Schlagwort: 1990s

“Asylum Compromise” Revisited

In the early 1990s, Andreas Zimmermann – then a PhD candidate at the Max Planck Institute for Comparative Public Law and International Law (MPIL) in Heidelberg – comprehensively addressed the question of the conformity of so-called “safe third country” (STC) practices in the field of refugee protection with international law. In 1992, in the run-up to the reform of the German asylum system, he wrote an expert opinion together with the institute’s director at the time, Jochen A. Frowein, for the German Federal Ministry of Justice titled “Der völkerrechtliche Rahmen für die Reform des deutschen Asylrechts” (“The International LawFramework for the Reform of German Asylum Law”)[1]. Following the adoption of Article 16a of the Basic Law – the German constitution –  in 1993, he wrote his PhD thesis titled “Das neue Grundrecht auf Asyl – Verfassungs- und völkerrechtliche Grenzen und Voraussetzungen” (“The New Fundamental Right to Asylum – Constitutional and International Law Limits and Requirements”).[2] The work was published in 1994 as part of the Heidelberg Max Planck Law book series “Contributions on Comparative Public Law and International Law”. It analyses in detail the legality of Article 16a of the Basic Law at the constitutional, European, and international level. Furthermore, in 1993, he also wrote an article on “Asylum Law in the Federal Republic of Germany in the Context of International Law”, which was published in the Heidelberg Journal on Comparative Public Law and International Law (HJIL).[3] In light of the current discourse surrounding the proposals for a reform of the Common European Asylum System (CEAS), it is worthwhile revisiting these works.

The “Asylum Compromise” of 1993

In 1993, the concept of STC was introduced to the German constitution in an effort to deal with the high influx of war refugees from the former Yugoslavia. Germany thereby followed an emerging practice that originated in Switzerland to ensure a sharing of responsibility for asylum seekers  among members of the international community. States most impacted by refugee influx could reject an application for asylum as inadmissible and could subsequently send asylum seekers to another state on the condition that the receiving state could offer them adequate protection in accordance with accepted international standards.

Through the so-called “asylum compromise”, the right to asylum was severely restricted by means of changing Article 16 of the Basic Law. The previous version, formerly enshrined in Article 16(2) of the Basic Law read:

“Persons persecuted on political grounds shall have the right of asylum.”

Paragraph (2) of Article 16a of the Basic Law (introduced in 1993) stipulates:

“Paragraph (1) [the right to asylum] of this Article may not be invoked by a person who enters the federal territory from a member state of the European Communities or from another third state in which application of the Convention Relating to the Status of Refugees and of the Convention for the Protection of Human Rights and Fundamental Freedoms is assured. […]”[4]

Safe Third Country Practices and International Law

Regarding the compliance of the STC concept enshrined in Paragraph (2) of Article 16a of the Basic Law with international law, Zimmermann’s main conclusions,[5] in 1993, are as follows:

Under customary international law, states are not subject to a duty to readmit persons other than their own nationals. Hence, sending an asylum seeker to a third state is only possible if that state has declared its willingness to readmit asylum seekers and provide them with the option to request refugee status in a legally binding agreement with the sending state.

Under the 1951 Geneva Refugee Convention (thereinafter: Refugee Convention), states are generally not precluded from returning asylum seekers to third states. In any given case, however, it must be assured that the principle of non-refoulement, enshrined in Article 33 of the Refugee Convention, is abided by. This means that asylum seekers must be protected from persecution in the third state as well as from deportation by the third state to the presumed country of persecution (protection from chain deportations).

While STC can be determined in a generalized manner, it is necessary that the states in question are themselves bound by the Refugee Convention without geographical reservations and that their procedure of asylum application is in line with the procedural minimum standard under international law. However, even if the procedural guarantees of the Refugee Convention are generally complied with, the returning state is obliged to examine whether the receiving state fulfils its obligations under Article 33 of the Refugee Convention bona fide. In case of concrete indications that the respective third country is, in practice, not complying with the prohibition of refoulement stipulated in Article 33 of the Refugee Convention despite its formal commitment to the Convention, the returning state is obliged to review whether a country may remain on the list of STC.

Furthermore, before being sent back to a STC, a refugee must be granted an opportunity to claim that, in his or her individual case, the third state in question would not be safe.

Under the Refugee Convention, an asylum-seeker needs to be sufficiently connected to the third country in question for his or her transfer there to be legal. For example, a state has no right to return a person to a country through the airport of which he or she has passed for purposes of transit only.

Whereas the returning state must ensure that the third state abides by Article 33 of the Refugee Convention, it is not responsible for ensuring that allprovisions of the Refugee Convention are complied with in the third country.[6] Nevertheless, the third state must guarantee the refugee some kind of minimum living standard.

