From land warfare to maritime warfare

Geneva Conventions I and II (GC I and GC II) both address the protection of wounded and sick (and, in the case of GC II, shipwrecked) members of armed forces. GC I pertains to land warfare,Footnote ⁵³ while GC II governs maritime or naval warfare.Footnote ⁵⁴ Even though these two conventions regulate distinct domains of warfare, the parallels between them are striking, both in the structure of their chapters and the phrasing of their provisions. Historical evidence suggests that these similarities are not coincidental but rather the result of a deliberate analogical reasoning between warfare on land and warfare at sea, at least in relation to the protection of people in wartime (Geneva law).Footnote ⁵⁵ Land warfare received the earliest attention from lawmakers, beginning in 1864 with the previously mentioned Geneva Convention for the Amelioration of the Condition of the Wounded in Armies in the Field.Footnote ⁵⁶ Thereafter, land warfare served as the analogical source to which maritime warfare – the analogical target – was compared. The legal framework developed for land warfare was used as a model for regulating maritime warfare, based on the assumption that the two war situations were sufficiently similar. However, where crucial differences were identified, the legal regime for land warfare was adapted to account for the specificities of maritime warfare.

The heritage collection on the First Geneva Convention is very illuminating in this regard. In his 1870 Etude sur la Convention de Genève, Gustave Moynier noted that the idea of analogizing maritime warfare with land warfare had already emerged in 1864.Footnote ⁵⁷ The draft convention submitted by the International Committee for Relief to the Wounded (ICRW)Footnote ⁵⁸ for the 1864 Diplomatic Conference included a provision on the issue in its Article 11, which stated: “Des stipulations analogues à celles qui précèdent, relatives aux guerres maritimes, pourront faire l’objet d’une Convention ultérieure entre les Puissances intéressées.”Footnote ⁵⁹ A few years later, in preparation for the 1868 Diplomatic Conference to revise the First Geneva Convention, the ICRW proposed, as a point of discussion, the extension of the existing land warfare framework to maritime warfare.Footnote ⁶⁰ The States attending the Conference approved ten additional draft provisions addressing this issue.Footnote ⁶¹ Commenting on these draft articles, Moynier wrote:

Carl Lueder, in his 1876 monograph, also examined these ten draft articles. He concurred with Moynier on the appropriateness of analogical reasoning between land and maritime warfare, stating:

Nevertheless, Lueder expressed reservations about the mapping step of the analogical process. Indeed, he appeared to disagree with some of the relevant dissimilarities identified by the 1868 Diplomatic Conference – although he did not elaborate on this point.Footnote ⁶⁴ Furthermore, his rather detailed analysis offered mild criticism on the analogical solution adopted at the Conference, suggesting that the transfer step could have been implemented differently. He proposed a more economical approach, advocating for a single provision prescribing the analogical application of the 1864 Geneva Convention to maritime warfare.Footnote ⁶⁵

During the second session of the 1869 International Conference in Berlin, which focused on voluntary rescue in maritime warfare, the analogy between land and maritime warfare was made explicit with a rare transparency on the analogical process. Dr Steinberg, rapporteur of the committee on signals for rescue vessels, explained the rationale behind the committee’s decision to adopt a common signal to notify rescue vessels to intervene on the battlefield. He stated: “Les Articles additionnels de la Convention de Genève relatifs à la guerre maritime ayant été calqués sur les articles relatifs à la guerre sur terre nous nous placerons pour ces motifs au point de vue de l’analogie.”Footnote ⁶⁶ Steinberg then identified four critical differences between land and maritime warfare to justify the need for a specific rescue signal at sea. These differences are grounded in the technological context of the time. The first difference touched upon the “action theatre”: on land, soldiers hors de combat could wait for combat to cease before receiving care, whereas at sea, immediate rescue was required in the face of shipwreck. The second difference concerned the scope and evolving nature of war theatres: land warfare was geographically limited by the number of combatants, while maritime warfare could shift in location and scale due to the use of steamboats. The third difference pertained to the nature of military power. On land, military power resided in armed forces, while at sea, it resided in vessels; the destruction of a vessel could end combat, even if the vessel crews were unharmed. The last difference referred to the consequences of defeat: on land, defeated soldiers were typically just wounded, but at sea, they were often also at risk of drowning and thus dying.Footnote ⁶⁷

Beyond the heritage collection on the First Geneva Convention, subsequent treaties on maritime warfare leave no doubt as to the importance of analogies in their development. At the turn of the twentieth century, the titles of the two Hague Conventions on maritime warfare made the lawmakers’ intent very clear. The 1899 Hague Convention III was entitled Convention (III) for the Adaptation to Maritime Warfare of the Principles of the Geneva Convention of 22 August 1864.Footnote ⁶⁸ Similarly, the 1907 Hague Convention X – which replaced the 1899 Convention – was titled Convention (X) for the Adaptation to Maritime Warfare of the Principles of the Geneva Convention.Footnote ⁶⁹ In its 1960 Commentary on GC II, the ICRC emphasized that the maritime conventions were “simply presented as an extension” of the land conventions.Footnote ⁷⁰ The ICRC has also underlined, in its updated 2017 Commentary on GC II, that the titles of the 1899 Hague Convention III and the 1907 Hague Convention X “reflected the spirit of both Hague Conventions: affirming that the protective rules agreed upon for land warfare extended to armed conflict at sea”.Footnote ⁷¹

Hague Convention X was eventually superseded in 1949 by GC II – which “received its own appropriate name” for the first time.Footnote ⁷² However, the same analogical approach persisted during the 1949 Diplomatic Conference. This approach was generally accepted by the ICRC and by State delegates who voted in favour of GC II,Footnote ⁷³ despite some disagreements regarding specific (dis)similarities between land and maritime warfare, such as those concerning the transfer of prisoners.Footnote ⁷⁴ Nonetheless, certain State representatives remained broadly sceptical. For instance, during a meeting of Committee I, the French representative argued that the analogy between the two conventions was superficial, asserting that the maritime convention “va beaucoup plus loin et … soulève des questions beaucoup plus importantes de droit international” compared to its land-based counterpart.Footnote ⁷⁵

From land and maritime warfare to air warfare

Unlike maritime warfare, which is at the very least regulated by GC II regarding the protection of wounded, sick and shipwrecked members of armed forces at sea, air warfare is not governed by a dedicated modern IHL treaty. Only two aspects of air warfare are addressed in non-specific IHL instruments:Footnote ⁷⁶ the protection of medical aircraft, as provided in Articles 36 and 37 of GC I, Articles 39 and 40 of GC II, and Articles 24–31 of AP I;Footnote ⁷⁷ and the status of aircraft occupants in distress, as regulated by Article 42 of AP I.Footnote ⁷⁸ The analogical mindset behind these provisions is less immediately apparent than that which informed the development of the maritime conventions (GC II and its predecessor treaties). There is no dedicated air warfare convention whose structure mirrors that of earlier treaties; additionally, the number of relevant provisions is limited, and they are dispersed across GC I, GC II, and AP I. These provisions also differ in form and content from other contemporary IHL norms. Further, the wording of the relevant articles in GC I and GC II is identical. At first glance, it may thus not be obvious that these air warfare provisions are based on analogies with other IHL regimes. Yet, and again, the analysis of historical documents reveals a consistent reliance on analogical reasoning for regulating air warfare (the target), drawing from both land and maritime warfare (the sources).

