This is the first installment of the Law of ATLA Conflict series, where I attempt to apply international humanitarian law to Avatar: the Last Airbender. See the series here. If you’re new to IHL, scroll down to see a list of concepts discussed in this post and brief explanations thereof. If you’re interested in learning more about IHL, I would recommend browsing the International Committee of the Red Cross’ website.
I think this is the post I’m most nervous to write, in that if I get this one wrong, I’m not sure that I’ll have the heart to continue. But I’ve already made a big deal about doing this, so I guess I better follow through.
As I mentioned before, I’m not watching every single episode of ATLA, just those that I think would provide the most IHL-relevant content. To that end, the episodes I watched from Book One are chapters two, three, six, eight, ten, fifteen, nineteen and twenty (the last two being the two-part season finale). I intended to watch Chapter Nine but other than to say that piracy is a jus cogens norm in international law (i.e., under no circumstances is it permissible), there’s not a whole lot that’s relevant.
One of the big issues that comes up with ATLA is that the main characters are all kids; Toph and Aang aren’t even teenagers. So, obviously, the main thing that comes up is the issue of child soldiering, particularly with Toph, Aang and Katara (the other kids, if ATLA lore is to be believed, are fifteen and up, which means they don’t count as child soldiers under Article 77 of Additional Protocol I of the Geneva Conventions). But – and here, I beg some suspension of disbelief – to begin with the premise that half of Team Avatar are child soldiers would kind of throw a wrench in everything. The issue of child soldiering is hugely complicated, and I don’t know that I’ll necessarily do a good job of navigating its nuances in a fantasy show intended for kids. Plus, I feel like it’d put a damper on…this whole project. So we’re going to set aside the issue of child soldiering and consider that Team Avatar are all combatants, without much reference to their ages. (If someone else reading this has insight to offer on what an ATLA analysis within the framework of Team-Avatar-as-child-soldiers would look like, I’d love to hear!)
We’re also going to assume that offensive bending has the same status as using arms in wartime. The reason this is important as least in the world of ATLA is that it would be hugely problematic to assume that all bending is the same as the wielding of arms, and thus that all benders are either present-combatants or future-combatants. This will also be important for establishing the boundaries of the principle of distinction, i.e., whether a person who has been targeted is a combatant or non-combatant.
I also should have clarified in the previous post – obviously, we’re assuming that all of the requisite conflict parties would be party to IHL in this situation.
There’s plenty more nuances to consider so instead of laying them all out here, let’s just get started!
Chapter Two: The Avatar Returns
I decided to start with chapter two because this is where most of the action actually starts. In fact, I might have even skipped chapter two, but there are a few interesting things that happen here. First of all, I assume from the get-go that Sokka is a combatant; he openly wields weapons, and makes his position as defender of the Southern Water Tribe very evident. For some reason, I had recalled that he was able to muster the kids and women of the Southern Water Tribe to help him defend against Zuko, and so one of the notes I had made for myself was levée en masse1, but that didn’t actually come to pass. Suffice it to say that had he done so, it would have been legal, so long as the tribespeople openly carried arms (and complied with IHL).
It’s also worth noting that at no point does Zuko actually violate any principles of IHL during his time searching for Aang. Sure, he intimates the people a little, but it’s clear he doesn’t actually want to hurt anyone (other than Sokka, who did attack first, admittedly – and got a few bonks on the head for it!). Nor does he actually destroy anything other than (somewhat inadvertently, if I remember correctly) Sokka’s makeshift garrison, which is fine. Eventually, Aang turns himself in – and since I’m assuming that the Avatar is a combatant during wartime, his detention by Zuko is governed by Geneva Convention III2 and while we don’t see a whole lot of his actual detention, things seem okay? It doesn’t necessarily seem like they were ever going to hold Aang in horrendous conditions. The most pertinent issue of note is the treatment of PoWs who try to/succeed at escape. Per Article 93 of GC3, an escaping PoW cannot be sanctioned for the escape attempt itself, but only for any offenses committed during the escape attempt (e.g., destroying property, or injuring/killing another person). This would have been relevant if they had managed to prevent Aang from escaping, but they didn’t. This is despite Zuko actively trying to prevent escape – and unfortunately, we don’t get a whole lot of guidance as to the means through which the detaining power is permitted to prevent the escape of a PoW. It’s pretty clear that Zuko isn’t above attacking Aang to prevent his escape (which could include injuring him). Intuitively, that doesn’t feel right; but I don’t have the points of reference available to me through GCIII to analyze the extent to which that’s wrong under IHL.