Under the European Convention on Human Rights (ECHR), there are further special restrictions, e.g., with regard to the protection of family life (Article 8 ECHR) and the prohibition (under Article 3 ECHR) to create a situation where a person is permanently sent from one state to another and thus becomes a “refugee in orbit”.

Most of these findings about the legal safeguards still apply to today’s changed legal environment and have been further developed by scholarship and jurisprudence over the years. Some questions, however, remain disputed to this day: this includes whether the third state needs to be a party to the Refugee Convention and to what extent the sending state must ensure that provisions of the Refugee Convention are complied with in the third country.[7]

The Safe Third Country Concept in EU Asylum Law

In 2005, the concept of STC was incorporated at the EU-level in the 2005 Asylum Procedures Directive, which was later replaced by the 2013 Asylum Procedures Directive (rAPD). Article 38(1) rAPD stipulates that a member state may apply the STC concept only when the competent authorities are satisfied that a person seeking international protection will be treated in accordance with the following principles in the third country:

„(a) life and liberty are not threatened on account of race, religion, nationality, membership of a particular social group or political opinion;

(b) there is no risk of serious harm as defined in Directive 2011/95/EU;

(c) the principle of non-refoulement in accordance with the Geneva Convention is respected;

(d) the prohibition of removal, in violation of the right to freedom from torture and cruel, inhuman or degrading treatment as laid down in international law, is respected; and

(e) the possibility exists to request refugee status and, if found to be a refugee, to receive protection in accordance with the Geneva Convention.”

Furthermore, according to Article 38(2)(a) rAPD, the applicant needs to be sufficiently connected to the STC, so that his or her return and seeking refuge there may be considered reasonable. If the conditions of Article 38 rAPD are met, a member state may consider an application for international protection to be inadmissible (Article 33(2)(c) rAPD).

Increasingly, restrictive immigration policies are aimed at deterring irregular arrivals and responsibility-shifting rather than responsibility-sharing. States are seeking to outsource migration management and international protection responsibility to third countries – mostly transit countries – already impacted by large refugee flows. These policies come with diminished safeguards, as demonstrated by the 2016 EU‑Turkey statement or the 2022 UK‑Rwanda agreement, the latter of which was recently declared unlawful by the UK Supreme Court for violating Article 3 ECHR.

While Article 16a of the Basic Law has lost its significance in German asylum law practice since the main regulations are now contained in international treaty law and EU law, the underlying question of the conformity of STC practices with international law remains a topical one. In June 2023, the European Council agreed on a negotiation position on the new Asylum Procedures Regulation. Under the latest[8] amended proposal for an Asylum Procedure Regulation (draft APR), the use of STC procedures shall be expanded by watering down and settling the legal safeguards of the concept on a low standard:

Article 43a(2) of the draft APR states that STC  need only provide “effective protection” to refugees but they are not obligated to grant them legal status, ensure full access to healthcare, or guarantee family unity. Furthermore, the concept of “de facto-protection”, on which Article 43a(2) draft APR is based, does not comply with international refugee law. However, whether the STC must be a party to the Refugee Convention (without geographical limitation) remains, as mentioned, controversial.[9] For instance, a STC could refer to a definition of refugee different than the one laid down in the Refugee Convention, as Zimmermann argues in his dissertation. Furthermore, as a non‐contracting state, a STC could claim to not be legally bound by the principle of non‐refoulement. Lastly, there would be no possibility for the UNHCR to intervene in cases of obvious breaches of the principle of non‑refoulement. Accordingly, at least in Zimmermann’s opinion, the STC must be a party to the Refugee Convention (as well as to other human rights treaties relevant to asylum).[10] With the concept of non-refoulement becoming customary in international law and, arguably, jus cogens, the argument Zimmermann makes may have lost some of its traction. However, an expansion of the circle of STC in the sense that it is no longer required for a third country to respect the principle of non-refoulement “in accordance with the Geneva Convention”[11] would be contradictory to Article 78(1) TFEU and Article 18 of the EU-Charter which require, respectively, that the EU’s asylum policy be “in accordance” with the Refugee Convention and that “the right to asylum shall be guaranteed with due respect for the rules of the Geneva Convention”.

According to Article 45(1)(a) of the draft APR, the safety of a country is to be assessed with reference to “non-nationals”. STC could therefore include countries persecuting their own citizens and producing refugees themselves. Additionally, a third country may be categorized as a STC “with the exception of certain parts of its territory or clearly identifiable groups of persons”, Article 45(1a) draft APR. This could lead to the relocation of asylum seekers to an unstable third state, where a protection zone equivalent to the size of a refugee camp is effectively managed and asylum seekers are held there with their subsistence ensured. This raises unresolved questions regarding the EU’s responsibility for ensuring adequate living standards in third countries as well as the inclusion of asylum seekers in line with the Refugee Convention (cf. Zimmermann’s findings above).