The retrieval step in this analogical process was neither straightforward nor uniform. In some instances, lawmakers drew from both land and maritime warfare as sources; in others, they debated which of the two was more appropriate. Such divergence is unproblematic, as lawmakers are not required to select one single, exclusive source. Lawmakers do not analogize at the general level of entire domains (in the present case, air warfare and land or maritime warfare). Even if this author identifies one or another entire domain as a source or target to simplify the language in this article, it is more accurate to consider, as previously introduced, that there are multiple analogies with distinct features of a domain as target or source situations.Footnote ⁷⁹ In other words, there are several sources and targets. Besides, disagreements on the source are to be expected because the retrieval step inherently involves subjective judgement: reasonable lawmakers may reasonably disagree on the most relevant source. Nevertheless, regardless of the source chosen, there is little doubt that analogies played a role in shaping the limited set of IHL provisions on air warfare.

A Commission of Jurists appointed by the 1922 Washington Conference adopted Article 17 of the 1922–23 Hague Rules, which addressed the control of wireless telegraphy in time of war and air warfare. This article provided that the principles applicable to land and maritime warfare should also apply to air warfare.Footnote ⁸⁰ Although not legally binding, the provision has likely acquired a customary nature over time.Footnote ⁸¹ It illustrates that the jurists did not choose between land or maritime warfare as the exclusive analogical source.

The following year, on behalf of the French Ministry of War, Dr Niclot proposed at the 11th International Conference in 1923 that the protection of medical aircrafts be included among the topics for future discussion.Footnote ⁸² In response, the ICRC prepared a draft convention in anticipation of the 12th International Conference in 1925. This draft adopted the same analogical approach as Article 17 of the Hague Rules: both land and maritime warfare were retained as source analogues. With the assistance of Charles-Louis Julliot, Paul des Gouttes – representing the ICRC – drafted a treaty entitled the Convention Additionnelle à la Convention de Genève de 1906 et à celle de La Haye de 1907 pour l’Adaptation à la Guerre Aérienne des Principes de la Convention de Genève.Footnote ⁸³ Article 1 of the draft Convention stated: “Sont applicables à la guerre aérienne toutes les prescriptions de la Convention de 1906 [land warfare] et de la X^e Convention de La Haye du 18 octobre 1907 [maritime warfare], qui peuvent lui être appliquées, et pour autant qu’elles ne sont pas modifiées par les dispositions suivantes.”Footnote ⁸⁴ The draft was approved by the 12th International Conference with only minor changes.Footnote ⁸⁵

Air warfare was not on the official agenda of the 1929 Diplomatic Conference, but both the French and British delegations raised the issue.Footnote ⁸⁶ The special committee tasked by Commission I with examining sanitary aviation voted, by six votes to one, in favour of incorporating a single provision into the revised Geneva Convention, rather than drafting multiple articles or a separate treaty.Footnote ⁸⁷ As a result, Article 18 was adopted in the 1929 Geneva Convention for the wounded and sick in the field.Footnote ⁸⁸ The first paragraph of this provision provides that “[a]ircraft used as means of medical transport shall enjoy the protection of the Convention during the period in which they are reserved exclusively for the evacuation of wounded and sick and the transport of medical personnel and material”.Footnote ⁸⁹ The incorporation of Article 18, which forms the chapter relating to medical transports together with Article 17 on mobile medical formations, reflects IHL lawmakers’ main intent to analogize air warfare with land warfare, and more specifically, medical aircraft with mobile medical units.Footnote ⁹⁰

The analogy between land and air warfare persisted beyond the Second World War. The relevant provisions of GC I and GC II remain closely aligned with Article 18 of the 1929 Geneva Convention.Footnote ⁹¹ However, discussions leading up to the 1949 Geneva Conventions show that maritime warfare was, at times, preferred over land warfare as the analogical source. For example, during the 1947 Conference of Government Experts, the Australian representative proposed in Committee II that civilian aircrew members be analogized with merchant seamen.Footnote ⁹² Similarly, in preparation for the 1949 Diplomatic Conference, the ICRC suggested defining hospital aircraft by analogy with hospital ships.Footnote ⁹³

Two decades later, the analogical approach remained prominent both for IHL experts and the ICRC. Regarding the status of aircraft occupants (in distress), the ICRC surveyed experts prior to the 21st International Conference in 1969. Some experts compared the “airman in distress” to “a shipwrecked individual”, and “armed parachutists” to “combatant[s] proceeding to attack or in flight”.Footnote ⁹⁴ Notably, the ICRC report even used the term “air-wrecked”.Footnote ⁹⁵ Ahead of the second Conference of Government Experts in 1972, the ICRC introduced a new provision on the status of aircraft occupants in its draft additional protocol. The commentary accompanying the draft stated:

Regarding the protection of medical aircraft, the report of the second Conference of Government Experts reveals divergent views, particularly concerning the capture of permanent medical aircraft crews. One expert asserted that such crews should enjoy the same status as hospital ship crews under GC II (i.e., not subject to capture), while others contended that they should be treated like other medical transport personnel under GC I (i.e., subject to capture). The latter opinion was ultimately adopted by Commission I.Footnote ⁹⁷

The final records of the 1974–77 Diplomatic Conference provide further examples of analogies. During the 39th plenary meeting, the Syrian delegate proposed analogizing a paratrooper to a land soldier attempting to escape – both, in his view, should be excluded from protection. However, he rejected the analogy between “an aviator trying to return to his territory” and a shipwrecked person, arguing that the former is “hors de combat and attempting to escape”, and therefore should not benefit from protection.Footnote ⁹⁸ During the 47th meeting of Committee II, the Canadian representative analogized a “forward helicopter” to a “wheeled ambulance”.Footnote ⁹⁹ Presenting a series of draft articles to Committee III, the ICRC affirmed:

Note that the prohibition on attacks against aircraft occupants in distress is not only established by Article 42 of AP I but also forms part of customary international law, as confirmed by Rule 48 of the ICRC Customary Law Study.Footnote ¹⁰¹ In its commentary to the rule, the ICRC highlights that “[a] parallel can be drawn here with the shipwrecked, who are considered to be hors de combat … even though they may swim ashore or be collected by a friendly ship and resume fighting”. The ICRC adds that “persons bailing out of an aircraft in distress have been called ‘shipwrecked in the air’”.Footnote ¹⁰²

To conclude on the regulation process of air warfare, these instances of analogical reasoning confirm that drawing analogies between air warfare and either land warfare – especially when addressing medical aircraft – or maritime warfare – especially when addressing aircraft occupants in distress – was the approach of both the lawmakers and the ICRC in respectively adopting or supporting successive treaty provisions and in shaping customary international law.

