Kenya Invades Somalia Invoking the Right of Self-Defence

Written by

Vidan Hadzi-Vidanovic is a doctoral candidate at the University of Nottingham School of Law.

At a press conference held in Nairobi on 15 October 2011, the Kenyan ministers of defence and interior announced that Kenyan security forces will engage in military operations against the Al-Shabaab militants in Somali territory. They invoked Article 51 of the UN Charter as a legal basis for these actions and pledged that all measures taken in the exercise of the right of self-defence will be reported to the Security Council. Additionally, they invoked the right to hot pursuit, although it is not clear whether it was regarded as a part of the right to self-defence or as a self-standing right. The announcement comes after a number of incidents which involved the incursions of Al-Shabaab elements as deep as 120km inside Kenyan territory, and abductions of several foreign nationals.

At the time of the writing of this post, reports indicate that the border between Kenya and Somalia has been closed and that Kenyan troops are gathering in the border area, preparing to invade Somali territory. It has also been reported that Kenyan troops have already entered 100km deep into Somali territory by Monday evening seizing two Al-Shabaab controlled towns. The Somali representative at the UN suggested that the Transitional Federal Government (TFG) would see the incursion as a violation of Somali sovereignty while expressing some understanding for the Kenyan legitimate concerns.

Some Problems with the Kenyan Invocation of the Right to Self-Defence

According to Article 51, every state has an inherent right to defend itself by employing military means if it has been subjected to an armed attack. The traditional view—which still has a significant support—is that an armed attack must be attributable to a state.

(For views prior to 9/11 see:A. Cassese, “The International Community’s ‘Legal’ Response to Terrorism” 38 ICLQ 589 (1989) at 596; J.L. Kunz, “Individuals and Collective Self-Defense in Article 51 of the Charter of the United Nations” 41 AJIL 872 (1947) at 878; I. Brownlie, “International Law and the Activities of Armed Bands” 7 ICLQ 712 (1958), at 712-713. For more recent support for such approach see: M. Bothe, “Terrorism and the Legality of Pre-Emptive Force” 14 EJIL 227 (2003) 233; E. Myjer, N. White, “The Twin towers Attack: An Unlimited Right to Self-Defence” 7 J. Conf. & Sec. L. 5 (2002), at 7; M. Williamson, Terrorism, Law and International Law, Ashgate, 2009).

Accordingly, the attacks performed by non-state armed groups cannot be qualified as armed attacks, preventing the right to self-defence to become operational in such situations. This position is seen by many to have been confirmed in the ICJ 2004 Wall Advisory Opinion and in the DRC v. Uganda case.

Al-Shabaab is obviously not a state, and their acts cannot be attributed to the Somali Transitional Federal Government by application of the general rules on attribution found in the ILC Draft Articles on State Responsibility. In fact, the TFG is itself engaged in a conflict with this armed group. If one accepts the traditional view, the Kenyan self-defence claim can only be rejected, as Kenya was not subjected to an armed attack by another state.

Ever since the events of 9/11, however, there has been vocal opposition to such a restrictive reading of Article 51 of the UN Charter. Bearing in mind that Article 51 does not textually require that the armed attack be attributable to state, many now claim that a new customary rule evolved, allowing the use of force in self-defence against non-state actors on the territory of another state in cases where the violent acts of these groups could be qualified as an armed attack (see: Dinstein, War, Aggression and Self-Defence, CUP 2005, at 204-8; Lubell, Extraterritorial Use of Force against Non-State Actors, OUP, 2010, at 31 et seq. The same opinion is held by Greenwood, Paust, Franck and many others). The above mentioned rulings of the ICJ do not necessarily contradict these positions despite the fact that they have been delivered after 2001 and that they appear to condition the existence of an armed attack with the imputability of these attacks to a state. In the Wall Opinion, Israel’s claim of self-defence was inappropriate simply because the attacks did not come from the territory of another state but from a territory which is under its occupation. In the DRC case, on the other hand, the Court refused to deliberate on the contentions that the right to self-defence can be exercised in response to the “large-scale attacks by irregular forces” (para. 147) providing, therefore, neither an affirmative nor negative answer to the question (for a discussion see: N. Lubell, Extraterritorial Use of force Against Non-State Actors, Oxford, 2010, at 31-34).