Fortunately, the “reasonable connection requirement” is to be upheld and its definition is subject to rules under national law. This prevents EU member states from concluding a UK‑Rwanda‑like agreement and sending refugees to faraway countries with which they lack any connection. Under Article 45(2b)(b), Recital 37 of the draft APR, a reasonable connection may be established when the applicant has settled or stayed in the third country. Proposals that included references to “transit” as conclusive evidence of a connection were unsuccessful. This is in line with the case law of the Court of Justice of the European Union (CJEU).[12] The examples provided of “settlement” and “stay” nonetheless give member states some room for interpretation in this direction. Having said that, the “reasonable connection requirement” as laid out by Zimmermann in 1994 should inform these interpretations.

Member states may presume the safety of a third country on the mere basis of an agreement between the EU and said third country as well as general assurances by that country that readmitted migrants will be “protected in accordance with the relevant international standards”, Article 45(3) draft APR. The EU-Turkey deal, however, is living proof that the mere existence of an agreement does not guarantee its safety in practice. This is why the 2007 Michigan Guidelines on Protection Elsewhere require, for permitting the referral of an asylum seeker to “protection elsewhere”, a “good faith empirical assessment”[13] by the sending state that refugees will enjoy Refugee Convention rights in the receiving state.[14] The burden of proof in this respect therefore does not lie with the asylum seeker but with the country where the asylum application was lodged, as it retains the responsibility for any action in violation of its obligations under international law, particularly the principle of non-refoulement (cf. Zimmermann’s findings above).

Conclusion and Outlook

After three years of negotiations, the European Parliament and the Council reached a political agreement on the key proposals of the Pact on 20 December 2023, including the Asylum Procedures Directive. This opened the door for further negotiations regarding technical details with a formal adoption expected before the European Parliament elections in June 2024.

The latest proposal for the APR is an example of an attempt at lowering standards with regard to the concept of STC and contradicts the EU’s endorsement of a positive contribution to the protection and promotion of human rights.[15] Once it is formally adopted (changes by means of the political agreement reached in December 2023 and further political discussions notwithstanding), refugee protection in the EU will largely depend on how member states interpret and enforce the regulation. It will be important for member states to consider the real circumstances in the third countries in question. Furthermore, due consideration should be taken of the principles of international cooperation and responsibility-sharing as expressions of international solidarity, a concept which is enshrined in the preamble of the Refugee Convention. Member states should conduct negotiations with all third countries along the migration routes and the EU should make more attractive offers of cooperation with regard to migration policy towards the relevant third countries. To this end, it should also utilize development policy instruments, for example supporting these countries in strengthening their asylum and migration policy capacities.

30 years after the German “asylum compromise”, migration remains at the forefront of the political debate. Concomitant with this is the desire of states to send asylum seekers and, by extension, responsibility to third countries. The debate about international human rights and refugee protection standards in this regard must likewise continue and it is worth bringing to attention some “old” arguments such as those put forward by Andreas Zimmermann and Jochen Frowein at the MPIL.

[1] Jochen A. Frowein, Andreas Zimmermann, Der völkerrechtliche Rahmen für die Reform des deutschen Asylrechts: Gutachten im Auftrag des Bundesministeriums der Justiz erstattet vom Max-Planck-Institut für ausländisches öffentliches Recht und Völkerrecht, Heidelberg, Bundesanzeiger 45 (1993).

[2] Andreas Zimmermann, Das neue Grundrecht auf Asyl – Verfassungs- und völkerrechtliche Grenzen und Voraussetzungen, Contributions on Comparative Public Law and International Law vol. 115, Heidelberg: Springer 1994.

[3] Andreas Zimmermann, ‘Asylum Law in the Federal Republic of Germany in the Context of International Law’, HJIL 53 (1993), 49-87.

[4] Translation: Christian Tomuschat et al., in cooperation with the Language Service of the German Bundestag, Basic Law for the Federal Republic of Germany in the revised version published in the Federal Law Gazette Part III, classification number 100-1, as last amended by the Act of 19 December 2022 (Federal Law Gazette I p. 2478).

[5] Compare: Zimmermann (fn. 2), 400-401.

[6] This is to be read against the background that, in Andreas Zimmermann’s opinion, it is required under international law that a STC is a party to the Refugee Convention.