From prisoners of war to civilian internees

It is hardly a secret that Title II, Section IV of Geneva Convention IV (GC IV)Footnote ¹⁰³ on the protection of civilians closely mirrors Title III, Section II of Geneva Convention III (GC III) on the protection of prisoners of war (PoWs).Footnote ¹⁰⁴ Even a non-expert in IHL would likely notice the striking similarities upon a first reading of the two conventions. In its 1958 Commentary on GC IV, the ICRC emphasized that “the regulations applicable to civilians reproduce almost word for word the regulations relating to prisoners of war”.Footnote ¹⁰⁵ The updated 2025 ICRC Commentary on GC IV also underlines that, “[f]rom the beginning, protections for civilians, in particular for civilian internees, were modelled on the protections afforded to prisoners of war”.Footnote ¹⁰⁶

This resemblance was not achieved by chance: IHL lawmakers built Title II, Section IV of GC IV upon an analogical reasoning between, on the one hand, the internment of PoWs and, on the other, the internment of civilians. In its preliminary remarks to the four Geneva Conventions, the ICRC explicitly indicated that “Section IV deals with internment. It is divided into twelve Chapters, the contents of which are in general analogous to the provisions adopted for prisoners of war.”Footnote ¹⁰⁷ Thus, in the analogical process, the internment of PoWs served as the source situation, while the internment of civilians constituted the target situation. Historical documents and preparatory works contain numerous direct references to this analogical reasoning.

The idea of analogizing civilian internees with PoWs dates back to 1934. The Tokyo Draft International Convention – concerning the condition and protection of civilians of enemy nationality in the territory of a belligerent or in a territory occupied by it – articulated the analogical reasoning in one single provision. Article 17 of the Tokyo Draft stipulated that “the Convention of July 27, 1929, concerning the treatment of Prisoners of War is by analogy applicable to Civilian Internees”.Footnote ¹⁰⁸ The same idea was considered crucial at the 1946 Preliminary Conference of National Red Cross SocietiesFootnote ¹⁰⁹ and at the 1947 Meeting for the Study of Treaty Stipulations Relative to the Spiritual and Intellectual Needs of Prisoners of War and Civilian Internees.Footnote ¹¹⁰ During both gatherings, however, the need for (additional) rules specific to civilian internment was also debated. The Preliminary Conference was divided on this issue,Footnote ¹¹¹ while the 1947 Meeting “decided that several distinct rulings should be adopted for Civilians”.Footnote ¹¹² Similarly, Commission III of the 1947 Conference of Government Experts – tasked with examining the protection of civilians in wartime – argued that simple cross-referencing to the legal regime applicable to PoWs was insufficient. The Commission maintained that most provisions concerning PoWs should be adapted when being transferred to civilian internees.Footnote ¹¹³

By the time of the 1949 Diplomatic Conference, the analogy between PoWs and civilian internees was largely taken for granted.Footnote ¹¹⁴ Rather than listing the positive references to this reasoning, it is more revealing to highlight the negative reactions, which underscore the centrality of the analogical mindset among lawmakers. Even those who opposed the analogy felt compelled to refer to it. They did not reject the relevance of analogizing as such, but they instead used analogical reasoning (and thus vocabulary) to highlight critical differences, invalidate the suggested analogy and justify their objections. For instance, during the 20th meeting of Committee III, the Canadian representative affirmed that “the analogy between internees and prisoners of war had been carried too far” in the context of war allowances. He contended that enemy aliens should not receive allowances similar to those granted to PoWs, as the latter “had earned a standard of treatment which had not been earned” by the former.Footnote ¹¹⁵ At the 22nd meeting of the same forum, the Canadian delegate also declared, in relation to another issue, that

In the same vein, the British representative considered that, in contrast to PoWs, civilian internees should not enjoy more favourable treatment than the Detaining Power’s own citizens. He remarked at the 31st plenary meeting that “it would be difficult to find a falser analogy than that between prisoners of war and internees in the present connection”.Footnote ¹¹⁷

In conclusion, the similarities between GC III and GC IV regarding internment are not historical accidents, but rather are the result of a deliberate choice made by the lawmakers in the drafting of the Conventions. Lawmakers intentionally constructed an analogical bridge between, on the one hand, the existing and revised legal regime for PoWs – dating back to the 1929 ConventionFootnote ¹¹⁸ – and, on the other, the newly established regime for civilian internees.

From international to non-international armed conflicts

Among the hundreds of articles in the Geneva Conventions of 1949, only one common provision addresses NIACs: common Article 3.Footnote ¹¹⁹ AP I contains 102 articles applicable to IACs, whereas Additional Protocol II (AP II) includes twenty-eight articles which govern certain types of NIACs.Footnote ¹²⁰ This significant imbalance in the number of provisions between the legal regimes applicable to IACs and NIACs may obscure the analogical connections between them. Nevertheless, the preparatory works of the Geneva Conventions and Additional Protocols show that the treaty legal framework for NIACs was conceived through analogical reasoning with IACs. Thus, in the analogical process, NIACs served (and often still serve) as the target situation, while IACs functioned as the source.Footnote ¹²¹ This article does not aim to provide a comprehensive historical account of the development of common Article 3 and AP II, but the following paragraphs, while not exhaustively indexing all analogical occurrences throughout the drafting history of common Article 3 and AP II, pinpoint a few key instances of analogical reasoning to highlight its importance in the lawmakers’ decisions.