This episode also provides us the glimpse of Aang’s Avatar state; I maintain that when Aang enters the Avatar state, he is not just a combatant, but a weapon in his own right, and that bears analysis on both fronts. This is consistent with my understanding of offensive bending as analogous to a weapon, and also accounts for the massive degree of power wielded during the Avatar state which – if not controlled – could threaten the principles of distinction and proportionality, as well as have consequences for non-combatants, civilian objects, and even natural life. And because the Avatar state encompasses the entire being of the Avatar, the Avatar themself must be considered a weapon.
Well, that’s my justification, anyway. And frankly, I don’t think it’s a bad idea to hold the Avatar – while in the Avatar state – to a greater level of scrutiny particularly given that Aang’s explanation of getting into the Avatar state was “I just sort of did it.” Per my notes: “not great.” The specific instance of the Avatar state being in triggered in this episode was far more measured, however; it was largely controlled, only targeted lawful combatants, and – well – Aang was drowning when it happened.
There is a moment that I was a little curious about; Aang, to prevent the Fire Nation ship from following them, redirects some fire bent at him towards an ice shelf, which breaks and submerges the Fire Nation ship. Ultimately, I don’t think it was disproportionate or nondiscriminatory in this specific instance. It did exactly what it was supposed to and nothing more. So I wouldn’t say that it was akin to dams, dykes or nuclear electrical generating plants in terms of not being able to control the forces unleashed by breaking the ice shelf (see Article 56(1), AP/I – “if such attack may cause the release of dangerous forces and consequent severe losses among the civilian population”). But given how tight that criteria is (literally just limited to those three situations – which I continue to think is ridiculous!), I guess it doesn’t matter anyway, especially since there aren’t any civilians nearby.
The last thing I’ll note is that this episode does catapult Katara out of civilian status and into (unprivileged) combatant status, which is a status that (I argue) she maintains throughout the series. With, um, consequences later on that might make Katara culpable under IHL but…we’ll get there when we get there. (Spoiler – highlight the following text: Blood-bending is definitely not Geneva-friendly, especially when used as a means of reprisal).
Chapter Three: The Southern Air Temple
There is actually so much to discuss here. This is generally a highly emotional episode, one that sets out the stakes for the entire series – and for our purposes, the terrain of the war. One thing that becomes abundantly clear is that the Fire Nation specifically intended to wipe out the Air Nomads in order to prevent the rebirth of the Avatar. Of course, they were too late; Aang wasn’t at the air temples. But that doesn’t nullify the genocide that took place: it fulfills both the actual crime element (in this case, extermination of a people) and the mental element (that of intent). As a grim aside, genocide is a difficult crime to prove. What makes it tricky is the mental element; how do you prove that someone/a belligerent side intended to kill an entire people? But it is not the only crime that deals with widespread murder: ethnic cleansing, mass extermination, massacres…these are all atrocious, unfathomably evil actions. We absolutely validly ascribe a particular enormity to genocide – with all the imagery that it conjures up – but how do you create a taxonomy of an mass murderous evil? It’s a chilling idea. That doesn’t mean it doesn’t happen, and that it isn’t the precise business of international criminal law to taxonomize evil.
Depressing digressions aside, we get more and more insight into this and other potential genocides perpetrated by the Fire Nation as the series progresses. Fire Lord Sozin’s mens rea is established in Book Three:
With Roku gone, and the great comet returning, the timing was perfect to change the world. I knew the next Avatar would be born an Air Nomad, so I wiped out the air temples. But somehow, the new Avatar eluded me. I wasted the remainder of my life searching in vain. I know he’s hiding out there somewhere. The Fire Nation’s greatest threat … the last airbender.”3Sozin, Book Three, Chapter Six
Obviously, the accusation of genocide is such that any further discussion of the circumstances surrounding – in particular – the Fire Nation’s attack on the Southern Air Temple might seem like minutiae, but, well, let’s get into it anyway. I would consider all of the air temples as protected civilian and cultural objects, particularly given the pacifism of air nomads. You could theoretically argue that if the Avatar can be considered both combatant and weapon, the air temples could be accused of harboring the Avatar. But this is problematic. The knowledge that the next Avatar was going to be born into the air nomads was common, but within which one of the temples was not a known fact. Moreover, there was no way of knowing whether the Avatar would have come into their power at this point. My argument is that until either a) the Avatar comes into their power; and/or, b) begins actively participating in hostilities (as Aang does in Chapter Two), they cannot yet be considered a weapon. You can argue that, by that same token, how was the Fire Nation to know that the Avatar hadn’t come into their power? But it’s precisely for that uncertainty that the Fire Nation should have exercised caution. Doubt as to an individual’s status redounds in favor of civilian – and therefore, protected – status.