[7] Rainer Hofmann and Tillmann Löhr, ‘Introduction to Chapter V: Requirements for Refugee Determination Procedures’ in: Andreas Zimmermann, Felix Machts and Jonas Dörschner (eds), The 1951 Convention Relating to the Status of Refugees and its 1967 Protocol: A Commentary, Oxford: Oxford University Press 2011, 1112-1113.

[8] Any changes that were agreed upon as part of the political agreement reached in December 2023 are not expected to be available on paper until February 2014.

[9] Hofmann and Löhr (fn. 7), 1112, para 79.

[10] Compare: Zimmermann (fn. 2), 174f.

[11] Article 38(1)(c) rAPD.

[12] See CJEU, LH v. Bevándorlási és Menekültügyi Hivatal, judgement of 19 March 2020, case no. C-564/18, ECLI:EU:C:2020:218; CJEU, FMS, FNZ, SA, SA junior v. Országos Idegenrendészeti Főigazgatóság Dél-alföldi Regionális Igazgatóság, Országos Idegenrendészeti Főigazgatóság, judgement of 14 May 2020 in joint cases no. C-924/19 PPU and C-925/19 PPU, ECLI:EU:C:2020:367; CJEU, European Commission v. Hungary, judgement of 16 November 2021, case no. C-821/19, ECLI:EU:C:2021:930.

[13] Emphasis added.

[14] See also: UN High Commissioner for Refugees (UNHCR), Guidance Note on bilateral and/or multilateral transfer arrangements of asylum-seekers, May 2013, available at: <>  (last accessed: 18 January 2024), para 3(iii).

Suggested Citation:

Laura Kraft: “Asylum Compromise” Revisited,, DOI: 10.17176/20240219-184505-0

Lizenz: CC BY-NC-SA 4.0 DEED


Fontes Juris Gentium. Ein völkerrechtliches Editionsprojekt

Fontes Iuris Gentium. An editorial project on international law


Einleitung. Zur Schaffung eines „Sinnsystems“

Vor etwa 100 Jahren war es wohl allenfalls Utopisten vorstellbar, dass riesige Quellen- und Textsammlungen digital verfügbar sein könnten. Zu jener Zeit waren Textsammlungen dessen, was man als Völkerrecht kannte, äußerst rar und ihre Anpassung an Rechtsentwicklungen allein schon aus technischen Gründen zeit- und natürlich auch kostenaufwendig. Die bedeutendste Sammlung völkerrechtlicher Quellen war seinerzeit der von Martens herausgegebene Recueil völkerrechtlicher Verträge, dessen erster Band bereits 1791 erschien.[1] Diese Publikation wird zurecht als die bedeutendste Grundlage des Völkerrechts und Martens daher als „Vater des positiven Völkerrechts“ angesehen.  Vor diesem Hintergrund war die Idee, weitere völkerrechtliche Quellen zu sammeln und zu publizieren, folgerichtig und es war naheliegend, dass dabei der Blick auf die völkerrechtliche Praxis gelenkt wurde. Die verlässlichste Aussagekraft über diese bot die internationale Gerichtsbarkeit, was zu dem Projekt führte, den völkerrechtlichen Gehalt aus Entscheidungen internationaler Rechtsprechungsorgane kontinuierlich zusammenzustellen und zu veröffentlichen. Der Urheber des Projekts, Viktor Bruns, Direktor des Kaiser-Wilhelm-Instituts für ausländisches öffentliches Recht und Völkerrecht (später Max-Planck-Institut, MPIL), umschrieb die Grundidee folgendermaßen:

„Es ist … Aufgabe der Wissenschaft, die Grundlagen und Grundprinzipien des Völkerrechts als einer Rechtsordnung zu ermitteln und ein System aufzustellen, das aus der Beobachtung der internationalen Praxis gewonnen ist und das ein Sinnsystem ist, das dem Charakter einer Rechtsordnung und im speziellen dem einer Ordnung für das Zusammenleben und den Verkehr der Staaten entspricht. So muß die Theorie des Völkerrechts aus der Erfahrung in der Praxis gewonnen, und so muß die Praxis an den so gewonnenen Erkenntnissen geprüft werden. Das ist nur möglich auf Grund eines umfassenden Überblicks über die internationale Praxis. Diesen Überblick zu verschaffen, ist Zweck der Herausgabe der Fontes Juris Gentium.“[2]