During the 1949 Diplomatic Conference, the Joint Committee – tasked with examining the common articles to all four Geneva Conventions – established a special subcommittee to consider, among other issues, the applicability of IHL to NIACs. This subcommittee instructed two successive working parties to draft a new common provision addressing this matter. According to the report of the 23rd meeting of the special subcommittee, the second working party ultimately abandoned the analogy between NIACs and IACs, even though it had initially explored this approach:

The Joint Committee’s report to the plenary session confirmed that the lawmakers in this committee considered but ultimately rejected the idea of analogizing NIACs with IACs.Footnote ¹²³ This is also clear from the fact that the Committee endorsed the second working party’s final determination to invalidate the analogy with IACs: indeed, the draft prepared by the said second working party – which “lays down a minimum of humanitarian rules which both Parties [to a NIAC] are bound to respect” – was amended by the special subcommittee and subsequently approved by the Joint Committee itself.Footnote ¹²⁴ Thus, in 1949, NIACs were not considered sufficiently similar to IACs to transfer the latter’s legal regime to the former. Nonetheless, even if considered unsuccessful, the analogical reasoning between the two types of armed conflicts was at the core of the discussions.

In 1972, the ICRC submitted two draft protocols to government experts, which “contained identical, or at least very similar, provisions on certain subjects”.Footnote ¹²⁵ Commission II of the Conference of Government Experts examined the protection of victims in NIACs. The first issue debated was the “degree of similarity in the protection of victims of both types of conflict and hence the question of whether one or two protocols are needed”.Footnote ¹²⁶ While some experts favoured a single protocol applicable to both IACs and NIACs, many others supported the adoption of two distinct protocol instruments – these experts argued that the “basic principles” of AP I “could be adapted, or purely and simply adopted” in AP II.Footnote ¹²⁷ They identified two crucial differences between IACs and NIACs: the “political aspects of the two types of conflict” and the “conditions affecting the implementation of the two Protocols”.Footnote ¹²⁸ Beyond these occurrences, the analogical vocabulary is present in the report of the second session of the Conference.Footnote ¹²⁹

The final version of AP II, with its twenty-eight articles, is significantly shorter than AP I. This disparity may suggest a lack of analogical intent, but this would be an incorrect conclusion. The approved version of AP II was the result of a last-minute compromise – the penultimate draft of AP II more largely resembled AP I and illuminates the analogical mindset between IACs and NIACs in the drafting process of the adopted version of AP II. In addition, several State delegates referred to analogical reasoning during the 1974–77 Diplomatic Conference.Footnote ¹³⁰ The 1987 ICRC Commentary on AP II notes that

However, a number of States were opposed to such a protective draft, either because it threatened national sovereignty and the principle of non-interference with internal affairs, or because it was too detailed and difficult to apply in NIACs.Footnote ¹³² There was thus a real risk that no consensus at all on AP II would see the light of day at the 1974–77 Diplomatic Conference. To avoid a complete deadlock on NIACs, the Pakistani delegation suggested a simplified (or rather reduced) version of AP II on 31 May 1977.Footnote ¹³³ The remaining provisions “did not include any drafting modifications” compared to the penultimate draft of AP II;Footnote ¹³⁴ therefore, the remnants of this draft still draw on analogies with IACs, even though the shrinkage of provisions makes it less explicit.

To conclude this section on the first category of analogies (analogies before IHL norms), it is clear that analogies underpin many well-known IHL norms. Although States are the formal IHL lawmakers, their representatives are human beings – and, like all human beings, they naturally tend to reason by analogy. As such, IHL lawmakers very frequently rely on analogical reasoning to design new rules. As the following section will show, they also sometimes oblige or invite those who apply or interpret the law to do the same.

Express obligation

IHL lawmakers sometimes adopt a norm using explicit language that compels the norm-appliers or norm-interpreters to reason by analogy. Such express obligations to reason by analogy are rare. The 1949 Geneva Conventions offer a very good example, although this approach was later abandoned in AP I. Articles 4 of GC I and 5 of GC II provide that

Interestingly, the US Law of War Manual mentions these provisions as examples of “treaty requirement[s] to apply rules by analogy”.Footnote ¹³⁷

In its 2016 Commentary on GC I, the ICRC notes that “[s]ince, by definition, neutral Powers are not Parties to the international armed conflict, the application expected of them is ‘by analogy’, as if they were Parties to the conflict (mutatis mutandis)”.Footnote ¹³⁸ In other words, the drafters of the Geneva Conventions identified the source situation as the reception and detention of the wounded, sick and shipwrecked by an adverse belligerent party, and the target situation as their reception and internment by a neutral power. However, they did not specify the (dis)similarities between the source and target situations (mapping), nor did they explain how to apply the provisions of GC I and II to neutral powers (transfer). Consequently, norm-appliers and norm-interpreters – including neutral powers and individuals such as judges – retain a degree of discretion in their analogical reasoning.

Although the terminology in Articles 4 of GC I and 5 of GC II was ultimately adopted, the phrase “by analogy” was not universally accepted and sparked debate during the 1949 Diplomatic Conference. The UK, in particular, was highly critical, describing the phrase as “difficult to interpret”, “confusing the minds of serious students”, “completely loose” or “far too wide”.Footnote ¹³⁹ The UK questioned the retrieval step in the analogical process and emphasized a fundamental difference between the source and target: while belligerents are fighting one another in the source situation, neutral powers provide assistance in a charitable manner in the target situation.Footnote ¹⁴⁰ At the 26th meeting of Committee I, the British delegate suggested an alternative: to explicitly list the provisions applicable to neutral powers, rather than referring to an analogy. This alternative did not convince all State representatives, including the delegate from the USSR, who responded that “no list could provide for all possible cases”.Footnote ¹⁴¹ The British proposal was eventually rejected by seventeen votes to six, with four abstentions.Footnote ¹⁴²

This was not the last time that the phrase “by analogy” faced opposition. The draft of AP I initially included a provision similar to Articles 4 of GC I and 5 of GC II; State experts retained the same wording during the second session of the Conference of Government Experts in 1972.Footnote ¹⁴³ However, during the 1974–77 Diplomatic Conference, State delegates reached a different conclusion. Other English-speaking countries, such as Australia, criticized the phrase “by analogy” as ambiguous in the English version (though acceptable in the French version).Footnote ¹⁴⁴ While the Drafting Committee also initially chose to maintain the terminology for the sake of consistency,Footnote ¹⁴⁵ New Zealand and other sponsoring States submitted an amendment to remove the disputed phrase.Footnote ¹⁴⁶ This amendment was adopted by twenty-seven votes to ten, with four abstentions.Footnote ¹⁴⁷ As a result, unlike Articles 4 of GC I and 5 of GC II, Article 19 of AP I does not expressly require the norm-appliers or norm-interpreters to employ analogical reasoning. It merely states that “[n]eutral and other States not Parties to the conflict shall apply the relevant provisions of this Protocol to persons protected by this Part who may be received or interned within their territory”.Footnote ¹⁴⁸