The ruling of the Court, however, could be understood to imply that “large-scale attacks of irregular forces” as opposed to any attack, could theoretically be regarded as “armed attacks” within the meaning of Article 51 of the UN Charter. If this reading is to be adopted, it would seem that it sets an especially high gravity threshold for irregular attacks to become armed attacks, even higher than the threshold required to transform any illegal use of force into an armed attack as set in the Nicaragua Judgment (para. 210). That said, it would hardly be new for international law to set a higher gravity threshold for non-state than for state violence in order to deem it as appropriate for regulation, as e.g. in the jus in bello framework related to the recognition of the existence of non-international armed conflicts. Alternatively, a gravity threshold could be just a component of the customary requirement of necessity as a precondition for the exercise of self-defence (See K. Trapp, “Back to Basics: Necessity, Proportionality, and the Right to Self-Defence against Non-State Terrorist Actors” 56 ICLQ 441 (2007), at 145 et seq).

It is doubtful whether the activities of Al-Shabaab against Kenyan territory have reached such a high threshold in order to be qualified as an armed attack, or alternatively if the attack of this magnitude necessitated self-defence. They are certainly nowhere near the scale of the 9/11 Al-Qaeda attacks or the Hezbollah attacks that triggered the 2006 Lebanon War. It is clear that, observed in isolation, none of the Al-Shabaab incursions could individually be understood as armed attacks.

In its argument, however, the Kenyan government invoked nine separate incidents from 2009 to 2011. It would seem that by invoking these incidents the Kenyan Government is relying on the doctrine, most vocally supported by Professor Dinstein (War, Aggression and Self-Defence, at 231), according to which a number of successive pin-prick attacks of lower intensity that show a distinctive pattern can constitute an armed attack when taken as a whole. But even if taken together, and if one could identify a distinctive pattern (bearing in mind that Al-Shabaab denied responsibility for the most recent abductions), it is doubtful if these incidents reached the required gravity threshold. There was no significant (if at all) destruction of property and no significant loss of life. The consequences that the Kenyan Government emphasizes are the infringement of security in the border areas and the severe economic losses due to the influence of the attacks on the Kenyan tourism industry on which the country heavily relies. There is, however, little support in international law that these would be sufficient, even in the cases of the attacks performed by a state, to reach the threshold of an armed attack.

Hot Pursuit

In explaining its intentions to send its military force into Somali territory, the Government of Kenya also invoked the right to hot pursuit. The right to hot pursuit on land is highly controversial and is generally rejected. In its original form, it is an institute of the law of the sea and it is regulated by Article 111 of the UN Law of the Sea Convention. The rule entails that hot pursuit must be uninterrupted and must commence while the offending ship is within the internal waters of the offended state. Most importantly, it does not allow for the incursion onto the territory of another state under any conditions.

Kenya had the option to invoke this right in the abduction cases of the French and British tourists in Lamu. In these cases, the offenders took the tourists from the beach on speed boats and headed towards international waters and further to Somalia. If the Kenyan navy started its chase while the offenders were still in its territorial waters and if it caught them while on the high seas, the claim of hot pursuit would be legally valid – but it is too late for that now. In any event, the announced operations are to be performed by land and will include incursion onto Somali territory. Therefore, hot pursuit seems to be inapplicable.


This most recent example of state practice on the use of force will only contribute to the uncertainties surrounding the legal regulation of the right to self-defense. On one side, it is one more example of a state’s understanding of Article 51 according to which self-defense can be exercised against non-state entities on the territory of another state without the consent of the territorial state. But the Kenyan intervention in Somalia might present a considerable challenge even for those of us who support such a reading of the Charter.

As already said, it is at best doubtful if the incursions of the Al-Shabaab on Kenyan territory amounted to an armed attack and if they necessitated the exercise of the right to self-defense. If it is accepted that they indeed reached the required threshold of gravity, one might have to concede that the threshold significantly dropped and that the requirement of necessity can be applied more loosely, which opens the door for even more extensive resort to unilateral force. On the other hand, this particular case might serve as an example that the gravity of cross-border non-state violence is only one among many factors that need to be addressed when assessing the necessity of the use of force in self-defense. In this particular case the inability of the TFG to prevent Al-Shabaab from crossing into the Kenyan territory certainly played a big role in Kenyan decision to invade Somalia.