Da bekanntlich Völkerrecht nicht nur Grundlage der Entscheidungen vor internationalen Gerichten ist, sondern auch nationale Gerichte bei ihrer Rechtsfindung Völkerrecht anzuwenden haben, war das Projekt Fontes Iuris Gentium nicht nur darauf angelegt, die Rechtsprechung internationaler Gerichte zu bearbeiten, sondern auch die Positionen nationaler Gerichte zu völkerrechtlichen Fragen aufzuarbeiten. Entsprechend wurden in der Serie der Fontes zur   Gerichtsbarkeit (Series A) zwei Abteilungen vorgesehen: die erste Abteilung (Fontes Iuris Gentium, Series A, Sectio 1) war der internationalen Gerichtsbarkeit gewidmet, das heißt dem Ständigen Schiedshof und dem Ständigen Internationalen Gerichtshof (StIGH) und dessen Nachfolger, dem Internationalen Gerichtshof (IGH). Die zweite Abteilung dieser ersten Serie (Fontes Iuris Gentium, Series A, Sectio 2) galt der Erörterung völkerrechtlicher Fragen in den Entscheidungen der obersten Gerichte der wichtigsten Staaten. Zusätzlich zur Serie über die internationale Gerichtsbarkeit wurde eine zweite Serie angelegt, die die politischen und rechtlichen Grundsätze aus dem Notenwechsel der wichtigsten Staaten enthalten sollte (Series B). Geplant, aber niemals ausgeführt wurden zudem eine dritte Serie zu Gutachten und Entscheidungen internationaler Organe, die nicht den Charakter internationaler Gerichte haben (Series C), und schließlich eine vierte Serie (Series D), in der die Vertragsklauseln der wichtigsten, seit Beginn des 19. Jahrhunderts abgeschlossenen Verträge enthalten sein sollten.[3] Dass dieses höchst ehrgeizige Vorhaben nur in langfristiger Zusammenarbeit mehrerer Wissenschaftler durchgeführt werden konnte, war offensichtlich. Daher war es ohne Frage bei einer Forschungseinrichtung wie dem MPILin den richtigen Händen und wohl auch nur hier zu bewältigen.

System und Aufbau der Fontes  

Meine persönliche Erfahrung mit der Arbeit an den Fontes betrifft nur die Serie A, Sectio 1, also die Bearbeitung der Entscheidungen des IGH, beginnend mit der Publikation von Band 6, der die Entscheidungen von 1959-1975 enthält, bis zur Beendigung des Projekts im Jahr 2005.

Als ich 1970 als wissenschaftliche Referentin am MPIL angestellt wurde, war dies in zweierlei Hinsicht etwas Besonderes, was mir allerdings erst später bewusst wurde. Zum einen war damals schon die Tatsache, dass Frauen im wissenschaftlichen Bereich eingestellt wurden, außergewöhnlich und kann als „Pilotprojekt“ für die dann später zunehmende Einstellung von Kolleginnen angesehen werden. Zum anderen, und wesentlich bedeutender, war die Tatsache, dass meine vita nicht den Lebensläufen bisheriger Mitarbeiter entsprach. Ich hatte nämlich in einem ersten Studium den Weg als Diplomdolmetscherin eingeschlagen und dann bei der EWG mit seinerzeit noch nur sechs Mitgliedstaaten als Simultandolmetscherin gearbeitet. Aber diese wenig kreative Tätigkeit erschien mir auf Dauer nicht befriedigend und so studierte ich dann Rechtswissenschaften an der Universität Heidelberg. Die Kombination aus Sprachkompetenz und erfolgreichem Jurastudium war dann offensichtlich ausschlaggebend für meine Einstellung –insbesondere mit Blick auf das Projekt Fontes, das aufgrund der zunehmenden Zahl von Fällen, die vor den IGH gebracht wurden, intensiveren Einsatz forderte. Erfreulicherweise blieb jedoch trotzdem noch Raum für die Fertigstellung einer Dissertation.[4] Neben den üblichen Aufgaben als wissenschaftlicher Referent am Institut blieb ich bis zur Einstellung 2005 „die Konstante“ des Fontes‑Teams, dessen Leitung ich 1985 übernommen hatte, mit Beginn der Fortführung des  Projekts in der von Professor Rudolf Bernhardt initiierten Form des World Court Digest. Die weiteren drei bis vier Mitglieder der Arbeitsgruppe waren jeweils nur vorübergehend eingebunden.

Wie schon erwähnt, begann meine Tätigkeit an den Fontes mit den beiden Teilbänden von Band 6 der Fontes Series A, Sectio 1. Diese waren (zusammen mit Band 7, der die Dekade von 1976 bis 1985 abdeckt) die letzten Bände, die noch unter dem ursprünglichen Titel und in der ursprünglichen Form der Fontes Iuris Gentium erschienen. Die Dokumentation der späteren Rechtsprechung des IGH wurde dann unter dem Titel World Court Digest fortgeführt, allerdings mit einigen wesentlichen Änderungen.