These historical developments in the drafting of GC I, GC II and AP I clearly highlight some opposition to the phrase “by analogy”, thereby illustrating how analogical reasoning, as an explicit treaty requirement, was undoubtedly at the core of the discussions. These developments also lead us to the next section of this article. The change and deletion of the phrase “by analogy” in AP I was justified by a fear that neutral powers would be considered to be on the exact same footing as belligerent parties, while it was undisputed that some provisions of AP I relating to the protection of persons could not apply to them.Footnote ¹⁴⁹ Nonetheless, in its 1987 Commentary on AP I, the ICRC aptly notes that this change was “only a question of wording, for basically it is quite clear that the situation under the Conventions was precisely the same”.Footnote ¹⁵⁰ Consequently, if analogical reasoning is, after all, not an express requirement within Article 19 of AP I, the drafters of that provision do implicitly oblige the norm-appliers or norm-interpreters to resort to such reasoning in continuation of Articles 4 of GC I and 5 of GC II. Such an implicit obligation is indeed another option for lawmakers, and one that we shall further explore in the following section.

Implicit obligation

Lawmakers rarely explicitly compel norm-appliers or norm-interpreters to engage in analogical reasoning. They more frequently impose such reasoning in an implicit way, particularly through the legislative technique of open listing.Footnote ¹⁵¹ This technique involves enumerating several situations as examples (or sources), thereby requiring the norm-appliers or norm-interpreters to apply the norm to analogous situations (or targets) not expressly mentioned in the text. In doing so, norm-appliers or norm-interpreters broaden the scope of application of a norm to encompass similar cases or situations.Footnote ¹⁵²

The selection of examples or sources by lawmakers is far from being a trivial step. Gérard Cornu notes that “un exemple légal n’est pas une application quelconque, mais une illustration choisie pour être topique”.Footnote ¹⁵³ Actually, lawmakers themselves reason by analogy to guide norm-appliers or norm-interpreters to these specific sources. All the listed sources share a common feature: they exemplify a general legal principle.Footnote ¹⁵⁴ To properly reason analogically, norm-appliers and norm-interpreters must first discern the general legal principle underlying the lawmakers’ choice of sources, and then determine whether the target situation constitutes another valid illustration of that principle.

This type of analogy through the legislative technique of open listing is typically signalled by specific vocabulary such as “and (other) similar…” or “and (other/all) analogous…”. Numerous examples can be found throughout history and the current framework of IHL, whether in declarations or treaties. For instance, the 1899 Hague Declaration (IV, 1) states that “[t]he Contracting Powers agree to prohibit, for a term of five years, the launching of projectiles and explosives from balloons, or by other new methods of a similar nature”.Footnote ¹⁵⁵ Article 5 of the 1922 Treaty Relating to the Use of Submarines and Noxious Gases in Warfare establishes signatories’ agreement on the prohibition of “[t]he use in war of asphyxiating, poisonous or other gases, and all analogous liquids, materials or devices”.Footnote ¹⁵⁶

Another illustrative example is to be found in Article 12 of GC I and of GC II. This common provision obliges parties to armed conflicts to treat and care for wounded, sick or shipwrecked members of the armed forces “without any adverse distinction founded on sex, race, nationality, religion, political opinions, or any other similar criteria” (in French: “sans aucune distinction de caractère défavorable basée sur le sexe, la race, la nationalité, la religion, les opinions politiques ou tout autre critère analogue”).Footnote ¹⁵⁷ The 2016 ICRC Commentary on GC I indicates that “[t]he drafters wisely anticipated a dynamic evolution of the catalogue of prohibited criteria and chose a sufficiently open formulation which could accommodate additional grounds”.Footnote ¹⁵⁸

Discussions during the 1949 Diplomatic Conference reveal that this kind of wording (“other similar …”/“autre … analogue”) in Article 12 entails analogical reasoning. The Afghan delegation, for instance, objected to the phrase “critères analogues” on the grounds that it did not see any similarity between the explicitly listed criteria.Footnote ¹⁵⁹ At the eighth meeting of Committee III, the Afghan delegate argued:

The Danish representative disagreed, however, stating at the 24th plenary meeting that

In any case, this type of wording now appears to be part of customary international law, as reflected in Rule 88 of the ICRC Customary Law Study.Footnote ¹⁶²

APs I and II provide further examples of an open-ended list. Article 51(5)(a) of AP I prohibits

Article 1(2) of AP II provides that “[t]his Protocol shall not apply to situations of internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature, as not being armed conflicts”.Footnote ¹⁶⁴

In contrast to the set of examples in the previous section (in which lawmakers explicitly require norm-appliers and norm-interpreters to engage in analogical reasoning), the language used in these examples containing open-ended lists does not initially appear to be particularly compelling. Nevertheless, it is reasonable to argue that norm-appliers are under an implicit obligation to apply these norms to cases that are analogous to those explicitly contemplated by the lawmakers. They cannot ignore a blatant similarity between two situations, one foreseen by the lawmakers and the other not, and choose not to apply the relevant IHL norm to the latter. To illustrate this point with some of the previous examples, when confronted with new means of warfare that are similar to existing ones whose use is already prohibited under a norm with an open-ended list, norm-appliers cannot simply disregard this similarity and associated prohibition, and proceed to use these new means on the battlefield. There is good reason to argue that such use would constitute a violation of IHL. Similarly, it is unacceptable and arguably unlawful for norm-appliers to discriminate against the wounded and sick based on a new criterion not explicitly mentioned by the lawmakers when that criterion closely resembles those already provided.

Implicit invitation

In addition to compelling norm-appliers and norm-interpreters to reason by analogy, IHL lawmakers may alternatively invite them to do so. Such an invitation amounts to an authorization or a discretionary faculty for norm-appliers, rather than an obligation to analogize. The collected data indicate that this type of invitation is implicit in IHL norms.

