has not finished simply because the announcement has reached its expiry date.

Law does not wait for the annex

Saint Lucia acceded to the Convention against Torture on February 12, 2026. The United Nations depositary notice records its entry into force for Saint Lucia on March 14, 2026.¹

Article 3 prohibits expelling, returning or extraditing a person to another state where there are substantial grounds for believing that the person would be in danger of torture. That rule is about the risk to a person. It does not provide a blanket right to remain indefinitely, and it does not establish the facts of any individual transfer.²

The Committee against Torture’s interpretive guidance calls for an individual assessment and effective independent review capable of suspending removal. It also addresses foreseeable onward-transfer risks and warns against coercive conditions that drive people toward an unsafe return.³

These duties do not disappear because a diplomatic memorandum is described as non-binding. A government still operates under applicable law. An agreement cannot make a forbidden act permissible simply by calling it cooperation.

This is an important limit on the language of national interest. The state has an interest in maintaining good relations, using resources well and protecting public safety. It also has legal duties. The negotiation must find a course within those duties, rather than treating them as another price the larger partner might persuade it to discount.

The practical question follows immediately: who can identify a protection concern, and what happens next?

If the answer is a referral, the referral must reach someone who can act. If the answer is an appeal, the person must be able to understand and use it. If the answer is review before removal, timing must permit the review to matter.

A right printed on a page needs a route into the working day.

The missing procedure is still a problem

UNHCR’s country profile describes Saint Lucia as lacking domestic asylum legislation and a national refugee-status determination procedure, with arrangements involving referral to UNHCR.⁴

That assessment should not be simplified into the claim that no protections apply. Constitutional protections, the Convention against Torture and other applicable rules are separate questions. Nor should we pretend that an agency referral automatically supplies every procedure a person may need.

The operational issue is how the system actually handles a case.

Who receives the first concern? Is there interpretation? Can the person obtain independent advice? Who records the claim? Which authority may make which decision? What suspends a proposed movement while the relevant issue is examined? Who pays for accommodation during that period? How are agencies informed that the case cannot proceed on the original timetable?

These questions can expose a gap between a general promise and a functioning safeguard.

They also show why a small state may need regional or international assistance. Sharing expertise can improve the quality of decisions. But an outside organization’s involvement should not become an excuse for unclear state responsibility. The country needs to know what assistance is actually available, what authority the organization has and what happens if its funding changes.

A phone number is useful. A dependable procedure is better.

A choice must be a real choice

If a return is described as voluntary, we should ask what makes it voluntary.

The International Organization for Migration’s framework for assisted voluntary return and reintegration emphasizes informed decisions and dignified processes. Its operational guidance considers the absence of coercive pressure and access to timely, reliable information.⁵

This does not mean that every lawful immigration removal everywhere requires the individual’s consent. Voluntary return and compulsory removal are different legal and operational categories. It does mean that a programme cannot claim the reassurance of voluntary choice while making the conditions of remaining deliberately unbearable.

A person who chooses between safe, explained alternatives is in a different position from a person who signs a form because shelter or assistance has been withdrawn. A signature alone cannot tell us which situation exists.

Officials therefore need a way to record the information provided, the person’s understanding, access to advice and any relevant pressure. That record should protect the person and the integrity of the process. It should not become a public spectacle.

The surrounding conditions matter. Can the person communicate with someone they trust? Do they know whether refusing a voluntary programme changes their status? Have the risks of the proposed destination been considered? Can they raise a concern without being treated as difficult?

These are questions about the quality of a decision. A state that respects its own people’s agency should understand why they matter to someone arriving from elsewhere.

Public safety without public humiliation

Governments must assess safety. They also must avoid turning uncertainty into collective suspicion.

A screening process should use relevant records, verified information and lawful criteria. It should identify what is known, what remains unclear and which further checks are required. A clean statement from one authority may need examination; an allegation may need correction. Neither blind trust nor automatic suspicion is a professional standard.