Aber zunächst einmal zum grundlegenden System der Fontes-Serie. Wie der Titel bereits sagt, handelt es sich um ein Quellenwerk. Das heißt, dass nur Exzerpte aus den Originaltexten der Entscheidungen des Gerichts wiedergegeben wurden. Das schließt natürlich auch Gutachten ein. Da die Entscheidungen jeweils in Englisch und Französisch abgefasst sind, wurden jeweils beide Sprachfassungen aufgenommen, auf der linken Buchseite die französische Fassung, auf der rechten die englische. Exzerpte des authentischen Texts der Entscheidung wurden mit einem Stern (*) gekennzeichnet. Auch Erklärungen und Sondervoten, die den Entscheidungen angefügt sind, wurden ausgewertet und zunächst in kleinerem Druck einbezogen, auch hier wurde jeweils die authentische Sprache kenntlich gemacht. Später, bei der Umstellung auf den World Court Digest, wurde allerdings die Einbeziehung der Sondervoten und Erklärungen zunächst grundsätzlich in Frage gestellt. Die Aufnahme dieser Texte wurde aber dann – glücklicherweise – doch beibehalten, da sie zum Diskussionsstand im Gericht und insbesondere auch zur Entwicklung der Rechtsauffassung des Gerichts außerordentlich informativ sind.


Die Systematik aller Bände folgt einem einheitlichen Aufbau: Der erste Teil der jeweiligen Bände gilt dem materiellen Recht, das in gleichbleibender Struktur von den Grundlagen und Grundfragen des Völkerrechts über spezielle Rechtsgebiete wie Vertragsrecht, Staatshaftungsrecht, und internationale Organe bis hin zu Krieg und Neutralität reicht. Gerade in diesem Teil hat sich natürlich die Bandbreite der Sachgebiete über die Jahre ganz wesentlich erweitert, da das Völkerrecht immer mehr Bereiche abdeckt, wie zum Beispiel Seerecht, Luft- und Weltraumrecht, Menschenrechte, internationales Wirtschaftsrecht, internationales Umweltrecht, internationales Strafrecht, um nur einige der neueren Aspekte zu nennen. Der zweite Teil eines Bandes gilt jeweils der Struktur der internationalen Gerichtsbarkeit, also dem StIGH und dem KaIGH und hat sich sachlich nicht erweitert. Er behandelt neben den Grundlagen der internationalen Gerichtsbarkeit die Bereiche Zuständigkeit, Verfahren, Urteile und Gutachten.

Wie bereits betont, trug die Publikation ursprünglich den Titel Fontes Iuris Gentium und den Untertitel Handbuch der Entscheidungen des (Ständigen) Internationalen Gerichtshofs. Dieser präzisierende Zusatz war durchaus sinnvoll, denn die Entscheidungen internationaler Gerichte sind als solche nicht „Quellen“ des Völkerrechts, da Gerichte bekanntlich nicht Recht setzen, sondern Recht sprechen. Allerdings sind die Grenzen hier oft fließend, und gerade im Völkerrecht gab es vor 100 Jahren nicht nur „weniger Recht“, sondern vor allem: weniger geschriebenes, kodifiziertes Recht. Aufgabe und Wert der Fontes lag insbesondere darin, genau herauszuarbeiten und zu dokumentieren, was das internationale Gericht als Teil des positiven Völkerrechts festgestellt hatte. Diese Zielsetzung musste bei der konkreten Arbeit immer wieder in Erinnerung gerufen werden, denn bisweilen war es schon sehr verlockend, ein Exzerpt kurz zu halten und damit die Aussage des Gerichts konkreter und definitiver zu fassen, als sie tatsächlich sein sollte. Doch der wissenschaftliche Geist und die Verantwortung des Teams und seines langjährigen Leiters Rudolf Bernhardt führten – wie wir fanden – immer dazu, die „Quelle“ nicht anzutasten und Aussagen so umfänglich wie erforderlich zu reproduzieren, um ihren tatsächlichen Gehalt getreu widerzuspiegeln. Dieses Bestreben führte auch dazu, dass beim Übergang zum World Court Digest längere Exzerpte aufgenommen wurden, um Ausgangspunkt, Parteivorbringen und Bewertung des Gerichts genauer abzubilden. Angesichts der Tatsache, dass die Zahl der Fälle und der Umfang der Entscheidungen im Laufe der Zeit stark zugenommen hatten – was eine erfreuliche Akzeptanz der internationalen Gerichtsbarkeit dokumentiert – wurden damit jedoch auch Einschnitte an anderer Stelle erforderlich. Konkret hatte dies zur Folge, dass die Zweisprachigkeit aufgegeben wurde und nur noch die englischen Texte der Entscheidungen ausgewertet und wiedergegeben wurden – unabhängig davon, ob sie die authentische Textfassung waren.  Neu eingeführt wurde eine Zusammenfassung der Fälle in englischer Sprache, was bei den oft recht komplizierten Grundlagen der Streitigkeiten und auch Gutachtenfragen sehr hilfreich war. Zudem wurde auch eine Liste der Richter und ad hoc Richter des behandelten Zeitraums erstellt, so dass die bearbeitete Periode im Gesamtüberblick leicht zu überschauen war.