A paradigmatic example of this subcategory can be found in common Article 3, which reflects the analogy at the origin of this provision and of AP II – namely, the analogical reasoning between IACs and NIACs.Footnote ¹⁶⁵ Paragraph 3 of common Article 3 states that “[t]he Parties to the conflict should further endeavour to bring into force, by means of special agreements, all or part of the other provisions of the present Convention”.Footnote ¹⁶⁶ This provision encourages parties to a NIAC to adopt special agreements that transfer norms applicable to IACs to NIACs. As Marco Sassòli has noted, it thereby prompts parties to engage in analogical reasoning between IACs and NIACs through the mechanism of special agreements.Footnote ¹⁶⁷ Although this invitation to analogy does not appear to be expressly articulated in the preparatory works of the Conventions,Footnote ¹⁶⁸ it nonetheless emerges as the practical effect of the provision. Beyond common Article 3, Article 19(2) of the Hague Convention for the Protection of Cultural Property similarly contains an implicit invitation to analogize through special agreements.Footnote ¹⁶⁹

In this second category of analogies (analogies in IHL norms), lawmakers still initiate the analogical reasoning process, either by imposing it (as in the first two subcategories) or by suggesting it (as in the third subcategory) to the norm-appliers and norm-interpreters. However, as will be shown in the third and final category of this taxonomy, lawmakers are not always involved in the analogical process.

Definition of concepts

Regarding the first scenario described above, numerous examples of such analogies have emerged following the adoption of the Geneva Conventions and their Additional Protocols. Norm-appliers and norm-interpreters quickly recognized that key concepts were left undefined in the treaties, necessitating interpretation to enable their practical application. This gap was often addressed through analogy with earlier or contemporary legal instruments that provided definitions for similar concepts and situations in different contexts. In his work on analogical reasoning in international law, Jean Salmon wrote that the source “est fréquemment trouvé[e] dans des textes conventionnels distincts (antérieurs, contemporains ou postérieurs) de celui qui fait l’objet [de la cible], mais qui portent sur la même matière ou la même notion juridique”.Footnote ¹⁷¹

Instances of third-category analogies can thus be found in successful attempts to clear up the remaining grey areas in the dense text of the four Geneva Conventions. To mention just one, GC IV refers to “occupied territory/ies” approximately forty times without offering a definition of the phrase.Footnote ¹⁷² It is now well accepted that this concept can be interpreted by analogy with Article 42 of the Hague Regulations, which now have a customary nature.Footnote ¹⁷³ This provision states that “[t]erritory is considered occupied when it is actually placed under the authority of the hostile army”.Footnote ¹⁷⁴ In its 2025 updated Commentary on GC IV, the ICRC declares that “the notion of occupation is derived from Article 42”.Footnote ¹⁷⁵ A similar approach was adopted by the International Criminal Tribunal for the former Yugoslavia (ICTY) in its Kordić and Čerkez judgment of February 2001. The Trial Chamber affirmed the following:

The legal developments since the adoption of AP II are particularly illustrative of this third category of analogies. As previously noted, AP II retains elements of a more comprehensive draft that was more explicitly inspired by AP I.Footnote ¹⁷⁷ Many provisions, including definitional clauses, were ultimately deleted.Footnote ¹⁷⁸ As a result, AP II employs several terms – such as “military objectives” (Article 15) and “medical personnel” (Articles 9 and 12) – without defining them. This has led norm-appliers and norm-interpreters to rely on definitions provided in AP I.Footnote ¹⁷⁹ For example, in its Statement of Understanding dated 20 November 1990, the government of Canada declared that it “understands that the undefined terms used in Additional Protocol II which are defined in Additional Protocol I shall, so far as relevant, be construed in the same sense as those definitions”.Footnote ¹⁸⁰

Analogical reasoning with prior legal regimes is a common feature of IHL and extends beyond the frameworks of the Geneva Conventions and their Additional Protocols. To take one such example, Protocol III to the Convention on Certain Conventional Weapons – the Protocol on Prohibitions or Restrictions on the Use of Incendiary Weapons – uses the notions of “civilian” and “civilian population” but does not define them, even in the definitional provision of Article 1. Therefore, upon ratifying Protocol III in February 1995, the UK stated that “[t]he terms ‘civilian’ and ‘civilian population’ have the same meaning as in Article 50 of the First Additional Protocol of 1977 to the 1949 Geneva Conventions”.Footnote ¹⁸¹

While the above examples show analogies with definitions already established in IHL, norm-appliers and norm-interpreters have sometimes drawn analogies with definitions established under other branches of international law. International judicial practice offers a salient example of this. Clear analogical occurrences with other branches of international law are relatively limited in international case law because States have the primary law-making authority and most significant legal reforms have happened through conventions or customary law.Footnote ¹⁸² In addition, international judges rarely acknowledge that they are filling a legislative gap and resorting to analogical reasoning.Footnote ¹⁸³ However, they offer some interesting examples of analogy, as illustrated by the following paragraphs.

IHL lawmakers did not define torture in wartime (target); consequently, the ICTY interpreted the notion of torture under IHL by analogy with international human rights law (IHRL). It retrieved a source (i.e., the situation of torture in peacetime), mapped it onto the target, and transferred the legal solution of the source to the target – i.e., it transferred and adapted the legal definition of torture under IHRL to IHL.Footnote ¹⁸⁴

This example is well known among IHL experts, although it is not typically analyzed through an analogical lens, as the ICTY judges did not present their reasoning as such. Nevertheless, the language used by the ICTY judges to explain their reasoning clearly reflects the steps of analogical reasoning. In the 2001 Kunarac judgment, the Trial Chamber held that “[b]ecause of the paucity of precedent in the field of international humanitarian law, the Tribunal has, on many occasions, had recourse to instruments and practices developed in the field of human rights law”.Footnote ¹⁸⁵ It further noted that “notions developed in the field of human rights can be transposed in international humanitarian law only if they take into consideration the specificities of the latter body of law”.Footnote ¹⁸⁶ The following year, in the 2002 Krnolejac judgment, Trial Chamber II emphasized that

These excerpts reveal that the ICTY employed terms such as “structural differences”, “common aspects” and “specificities” – all of which are central to the mapping process. Moreover, the use of verbs such as “transpose” closely corresponds to the transfer process. It is difficult to deny that the ICTY reasoned by analogy to define torture under IHL, and this definition, inspired by IHRL, now appears to be accepted as part of customary IHL.Footnote ¹⁸⁸

Extension of applicability

The definition of terms is not the only challenge encountered in the application and interpretation of IHL norms. At times, the scope of a provision may appear unduly restrictive to norm-appliers and norm-interpreters, prompting them to reason by analogy to extend its applicability.

One example of this concerns the exception to the special protection of medical units, as provided in Article 21 of GC I, Article 13 of AP I and Article 11 of AP II.Footnote ¹⁸⁹ These provisions stipulate that medical units may lose their protection if they engage in acts harmful to the enemy. This exception to the special protection of medical units has been extended to medical personnel; although the articles specifically addressing the protective status of medical personnel – namely Article 24 of GC I, Article 15 of AP I and Article 9 of AP IIFootnote ¹⁹⁰ – do not mention such an exception, it is nonetheless accepted that it applies to medical personnel by analogy.Footnote ¹⁹¹ Consequently, medical personnel may also lose their protection if they commit acts harmful to the enemy.