Kenya had the option of informing the TFG and getting its consent for invasion and it still has the opportunity to do so. In all likelihood, Kenya would have gotten it. Showing a strong arm in international relations might be a good move for the government with an eye on the upcoming elections in Kenya, but it also opens a number of difficult questions for international law and unnecessarily brings more uncertainty into an already unclear legal framework.

Print Friendly, PDF & Email


No tags available

Leave a Comment

Comments for this post are closed


Dawood I Ahmed says

October 18, 2011

Of course, there is an emerging consensus in state practice and scholarly opinion, especially American, (though not at the ICJ unless one considers the separate opinions of Judge Kooijmans in Armed Activities of Congo) that self-defence actions can be taken in the "unable or unwilling" state with or without the consent of the state provided that there has been an "armed attack" by non-state actors from within that state. As you mention though the real question is what the threshold of armed attack is and it does seem that this bar is getting lowered with the passage of time in response to terrorism threats emanating from weak and failed state.

Mihai Martoiu Ticu says

October 19, 2011

If states have a right to self-defence against non-state actors, the individuals should be granted the same right. States attack us, the individuals, on regular bases. And they attack us with drones, thus there is no way we can defend ourselves by just attacking combatants wearing distinctive signs and their weapons openly. Therefore we should get the right to blow up their civilians, what they now negatively call terrorist attacks.

Ok, we the individuals of the world are reasonable and give the states a fair alternative: the states should cooperate in creating international courts where we can sue states and force them to do or refrain from certain things. See the ECtHR as a model and the Al-Skeini case as an example of what I have in mind. In short, international law should be taken from the hands of states and should contain the following rule: individuals should have the possibility to force states to certain actions through binding decisions of international courts, or they should be free to carry terrorist attacks against states.

Vidan Hadzi-Vidanovic says

October 19, 2011

First of all, thank you for your comments. My reply might be a bit longer, but bear with me. It might spark some additional discussion.

Dawood - My opinion is that gravity of attack is only one among many elements that need to be taken in consideration when assessing if the attack was indeed an armed attack and if the requirement of necessity has been met for the right to self-defense to become operational. This stand both for the armed attacks committed by states and non-state entities. Consider this scenario:

Imagine that a huge dam retaining a massive volume of water is located in the border area on the territory of State A. A nearby town is located on the territory of State B. Now imagine that the dam cracks, and this huge mass of water pours out rushing towards the town with 50.000 inhabitants on the territory of State B. Let us say that it is certain that if this water reaches the town, it will be swiped out from the face of the earth killing all of its inhabitants. But between the dam and the town, there is a tick forest and a huge highway bridge on the territory of State A. If these are destroyed a new dam could be made out of ruins and timber and the town on the territory of State B could be saved (I am not a civil engineer, but let us assume that this would be possible). So, State B sends a squadron of its war planes, violates the airspace of State A, destroys the forest and the bridge with the massive missile strike and stops the water, and returns its planes to the base. The entire forest is destroyed along with the bridge. To make things even worse, some 50 civilians were on the bridge at the time of destruction and two trucks of the members of armed forces of the State A rushing to the spot of accident in an effort to assess the damage. Obviously, they are all killed. In return, a town of 50.000 people has been saved.

Now a) is this the use of force contrary to Article 2(4) of the UN Charter; b) if it is, is it an armed attack and c) even if it is, does it necessitate the use of force in self-defense?

As an answer to the first question, I will follow what Judge Higgins said in her book Problems and Process and assume that a single missile dropping on the territory of another state violates that state's territorial integrity. In this case, State B also violated sovereignty of State A by destroying its property and depriving it from the right to make its own decisions with regard to the issue occurring under its own jurisdiction.

Now, is this instance of the use of force an armed attack or not is more complicated question. If it was only gravity that needs to be taken into consideration, one might rush to conclude that the destruction of the entire forest as a natural resource, a demolition of a highway bridge which is of the obvious infrastructural importance and the killing of civilians and soldiers is certainly an armed attack. Nobody sane can claim that it was, to use the ICJ's construction from Nicaragua, a mere frontier incident. Here, however, we might be tempted to borrow the approach taken in the Definition of Aggression, something what is routinely done by both ICJ and scholars despite the fact that armed attack and aggression are not the same (the difference is well explained by Alexandrov in his book on self-defense), and observe that "other circumstances", beside the gravity, are also relevant to assess if a particular incident of the use of force is also an aggression (or armed attack). What these other circumstances might be, we don't know because nobody tells us (but bear in mind that they are not violation of territorial integrity and sovereignty for the reasons stated above).