Editorische Arbeit im vordigitalen Zeitalter: Mit Schere, Papier und Klebstoff

Wie Viktor Bruns schon im Vorwort zum 1. Band der Serie A, Sectio 1, erwähnt hatte, konnte nur ein Team von Wissenschaftlern, in der Regel vier bis fünf Kollegen, die Arbeit an den Fontes leisten. Das erforderte zunächst einmal vor allem die intensive Lektüre der Entscheidungen und dann, in nicht minder intensiven Diskussionen, die Entscheidung darüber, welche Passagen an welcher Stelle der Systematik wiedergegeben werden sollten. Hierbei war eine gewisse Kontinuität zumindest eines Teils des Teams hilfreich, so dass Erfahrung mit der Kultur der Fontes und neuer Input durch neue Mitarbeiter gleichermaßen permanent gesichert waren. Für neue Teamkollegen, vor allem jüngere Mitarbeiter, bot dies Gelegenheit, sich intensiv mit den Entscheidungen des „Weltgerichtshofs“ zu befassen, die bis heute ein wesentlicher Spiegel des Standes des Völkerrechts und seiner Entwicklung sind. Kritische Auseinandersetzung mit dem Völkerrecht konnte auf diese Weise hervorragend vermittelt und eingeübt werden.

Dorothee Bender (1970er) (Foto MPIL)

Neben dem „wissenschaftlichen Team“, dem die Auswahl der Exzerpte und ihrer Position in der Systematik oblag, hing die Herstellung der einzelnen Bände natürlich ganz wesentlich von dem „technischen Team“ ab, das in der Regel aus nur zwei oder drei Personen bestand. Die Herstellung des Manuskripts im „vordigitalen Zeitalter“ mag heute geradezu abenteuerlich erscheinen. Sie verdient an dieser Stelle ausdrückliche Würdigung, denn ohne die akribische Mitarbeit der Sekretariate wäre wohl kaum je ein Band erschienen. Natürlich können hier nicht alle einzelnen Mitarbeiterinnen genannt werden, die sich da höchst verdienstvoll betätigt haben, aber stellvertretend soll zumindest Dorothee Bender namentlich erwähnt werden, die über Jahrzehnte der nicht-digitalen Welt diese Arbeit in der ihr eigenen Perfektion hauptverantwortlich bewältigt hat. Und das lief folgendermaßen ab: Wenn das Team sich auf ein Exzerpt und dessen Einordnung in die Systematik geeinigt hatte, wurde dies im Text der Entscheidung kenntlich gemacht und seine Zuordnung vermerkt. Händisch musste dann dieses Exzerpt mit einer Schere aus dem Text ausgeschnitten und auf ein Blatt Papier geklebt werden; die entsprechende Rubrik des Exzerpts sowie die Angabe der Seiten im Urtext waren dann manuell und gut leserlich einzufügen. War ein Exzerpt für mehrere Rubriken vorgesehen, so musste es mehrfach abgetippt werden und den gleichen Weg gehen. Die Entscheidungen des Gerichts sahen nach diesem Prozedere sehr löchrig aus.  Und das Endmanuskript der Fontes bestand schließlich aus den gesammelten Seiten mit den aufgeklebten Exzerpten. Für die beiden Teile von Band 6 zum Beispiel – insgesamt über 830 Seiten – war das ein eindrucksvolles Konvolut, das natürlich noch ungebunden war und daher sorgfältig gehütet werden musste, damit kein einziges Blatt verloren ging oder falsch einsortiert wurde. Dieser zugegeben aufwendigen, aber verlässlichen Art der Manuskriptherstellung haben wir schmerzlich nachgetrauert, als das erste im Institut am Computer verfasste Manuskript, der World Court Digest 1986-1990, plötzlich auf mysteriöse Weise „verschwunden“ war, kurz bevor es zum Verlag gehen sollte –  und auch nur teilweise wieder hergestellt werden konnte. Allerdings blieb dieses Missgeschick einmalig – und die Herstellungsweise konnte insgesamt den technischen Fortschritten angepasst werden, die es bald auch ermöglichten, eine digitale Online-Version zur Verfügung zu stellen.[5]