A second example relates to the applicability of common Article 3 to IACs. Historically this provision has applied only to NIACs, but it is now considered to be also applicable to IACs by analogy. This was put forward by the International Court of Justice (ICJ) in its 1986 Nicaragua judgment.Footnote ¹⁹² However, the Court’s analogical reasoning involved the use of “fundamental principles of humanitarian law”: common Article 3 governs IACs because it reflects such principles, also called “elementary considerations of humanity”, which are relevant for both NIACs and IACs.Footnote ¹⁹³ The analogical bridge between NIACs and IACs is more directly established by the ICTY in its Kuprečkić judgment through an analogy a minore ad maius (literally, from the smaller to the greater) – or an analogy a fortiori.Footnote ¹⁹⁴ In this case, the Trial Chamber examined the prohibition of reprisals against civilians in IACs. Among other developments, it considered that common Article 3 prohibits reprisals against civilians in NIACs and then declared that “it would be absurd to hold that while reprisals against civilians entailing a threat to life and physical safety are prohibited in civil wars, they are allowed in international armed conflicts as long as the civilians are in the combat zone”.Footnote ¹⁹⁵ The ICTY judges were thus of the opinion that, if reprisals against civilians are prohibited in NIACs, the least regulated form of armed conflicts, then they must be prohibited in IACs, which are governed by a more detailed legal regime. Thereby, the international judges agreed to extend the application of common Article 3 to IACs.

To conclude the historical research and taxonomy, this analysis has identified three categories of analogies in IHL: analogies before, in and after IHL norms. While the first two categories correspond respectively to a standard legislative technique and a common method of reasoning to be used in the application of IHL norms, the third pertains to analogical interpretation. If the three categories are similar in terms of frequency (all three are very common in IHL, and an in-depth analysis of State practice would probably confirm this observation), they differ in their significance for IHL development. The first category often involves sets of analogies that contribute to the evolution or the emergence of entire legal regimes, whereas the second and third typically concern more isolated or single IHL rules – at least in the current and accepted IHL framework. The following section offers an axiological analysis of the data collected on analogical reasoning in IHL and, thus, of the three categories previously suggested.

Internal coherence

At first sight, two features of IHL’s past development appear to challenge coherence. First, similar aspects of warfare were regulated at different points in time. In other words, IHL’s normative content evolved in successive layers over two centuries: the protection of the wounded and sick first on land, then at sea; the protection of other vulnerable groups such as PoWs and later civilians, including civilian internees; the regulation of various types of armed conflicts, beginning with IACs and later extending to NIACs, etc. Second, in more recent decades, IHL has expanded rapidly, with a proliferation of treaties and accepted customary norms in the late twentieth and early twenty-first centuries. While this layered evolution could have led, over time, to fragmentation, and the recent rapid growth risked creating a cluttered framework where each addition did not take into account previous achievements,Footnote ²⁰³ analogies have served as a powerful tool to preserve internal coherence – i.e., a sense of uniformity and harmony across IHL legal regimes.

The three categories of IHL analogies very often have a circular nature. Both the target and the source are usually found within IHL itself; in other words, the inspiration to develop IHL through analogical reasoning directly comes from IHL. This self-referential approach has proven effective in preserving IHL’s internal coherence, as analogies have promoted consistent treatment of similar cases through a principle of replicability.Footnote ²⁰⁴ In doing so, analogies have also contributed to the “perpetuation of [IHL’s] legislative architecture”.Footnote ²⁰⁵ They have supported internal coherence not only in IHL’s content, but also in its normative structure. The above section on the taxonomy of IHL analogies highlighted numerous structural alignments across conventions: GC II mirrors GC I, Title II(IV) of GC IV replicates Title III(II) of GC III, and AP II is essentially a remnant of an original draft closely resembling AP I.Footnote ²⁰⁶

Moreover, consecutive analogies reflect a collaborative effort of multiple actors over time, whether they are successive lawmakers or some lawmakers together with norm-appliers and norm-interpreters. These actors gradually accept and confirm legal ideas in the course of time.Footnote ²⁰⁷ For instance, lawmakers drew inspiration from land warfare to regulate maritime warfare, and subsequently from maritime warfare to address air warfare. This iterative process fostered coherence among the legal regimes governing these domains. Different State representatives participated in the Diplomatic Conferences over time, and the underlying legal principles were endorsed repeatedly: first for land warfare, then for maritime warfare, and finally for air warfare.

These observations allow us to conclude that some internal coherence exists within IHL and that analogical reasoning has significantly contributed to it. When selecting the analogical source, the analogizers – often IHL lawmakers in the present case – exercise discretion. They have “a choice of source analogs and selec[t] one from among multiple possibilities”.Footnote ²⁰⁸ Yet, they have consistently selected sources within IHL, thereby reinforcing its internal coherence.

Nonetheless, it is reasonable to acknowledge that the circular nature of IHL analogies may stem more from a combination of circumstances than from a pledge for internal coherence. First, wartime is an exceptional context, and IHL is a legal regime designed to operate under exceptional circumstances rather than as a default framework.Footnote ²⁰⁹ Given IHL’s specificity, it may be difficult to draw analogies from other branches of law. This raises the issue of IHL exceptionalism – i.e., its distinctive nature compared to other branches of (international) law, characterized by the establishment of rules that diverge from the normative standards typically and normally governing international society. Second, modern IHL began developing in the mid-nineteenth century – earlier than related fields such as IHRL. For decades, wartime situations were regulated before similar peacetime situations, leaving few – if any – external analogical sources. Third, the necessarily limited legal knowledge of IHL stakeholders may also explain the circular nature of IHL analogies. Since analogies move from the “known” to the “unknown”, analogizers naturally draw from their own “informational and experiential background”Footnote ²¹⁰ – and IHL experts are most familiar with IHL itself rather than with any other branch of international law or domestic law.