Finally, if this particular incident could be understood as an armed attack from Article 51, does it really necessitates self-defense? There are absolutely no indications that State B will send again its planes and bomb the territory of State A. It already achieved what it wanted to achieve, namely, it prevented drowning of 50.000 of its citizens. Let us say that it promptly sent a formal apology to the State A as well as stating that it will offer reparations for all damages (it accepts, therefore, that it cannot invoke necessity as exculpatory factor within the meaning of the rules of state responsibility since no forcible measures can be taken in necessity as the ILC concluded). The requirement of necessity as a part of self-defense in this particular case would simply not be fulfilled, and State A could not respond by force against State B.

The key is in necessity and, in my opinion, gravity is just one among many factors that reasonable state officials need to take into consideration when assessing the appropriateness of taking self-defensive measures. Other factors are the ongoing threat, inability or unwillingness of territorial state to deal with the problem, maybe even inferred or presumed motives of the attacking state and a number of other factors that only practice can reveal and that cannot be predetermined through the abstract legislation.

Mihai - You are basically raising the issue of legitimacy of violence from the perspective of general theory of law. Your proposition would be acceptable if the modern law and modern international law were not resting on the Weberian premise that states are sole legitimate monopoly of force. Why the states acquired such a status is different story, and there are some great classical works on the subject (see, for example, H. Spruyt, The Sovereign State and Its Competitors: An Analysis of Systems Change, (Princeton: Princeton University Press, 1994) from the perspective of international relations; C. Tilly, Coercion, Capital and European States, AD 990-1992 (Oxford: Blachwell, 1992), from the perspective of political economy). A major historical argument for the emergence and later domination of the “sovereign territorial state” and its prevailing over other forms of human associations is its effectiveness in preventing the “defection by its members, reducing internal transaction costs and making credible commitments to other units” (H. Spruyt, “Institutional Selection in International Relations: State Anarchy as Order” 48 International Organization 527-557 (1994), at 527). How did the states acquire such monopoly of force by which they achieved their status of a credible “racketeers” and “guarantors” is separate issue, and one can turn to social contract theories, Marxist interpretations etc. But for now, it is agreed that a state protects us and guarantees for us unless if it delegated its powers to some supra-national entity (like ECtHR that you mention or the UN Security Council in some situations) or to sub-national one (private security companies, for example).

Of course that this is not the only solution – states are relatively new concept themselves and they will probably disappear as a concept just as they appeared once they are not able to fulfill its tasks. I am afraid (or maybe I am relieved) that we are not going to be around to witness that change. It is true that state-system is challenged by globalization and the emergence of powerful non-state actors, but I don’t think it is anywhere near the point of breaking. Institutional change is an infrequent phenomenon, it “imposes costs, and hence social groups and political actors will be unwilling to experiment with new institutions unless a serious exogenous shock alters internal political alignments” (Spruyt, at 7). Unless we face a dramatic change which cannot be handled within the established system, it is an idealistic presumption that a mere potential of the new system to deal with the problem better than the old one will lead to some fundamental changes. The state as a central institution of international system as we know it today showed a remarkable degree of flexibility which could have only contributed to its durability as a social concept. And while its significance is maybe fading in some particular spheres of the system, such as international trade for example, it does not mean that it is disposable in the others, such as security.

That said, Al-Qaeda, Al-Shabaab and other organizations legally or simply pejoratively branded as “terrorists” or more precisely “Islamic fundamentalists” are precisely one of the emerging competitors of the national state as a concept. The main idea of this “Islamic fundamentalism” that has been promoted by Muslim Brotherhood for century now is the establishment of the ummah, a community of Muslims governed by the laws of Prophet that knows no borders and state institutions as we understand them (some references to the place of concept of ummah in the Al-Qaeda ideology and the ideology of other Islamic organizations can be found in L. Wright, The Looming Tower, (Knopf, 2006), at 164 and 349). Another competitor, to which I think you implicitly refer, is the more powerful international community, a Federal Republic of Earth or similar in which you still envision states as a player, although less powerful one. Whether one of these will really prevail (but bear in mind that one negates the other), only time will tell. Maybe another mode of social organization, the one we cannot even grasp with our current knowledge, will emerge.

