Revisiting the UK’s 2011 Rescue of Nationals from Libya. Part I

In 2011, the UK, Germany and (seemingly) the Netherlands launched targeted rescue operations to rescue their nationals (before the passing of UNSC Resolution 1973, which authorised “all necessary measures [short of occupation]… to protect civilians and civilian populated areas…”) from Libya, in the midst of worsening civil strife amidst the collapse of the Gaddafi regime. While Germany later justified its operation on the basis of consent, the UK and the Netherlands failed to publicise any legal basis for their operations. Notably, the UK has remained ambiguous about whether its rescue was conducted with Libyan consent. Nevertheless, authors Grimal and Melling highlighted contemporaneously to the events that the intervention was conducted “presumably without [Libyan consent]”. This view was supported by other contemporary evidence, including interviews with pilots.

They appear to have been correct. In response to an individual enquiry I made to the UK’s Foreign Commonwealth and Development Office (FCDO), the UK appears to have confirmed, seemingly for the first time, the nonconsensual nature of its rescue operation in Libya. Here is part of their response, provided in January 2026 (emphasis is my own).

The safety of British nationals abroad is always the highest priority for the Foreign, Commonwealth & Development Office.

During the rapidly deteriorating situation in Libya at this time, the UK undertook evacuation flights supported by specialist personnel to ensure British nationals could be moved to safety as quickly as possible, as widely reported at the time.

Regarding your questions on consent from Libyan authorities, in situations of acute instability, the UK’s overriding responsibility is to protect its citizens, and actions taken were focused on ensuring their immediate safety.

Remarkably, the UK made no attempt to justify its operation as a lawful exercise of self-defence. Instead, it appears to rely on (a) the existence of a situation of acute instability, and (b) the need to ensure British nationals’ “immediate safety”. This departs from the prevailing state practice of relying on self-defence as the legal basis for the conduct of rescue operations. However, this piece argues that the UK’s response is consistent with state practice in respect of a novel sub-category of rescues which reasonably do not anticipate violent conduct, and which only employ violence in ancillary, “on-the-spot” reactions. These “non-violent evacuation operations” (NVEOs), which have not been justified by states as lawful exercises of self-defence, are best justified by the existence of a state of necessity precluding wrongfulness.

Rescue Operations

States have occasionally launched operations, without the consent of the targeted state, to rescue their nationals abroad from imminent danger. The archetypal example of such a rescue operation is Operation Entebbe, in which Israeli personnel were deployed to Uganda to rescue civilians held hostage by PLO hijackers. Such operations have been conducted by a range of actors, including by the US in Cambodia (the Mayaguez Incident) and by Belgium in the DRC (Operations Dragon Rouge and Dragon Noir).

The legality of these operations is contested. While Operation Entebbe was near-universally condemned by states, some other rescues, such as the Mayaguez Incident, appear to have been largely tolerated by the international community. Academics also disagree on the legality of rescue operations. Forteau and Ruys have argued it is not possible to assert a customary right to conduct such rescue operations (although Forteau concedes that establishing a prohibition of such rescues would be difficult), while Peters has written that such operations are generally permissible. Existing legal assessment appears to have generally relied on the assumption, regularly found in the literature, that rescue operations constitute uses of force, meaning that the jus ad bellum applies.

Non-Violent Evacuation Operations

While most operations conform to this dogma, it is not universally true. The jus ad bellum may be irrelevant to a minority of operations which do not constitute uses of force. These operations share many characteristics: they are nonconsensual, have (thus far) universally been conducted during breakdowns of authority, reasonably do not anticipate violence, and do not employ violence beyond in ancillary, “on-the-spot” reactions. Because of their humanitarian nature, accompanied with the lack of expectation of and employment of violence, such operations do not appear to constitute uses of force under Article 2(4) of the UN Charter.