In any case, regardless of whether this circularity is intentional or incidental, analogies have effectively safeguarded IHL’s internal coherence. This internal coherence, in turn, goes hand in hand with other key values. It fosters stability in the IHL framework,Footnote ²¹¹ enabling IHL norm-appliers to align their behaviours with a stable legal structureFootnote ²¹² and anticipate the direction of future development in the field; there is some predictability.Footnote ²¹³ Analogies have served as a “tool of legal planning”Footnote ²¹⁴ for IHL stakeholders, and the resulting stability and predictability have enhanced legal certainty and security.Footnote ²¹⁵ As Martti Koskenniemi highlighted in the finalized report of the Study Group of the International Law Commission on fragmentation of international law, “[c]oherence is valued positively owing to the connection it has with predictability and legal security”.Footnote ²¹⁶ Furthermore, coherence and acceptance are usually two sides of the same coin, and this remains true for IHL. IHL stakeholders are more likely to respect the rules that they accept and more likely to accept rules that are coherent.Footnote ²¹⁷ This dynamic ultimately enables IHL to fulfil its double purpose: protecting victims of war (humanity) while accommodating the necessities of warfare (military necessity).Footnote ²¹⁸

External coherence

Internal coherence is essential, but it is not enough. IHL does not operate in a legal vacuum – it is embedded within the international legal order, alongside many other branches of international law. Therefore, IHL cannot develop in isolation, disregarding the existence and evolution of other legal fields. Analogies helped in coordinating IHL with these other branches, thereby participating in the overall coherence of the international legal system. In this way, they ensured IHL external coherence – i.e., a sense of unity and harmony between IHL and other areas of International Law.

Although most IHL analogies are circular in nature, a limited number of analogical instances demonstrate that the analogical source may sometimes lie outside IHL, for example in IHRL. This is exemplified by the definition of torture discussed in the third category of the taxonomy. In that case, the analogizers, namely the judges, sought inspiration beyond IHL, turning to IHRL. This approach ensured coherence between torture situations in peacetime as governed by IHRL and torture situations in wartime as regulated by IHL. In this instance, external coherence was facilitated by the indeterminate nature of the IHL norm in question.Footnote ²¹⁹ Although lawmakers did not intend to leave uncertainty, there was nonetheless a lacuna. International judges intervened to fill this lacuna through analogical reasoning, thereby contributing to the preservation of the global coherence within the international legal order. However, when the analogizers are not IHL lawmakers but judges, analogies raise a distinct issue: that of legitimacy.Footnote ²²⁰ Judges are primarily tasked with adjudicating disputes by applying existing law, not with engaging in lawmaking through analogies. The dividing line between law-applying and law-making is sometimes blurry.

In addition to judges, IHL lawmakers themselves have also drawn inspiration from IHRL – despite the limited examples presented in the taxonomy, which might suggest otherwise. For instance, the IHL provisions on fundamental judicial guarantees, such as those in Article 75(4) of AP I,Footnote ²²¹ reflect analogies with the IHRL framework. The right to a fair trial is a classic human rights normFootnote ²²² which was later incorporated by analogy into IHL.Footnote ²²³ Thus, whether invoked by lawmakers or judges, situations governed by IHRL often serve as a convenient and compelling analogical source. This is facilitated by key commonalities between the two branches of international law – notably their shared commitment to the protection of human dignity.Footnote ²²⁴

Nevertheless, drawing analogies from outside IHL is not without risk. Over-reliance on analogical sources from outside this field may lead to the absorption of the logic of the source(s) into IHL, potentially eroding its distinctiveness.Footnote ²²⁵ Indeed, the transfer step implies the extension of a legal solution from the source to the target, yet the legal solution governing the source fits into a particular legal regime and is thus driven and fuelled by the logic of that regime. Therefore, it is crucial that analogizers exercise caution during the transfer step of analogical reasoning, carefully considering the differences between the source and the target as well as seriously adapting the solution to IHL’s own logic. For instance, the ICTY Trial Chamber considered in its Furundžija judgment that the definition of torture under IHL requires that “at least one of the persons involved in the torture process must be a public official or must at any rate act in a non-private capacity”.Footnote ²²⁶ However, this element of the definition was connected to IHRL logic, because that branch of international law imposes obligations on States in their relationship with individuals who are within their jurisdiction. Therefore, the ICTY Trial Chamber later made clear in the Kunarac judgment that, in opposition to IHRL, IHL creates obligations not only for States but also for individuals.Footnote ²²⁷ Observing this difference in logic, it concluded that “the presence of a [S]tate official or of any other authority-wielding person in the torture process is not necessary for the offence to be regarded as torture under [IHL]”.Footnote ²²⁸ This example illustrates that IHL’s external coherence should not be pursued at the expense of its own internal coherence.

In conclusion, this author readily concedes that complete coherence in IHL is likely “bound to remain a fantasy” as “the idea of unity on which it is built … is always [destined] to collapse”.Footnote ²²⁹ However, she maintains that the pursuit of coherence in this field is not futile – rather, it is a healthy and desirable endeavour. The sample of data presented in the previous section demonstrates that analogies have contributed to a balanced form of internal and external coherence. Analogies thus emerge as a powerful tool for promoting coherence in IHL.

However, three caveats must be noted. First, the relationship between analogies and coherence is likely bidirectional. While analogies may foster coherence in IHL (downstream), their widespread use may also reflect IHL stakeholders’ responsiveness to a broader “general pressure for cognitive coherence – that is, the tendency for people to be consistent in their reasoning”.Footnote ²³⁰ In other words, analogizers may reason by analogy in IHL not because they want or wish to create coherence, but because they feel compelled to think coherently in the first place (upstream). In this sense, coherence is both a cause and consequence of analogical reasoning. Second, analogies may promote coherence, but not necessarily truth or – to put it differently – objective legal solutions which perfectly match the reality on the ground. Analogical reasoning does not always guarantee accurate or definitive conclusions; rather, it offers a probabilistic perspective on reality – i.e., on what is going on the battlefield and on how this battlefield materiality is regulated by law.Footnote ²³¹ Analogical conclusions are probable, but not certain.Footnote ²³² Third, coherence should not be pursued as an absolute or exclusive goal which foolishly prevails over any other important goals within IHL.Footnote ²³³

Because analogies contribute to coherence in IHL, it may be argued that they introduce a conservative dynamic into this field of international law.Footnote ²³⁴ Admittedly, at first glance, IHL analogies appear to be a tool for legal continuity rather than change: they rely on existing sources rooted in the past and may slow the pace of IHL development. They also serve the goal of legislative economy.Footnote ²³⁵ Even when they lead to the creation of new legal texts rather than mere cross-referencing, “l[es] analogie[s] produi[sen]t moins de droit”: they generate fewer distinct rules and more similar norms.Footnote ²³⁶ Yet, despite their inherently conservative dimension through their reliance on sources, analogies do not necessarily confine IHL to the past. Their conservatism is only moderate; analogies also leave room for flexibility, allowing the law to evolve in response to new challenges.