Until that happens, states are here to stay for the time being, and if changes are to be made, they would rather include centralization of force above states rather than bellow them, towards the individuals. Looking at the examples where no state organization exists, like in Somalia, it seems to me that the Hobbesian natural state is more realistic than Lockian, and I, for my part, would not like to live in such a world. Although, Virgin Galactic now has competition so the space travel might be even cheaper, and if the CERN scientists are right and we can travel faster than light, I might have additional options.

Mihai Martoiu Ticu says

October 19, 2011


I don't rise "the issue of legitimacy of violence", I rise a question of common sense. I know one thing. If I perceive an individual or a group as a threat to my security or a violator of my rights, there are only two possibilities: I can either force the individual/group through a court to do/refrain from something, or I'm free to blow them up. A state is nothing than a group of people.

And I'm not the only one stating that individuals should be able to sue states. See for instance:
Trindade Cançado, A. A. (2011). The access of individuals to international justice. Oxford University Press
Lauterpacht, H. (1968). International law and human rights. Archon Books

Vidan Hadzi-Vidanovic says

October 19, 2011

Mihai - Thanks for clarification and sorry if it appeared that I neglected your argument about the desirability of greater access to international justice for individuals. I completely agree with you with regard to that and that is why I didn't respond. I was only referring to the second part of your either/or argument in which it seemed to me that you say that the absence of the access to justice legitimizes non-state violence against states (and their civilians etc, but I would not go as far as that since I don't think that state as organization is simply group of people).

Dawood I Ahmed says

October 19, 2011

Vidan - Thanks for the reply. Its an interesting hypothetical, not least because you have avoided the conventional human agency example i.e. unpredictable terrorists operating within State A; though it does raise more questions than answers e.g. how likely was it that the dam would break? Was State A informed of this occurrence, did it exercise due diligence in this responsibility to prevent harm? was it unable to prevent the harm? Why did State B not build a strong wall on the border etc? For what its worth, here's my two cents:

Assuming that State A was yet unable to prevent the harm from occurring and all other avenues had been exhausted by B (unlikely), one would still query what harm was being acted again - no dam had broken so one would be engaging in semantics to say that State B's action was in response to an armed attack - at best it would have to be viewed within the less than fully accepted pre-emptive/anticipatory framework. Also, it is doubtful whether there was any necessity - unlike human action which can be unpredictable and come as a surprise, preventing the dam would probably require monetary resources and one could scientifically predict the foreseeability of harm and the costs of controlling the "harmful element" - if State B derives great benefit it could foot the cost in a Coaseian sense. If State A did nothing about this despite knowledge and the provision of financial resources and it was almost certain that the dam would collapse and the deaths would occur then perhaps one could still better place this within the state responsibility paradigm anyway as a necessity counter-measure (some commentators accept this e.g. Malanczuk, 'Countermeasures and Self-Defence as Circumstances Precluding Wrongfulness in the International Law Commission's Draft Articles on State Responsibility', in M. Spinedi and B. Simma (eds.), United Nations Codification of State Responsibility (1987)) - of course the rules of proportionality would still apply.

On the other hand, it would appear that State B has engaged in an armed attack but as you point out the harm is already done so unless future attacks are being planned, its use of force may count as a reprisal and be outside the Article 51 exception.

Keeping aside doctrinal issues though, one could query in terms of legitimacy which of the two hypothetical uses of force would invite greater international acquiescence or condemnation. I would be tempted to say that State B's actions here, even if State A was unable, may not be viewed favorably especially if there are other things that state could always do before force becomes necessary e.g. establish a wall on its border or some other engineering marvel which could prevent the 50,000 deaths.

jpaust says

October 19, 2011

That is NOT the traditional view. The traditional view is that non-state actor armed attacks can trigger the right of self-defense -- ever since and before the 1837 Caroline incident!
Please see
Self-Defense Targetings of Non-State Actors and Permissibility of U.S. Use of Drones in Pakistan.
There is no need for special consent from the foreign state from which non-state actor armed attacks emanate. There is no need for attribution or imputation of the non-state actor attacks to the foreign state. There is no need for the existence of a war. There is no need to prove that the foreign state is unwilling or unable to control its territory.
please see also for a free download.