Also, by any stretch, the Fire Nation’s attack was entirely disproportional. They destroyed the population of every single air temple, intentionally – combatant or otherwise. Even if there was a fully realized Avatar at any of these temples, I don’t think you could argue that the destruction of the entire air nomad population was a matter of military necessity such that it warranted such extensive massacre. Of course, just because the air nomads were pacifist, that doesn’t mean they were defenseless. The fact that Monk Gyatso’s body was surrounded by several Fire Nation soldiers shows that at least the masters were perfectly capable of combat – but does that make them combatants? Or could you argue this is another instance of levée en masse? My understanding is that the air nomadic population as a whole was not constituted by lawful combatants. Therefore, the Fire Nation’s attack on the air temples was a violation of the laws of war, even setting aside the genocidal nature of it.
This episode also provides the main evidence that helped me come to my conclusion that the Avatar state is a weapon – and a potential weapon of mass destruction, at that. Aang – understandably – totally loses control of himself upon seeing his mentor and friend Monk Gyatso’s body. He could have destroyed everything, including – potentially – his friends, if Katara hadn’t been able to get through to him. I wonder if you could extend that analogy to dams, dykes and nuclear power.
Phew. That was a heavy episode.
Chapter Six: Imprisoned
One of the recurring themes in ATLA (at least w/r/t IHL) is that of belligerent occupation. The war has been going on for so long at this point, and the most active hostilities are between the Earth Kingdom and the Fire Nation. As a result, there are several Earth Kingdom territories that are de facto occupied by the Fire Nation. A territory is said to be occupied by a belligerent power if the following test4 is satisfied:
- The belligerent power is present in the territory without consent;
- The belligerent power is able to displace the rightful power;
- The belligerent power is able to put in place an administration that adequately substitutes that of the rightful power.
I think this test is pretty clearly satisfied. I also argue though that by the end of the episode, the occupation is on tenuous ground again, given that Katara is able to pretty competently resurrect the resistance against the Fire Nation occupation among the interned earth benders. Depending on how effective the resistance is (I argue, pretty damn effective), it could be enough to challenge if not dismantle the occupation.5
The biggest issue of note in this episode is the internment of earth benders. Immediately, we have an issue: any individual who is a bender is considered a potential combatant, and thus interned. This is problematic, not only from a moral standpoint, but also from a regime standpoint – do we then assess the conditions of internment from the standpoint of prisoners of war (GCIII) or from the standpoint of civilian internment (GCIV)? I’m going to err on the side of conservatism here and say that the people interned are at least civilians.
Well, it doesn’t look good for the Fire Nation. First of all, protection persons may only be assigned to residences or interned for “imperative reasons of security“. Now I suppose the Fire Nation could – and does, kinda – argue that all benders constitute a security threat. But for the most part, this is clearly overblown. For example, Haru only uses his bending to save another civilian from certain death. And he’s immediately abducted by the occupying Fire Nation force and interned. Certainly not imperative by any stretch. And there has to be some degree of process: interned people should be able to appeal, and the decision to intern them must be subject to review (every six months if possible). Obviously, this is not the case. So, boop, bad.
Second, you cannot intern them outside the occupied territory. Based on the description of the prison rig, it doesn’t seem like the prison rig that internees are sent to is anywhere on the actual occupied territory. Per the wiki, the rig is “several miles off the coast of the Earth Kingdom” – which does not, to me, read is being on occupied territory. Moreover, internees are put to work performing labor that benefits the Fire Nation’s military; thing is, I thought I remembered that this wasn’t okay. But it turns out that within the context of an occupation, per Article 51(2) of GCIV:
The Occupying Power may not compel protected persons to work unless they are over eighteen years of age, and then only on work which is necessary either for the needs of the army of occupation, or for the public utility services, or for the feeding, sheltering, clothing, transportation or health of the population of the occupied country.Article 51(2), Geneva Convention IV
Now… the wording “needs of the army of occupation” is a little confusing. Does that mean the needs of the entire army of the occupying power are fair game? Or is it limited to the needs of the army insofar as it is deployed to maintain the occupied territory? I’m partial to the second interpretation, and I think it’s the more conservative interpretation anyway. By the International Committee of the Red Cross’ own admission, the wording is open to interpretation.6 Per my interpretation, forcing internees to refuel any Fire Nation ship is a breach of GCIV.