The public can be told what the screening process is designed to do and which authority is responsible. It does not follow that the public should receive identifiable medical histories, personal protection claims or untested allegations about each person.

Disclosure has a purpose. Exposing vulnerable people to gossip does not improve accountability for the agreement.

The same principle applies when a problem occurs. The government should report material failures, corrective action and aggregate outcomes where possible. It should protect privacy while allowing appropriate courts, auditors and oversight bodies to examine the records they lawfully need.

We can insist on public accountability without demanding that strangers surrender their private lives as an admission fee.

That distinction will be important in any small community, where a detail that appears anonymous on a national report may identify someone locally. The reporting design should consider that risk before publication.

Money at the time it is needed

Funding has several states. It can be discussed, promised, authorized, transferred, received, available to spend, spent and audited. These are not interchangeable.

A government may expect reimbursement and still need cash today. A partner may have approved support that arrives too late for the supplier’s invoice. A contribution may cover transport but not accommodation, or an initial period but not a delayed case.

A serious financial schedule should specify the categories, timing, currency, responsible payer and process for exceptional costs. It should distinguish fixed readiness costs from costs that vary by person and duration. It should explain what happens if the programme pauses or ends.

This is not a request to place a price on a human being. It is a request to fund the services and duties that governments have agreed to provide.

The distinction matters morally and practically. A per-person payment can create incentives to accept more cases while overlooking whether the system is ready. A budget tied to actual necessary services, with independent review and clear limits, can be designed to reduce that pressure. No payment structure removes the need for scrutiny.

Local providers also need certainty. A guest-house owner should not have to become an involuntary lender to a diplomatic programme. A clinic should not be expected to absorb new duties invisibly. Staff should not be praised for improvising indefinitely around an unfunded promise.

Someone will carry the cost. The agreement should identify who before the bill arrives.

Prepare the whole route

A well-designed process begins before arrival and continues beyond it.

Before acceptance, the responsible authorities need the information required to decide lawfully and assess readiness. Before travel, they need to confirm the practical arrangements and ensure that essential questions have not been left for the airport. On arrival, the person needs understandable information, safe accommodation and access to appropriate assistance.

Afterward, the work becomes less visible but no less important: documentation, case review, health needs, communication, lawful opportunities to support oneself where available, and a clear next decision. A person should not have to spend each week asking a new official to explain the same uncertainty.

The process also needs an owner for the connections between agencies. When everyone performs only their own narrow task, the spaces between tasks can become the places where people disappear administratively.

A coordinator can help, but should not become a substitute for defined authority. Who can resolve a conflict? Who can authorize a necessary expense? Who can halt the next intake when existing capacity is full? Who answers when an agency does not respond?

Test these questions with a difficult hypothetical file before the difficulty arrives in a person’s life. Give the file a missing document, an urgent medical issue, a disputed destination and a funding delay. See where the process stops. Then repair that part.

A government that tests its system is showing respect for the people who will depend on it.

What a good result would look like

Success would be a lawful, adequately funded arrangement that the country can manage and that treats each person with dignity. It would include the ability to decline unsuitable cases, correct errors, explain costs and stop expansion when the conditions are not met.

Some cases might end in a safe return. Others might follow a different lawful route. The outcomes should be reported honestly, without forcing them into a single story of triumph or failure.

The public should be able to distinguish the number proposed, accepted, arrived, currently supported and departed. It should know whether funds promised have become funds available. It should see whether review and complaint systems work. It should know who remains responsible after intake ends.

These measures will never tell the whole human story. A person’s relief, fear or renewed sense of direction cannot be reduced to a completed form. But careful records can help prevent the state from losing sight of people whose circumstances do not fit an easy headline.

The practical and moral questions return to the same place. Can we keep the promise we have made? If we cannot, what must change before we make another?

There is a person inside every case number. There is a public duty inside every signature. A small country should be determined enough to remember both.

Sources and notes