Vidan Hadzi-Vidanovic says

October 20, 2011

UPDATE - Following the meeting between Kenyan and TFG officials in Mogadishu, it would seem that Kenya acquired Somali consent for invasion. See: . The news came several hours after writing this post.

If this is correct, this particular instance of the use of force cannot be perceived anymore as an act of self-defense but as an intervention with the consent of the territorial government. This, however, raises the issue of the retroactive effect of consent. The attention should also turn to the issue of classification of armed conflict. Dapo Akande recently posted his draft chapter on the issue on this blog

Alexander Eichener says

October 20, 2011

I do not see any territorial government there, east of Daadab. You keep imagining a status that does not apply. This is not the Turkish-Iraqi border. It is more comparable with the Israeli invasion into Southern Lebanon during the dissolution of the former Lebanese state, and the subsequent creation of an occupied buffer zone, in cooperation with local pro-Israeli militia. Which is exactly what Kenya aims at.

Vidan Hadzi-Vidanovic says

October 20, 2011

Jordan - I largely agree with your position on the issue in general. I, however, do think that factors such as inability or unwillingness of the territorial state to address cross-border armed activities of non-state groups do play a role, not as juridical concepts (for the purpose of attribution or the establishment of state responsibility through the due diligence principle because I think as well that this is not needed) but as factual circumstances that need to be taken into consideration when assessing the existence of the customary requirement of necessity which conditions the legality of self-defense. But as I said in the post, my opinion is that these are only some of many factors that need to be taken into consideration when assessing necessity and sometimes they might not even be applicable.

Alexander - To be honest, I am not sure that even Kenya knows what exactly they want to achieve with this operation, but that is not a legal question. It is however interesting what you said about the "imaginary status".

Somali government definitely does not exercise its sovereignty over this (if any) part of its territory and if only facts were relevant for determination of the existence of a state, I would completely agree with you. If this was the case, we couldn't even talk about the use of force within the meaning of Article 2(4) of the UN Charter to which self-defense is exception, because there would be no territorial integrity and political independence of Somalia in the area of Kenyan operations to be threatened by the use force in the first place.

But, as most of the writers agree, the existence of state depends both on the empirical and juridical elements. In other words, statehood is not just a matter of facts. Principle of effectiveness is very loosely applied once a state has been established and recognized. This is due to the fact that, while in accordance to Montevideo principles, territory, population and internal and external sovereignty are necessary elements for the existence of states, they are also most fundamental rights of the states. Infringement of these rights cannot automatically lead to the cessation of a state. Accordingly, Somalia is state, however failed, and as such it is its right to claim sovereignty over the entirety of its nominal territory as recognized by international law. I agree that this is imaginary status but application of international law as of any law can lead to unrealistic solutions. I might be heretic, but I do not see too many obstacles for a state to simply break the law in these situations. Nobody can say that it doesn't happen.

Alexander Eichener says

October 20, 2011

"...nisi magna latrocinia?"
'Nuff said. :-)

jpaust says

October 20, 2011

Vidan: yes, as another aspect of circumstance with respect to overall consideration of reasonable necessity regarding methods and means (necessity for a response has already occurred once the victim state, its embassies abroad, its miltary personell and/o other nationals abroad are attacked or under a relatively continuous process of armed attack) -- but not as a determinative circumstance regarding permissibility or impermissibility of responsive force.
my Denv. J. piece is now up on ssrn (cite is in comment above), it was substituted for the older draft.
Comments welcomed.

jpaust says

October 20, 2011

Oh, but Article 2(4) does not merely prohibit uses of force "against" territorial integrity and political independence. It also prohibits the use of force in any other manner inconsistent with the purposes of the Charter (such as peace, security, self-determination of peoples, human rights, that was shall not be used save in the common interest). This third category of prohibition is not tied to the existence of some state (e.g., self-determination of peoples, 1970 Dec. on Principles of Int'l Law re: no use of force to deprive such peoples, etc.). Further, those who intervene might argue that, on balance, most of the purposes of the Charter are served with regard to some forms of "humanitarian" or "R2P" intervention. Makes it all the more interesting.