But anyway, this is not relevant. If we agree – and we do! – that the very basis for interning benders is illegal, then it doesn’t matter what they’re made to do; it’s illegal by extension under any interpretation.
I noted the fact that the Fire Nation basically extorts taxes from occupied persons for the upkeep of the Fire Nation military, rather than the needs of civilians and upkeep of the occupied territory – but it doesn’t seem as though there are any rules (in the Geneva Conventions or CIL, which – again – is the legal world I’m operating in here; THOUGH PLEASE CORRECT ME IF I’M WRONG) governing taxation in occupied territories.
Also, clearly, the actual conditions of internment suck as well. They’re forced to wear rags, blankets are scarce and need to be rationed out by the internees themselves, etc. Suffice it to say, this whole episode exemplifies breaches of GCIV.
Okay, I’m exhausted. This is the first half of my Book One analysis. Look forward to my accusation against Jet as a horrible war criminal, whether attacking an abbey is okay, and whether or not the spirit of the moon is a combatant.
Why did I undertake this project…
Proportionality: “Launching an attack which may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated […]”
Distinction: “The parties to the conflict must at all times distinguish between civilians and combatants. Attacks may only be directed against combatants. Attacks must not be directed against civilians.”
Prisoner of War: Any persons defined under Article 4 of GCIII who fall into the power of the enemy.
Combatant: “All members of the armed forces of a party to the conflict are combatants, except medical and religious personnel.”
Civilian/non-combatant: Pretty much the negative space around combatants. If you’re not a combatant, you’re a civilian and therefore protected under IHL (until you directly participate in hostilities, at which point you lose protected status).
Mens rea: The knowing or intention of wrongdoing – knowing that you’re about to commit the act which is wrong. Necessary for war crimes, including genocide (see Article II of the Genocide Convention).
Belligerent Occupation: When a belligerent power, without consent, displaces or substitutes (or at least has the potential to substitute) the normal authority and administration in a given territory. An occupier power incurs certain responsibilities towards the civilians in the occupied territory. (There’s no real definition for me to refer to, so I’m trying my best!)
 Geneva Convention III, Article 4(6). The definition of levée en masse is as follows: “Inhabitants of a non-occupied territory, who on the approach of the enemy spontaneously take up arms to resist the invading forces, without having had time to form themselves into regular armed units, provided they carry arms openly and respect the laws and customs of war.”
 Geneva Convention III. I will also note that PoWs are meant to be detained in the same conditions as the detaining power’s own armed forces (Article 25).
 There are three situations in which international humanitarian law applies: international armed conflict (IAC), non-international armed conflict (NIAC), and a belligerent occupation (which can only take place as a result of an IAC). For more on the belligerent occupation test, see Armed Activities on the Territory of the Congo (Democratic Republic of the Congo v. Uganda), International Court of Justice (2005), https://www.icj-cij.org/en/case/116. Of note is a separate onion in the judgment of the case, in which a judge argues that it is enough that the belligerent power have the capacity to substitute the rightful power, even if it does not do so in actuality. This might seem counterintuitive, but think about this: the stigma of being an occupier, in an ostensibly post-colonial era of the world, is intense. So are the responsibilities of an occupier. Wouldn’t it be relatively easy for the would-be occupier to claim they are not so because, hey, look, we haven’t imposed our own administration on this territory? Judge Kooijmans argues that the potential of substitute is sufficient. I have to agree. Anyway, we’re not going to really go in that deep since I don’t think we have all the necessary information to begin with, but it’s just an interesting aside.
 See Prosecutor v. Prlić, ICTY, Case No. IT-04-74-A, Judgment (2017). The Appeals Chamber found that if the resistance against the would-be (or actual) belligerent occupier is such that it displaces authority, then you cannot say there is an occupation.