These operations, termed “NVEOs” above, appear to exist as a distinct legal category of rescues, although they have received little attention from international lawyers (one exception is the work of Gill and Tibori-Szabó). The distinctiveness of NVEOs (vis-à-vis other rescue operations) is reflected by state practice. Unlike with other rescue operations, states conducting NVEOs have refrained from justifying them under the heading of self-defence. The clearest articulation of the legal basis for an NVEO originated from Germany, which justified its conduct of Operation Libelle in Albania by reference to the existence of a customary right to protect nationals (as opposed to self-defence and necessity). Furthermore, while most hostage rescues have been controversial, encountering at least some adverse state reaction, NVEOs have been near-universally acquiesced to (and often accepted) by states, seemingly suggesting that states view them as distinct from “conventional” hostage rescues.

Gill and Tibori-Szabó appear to have most closely engaged with NVEOs as a distinct legal category of rescues. They suggest that a rescue operation might not constitute a use of force (and thus be describable as an NVEO, justifiable by the existence of a state of necessity) if (a) consent from the host state is infeasible, (b) “as little force as the situation strictly requires” is used, and (c) the force used does not amount to a use of force against the state. The difficulty with their approach is that the requirement that “consent from the host state is infeasible” wrongly implies that the nature of the failure of the targeted state to consent might be relevant to whether a use of force has occurred. It is also unclear what the requirement that “as little force” as possible is used contributes – even if we were dealing with a use of force, unnecessary force could not be used. The better view is that state consent is binary – either it has been provided or it has not – and is not strictly relevant to whether a use of force has occurred. The question as to whether there is a use of force must be answered with respect to more basic principles in the jus ad bellum.

Corten has briefly discussed the potential of NVEOs existing as a novel category of rescues. He appears to accept that such operations might exceptionally fall outside Article 2(4)’s scope if they do not “[involve] fighting or [cause] any damage to a state”. In my view, Corten is wrong to suggest that any violent conduct would disqualify a rescue operation from falling outside the scope of Article 2(4). Corten relies in part on the American NVEO in Liberia as an example in support of his position, but evidence suggests that the operation encountered some ancillary violence, and it is unclear whether the US responded in immediate self-defence. To disqualify an operation from constituting an NVEO on the basis of whether it responded to this ancillary, unexpected violence (given that Corten accepts that the Liberia operation may have constituted an NVEO), would be incoherent.

The better view is that ancillary violence employed exclusively as an “on-the-spot” reaction, deriving from Dinstein’s conception (where a “small-scale” attack encounters immediate opposition which “closes the incident”), does not more broadly implicate the NVEO as a use of force under Article 2(4). This view reflects state practice. States have not argued that such “on-the-spot” reactions ancillary to the broader operation have rendered rescues as potentially breaching Article 2(4). This is seen, inter alia, in the relatively muted state responses to the initial “rescue” component of the 1965 US intervention in the Dominican Republic and the above-mentioned NVEO in Liberia. As long as any violent conduct employed is exclusively part of an “on-the-spot” reaction, the NVEO need not constitute a use of force.

Corten nonetheless acknowledges that his criteria of a lack of fighting or damage alone may not be determinative that an extraterritorial operation is not a use of force. This must be correct. An important indicator of whether the rescue operation constitutes a use of force is whether the conducting state anticipates that it will employ violent conduct as part of the operation. Any operation which does not anticipate employing violent conduct will likely not constitute a use of force because there is a lack of the requisite hostile intent.

Any operation which does anticipate the employment of violent conduct would inherently possess sufficient hostile intent that, particularly when combined with the violation of the targeted state’s sovereignty, would likely qualify as a use of force, and thus require justification under the jus ad bellum regardless of whether violent conduct actually occurred. The legal assessment cannot be modified just because the state was sufficiently fortunate to encounter no resistance. The converse view would introduce significant uncertainty and inequality into the jus ad bellum. Equally, even if a rescue operation reasonably does not anticipate the employment of violent conduct, the employment of force beyond in ancillary, “on-the-spot” reactions will clearly be sufficient for the operation overall to constitute a use of force, even without any hostile intent. Any other suggestion would wrongly undermine the Article 2(4) prohibition on force.

Therefore, I suggest that any operation that (a) reasonably does not anticipate the employment of violent conduct, and (b) does not employ force beyond in ancillary, “on-the-spot” reactions, will not constitute a use of force. The next part of this post will assess the appropriate legal basis for these “NVEOs” with reference to the UK’s 2011 rescue from Libya.