Der durch die digitale Entwicklung erleichterte Zugriff auf die Originaltexte der Entscheidungen des IGH ist fraglos sehr zu begrüßen, kann meines Erachtens aber die Fontes Iuris Gentium bzw. den World Court Digest nicht völlig ersetzen. Denn diese Publikationen machten es möglich, die Entwicklungen des Völkerrechts zu bestimmten Sachbereichen, so wie sie in den Entscheidungen des IGH zum Ausdruck kamen, kontinuierlich über Jahrzehnte zu verfolgen. Gerade in den neueren Bereichen des Völkerrechts ist ein solcher „historischer“ Überblick außerordentlich lohnend und hilfreich. Und nebenbei sei bemerkt, dass die Bearbeiter dabei vielfach feststellen konnten, dass ihre Versuchung, Exzerpte kürzer zu fassen, und damit die völkerrechtliche Aussage kategorischer erscheinen zu lassen, als sie seinerzeit wirklich war, der tatsächlichen Entwicklung des Rechts oft nur vorgegriffen hätte. Die mit den Fontes eröffnete Möglichkeit, die Entwicklung des Völkerrechts in einzelnen Bereichen minutiös nachzuverfolgen, reicht nur bis 2005. Danach wurde das gesamte Projekt eingestellt, was, wie alle derartigen Entscheidungen, zwar nachvollziehbar ist, aber eben auch bedauerliche Nebenwirkungen hat.

[1] Georg Friedrich von Martens, Recueil des principaux traités d’alliance, de paix, de trêve, de neutralité, de commerce, de limites, d’échange etc. conclus par les puissances de l’Europe tant d’entre elles qu’avec du monde depuis 1761 jusqu’à présent, Bd. 1, Göttingen: Jean Chretien Dieterich 1791; Ab1819 wurde das Werk von unterschiedlichen Herausgebern bis 1944 fortgeführt unter dem Titel „Recueil Martens“.

[2] Viktor Bruns, Vorwort zu Band 1, Series A, Sectio 1, S. XI.

[3] Erschienen sind in Serie A, Sectio 1 Band 1 (1931), 3 (1935) und 4 (1964) zum Ständigen Internationalen Gerichtshof; Band 2 (1931) zum Ständigen Schiedshof und zum Internationalen Gerichtshof; die Bände 5 (1961), 6 (1978) und 7 (1990) in der „alten“ Fassung unter dem Titel Fontes Iuris Gentium und beginnend mit der Rechtsprechung ab 1986 unter dem Titel World Court Digest die Bände 1-4, die die Rechtsprechung bis einschließlich 2005 umfassen. In der Serie A, Sectio 2, sind in Band 1 (1931) die Entscheidungen des Deutschen Reichsgerichts von 1879 – 1929 bearbeitet worden; in Band 2 (1960) die Entscheidungen  des Reichsgerichts unter Einbeziehung der Staatsgerichtshofs für das Deutsche Reich, 1929-1945; Band 3 bis Band 9 betreffen dann die deutsche höchstrichterliche Rechtsprechung in völkerrechtlichen Fragen bis einschließlich 1985. Die Serie B Sectio 1 umfasst in Band 1 die diplomatische Korrespondenz von 1856 ‑1878 und in Band 2 die diplomatische Korrespondenz von 1871‑1878.

[4] Karin Oellers-Frahm, Die einstweilige Anordnung in der internationalen Gerichtsbarkeit, Beiträge zum ausländischen öffentlichen Recht und Völkerrecht Bd. 66, Berlin: Springer 1975.

[5] Diese ist zu finden auf: <>.


Introduction. On the establishment of a “material system”

Some 100 years ago, only utopians could have imagined that huge collections of legal sources and texts would become available digitally. At that time, collections of texts on what was known as international law were extremely rare and their adaptation to legal developments was – for technical reasons, to begin with – time-consuming and, of course, costly. The most important collection of international law sources at the time was the Recueil of international treaties edited by von Martens, the first volume of which was published as early as 1791.[1]  This publication is rightly regarded as the most important basis of international law and Martens is therefore considered the “father of positive international law”. Against this background, the idea of collecting and publishing further sources of international law was logical, and it was only natural that the focus was to be put on the practice of international law. Here, international jurisdiction offered the most reliable information, which led to the project of continuously compiling and publishing excerpts relevant to the determination of international law from the decisions of international judicial bodies. The initiator of the project, Viktor Bruns, Director of the Kaiser Wilhelm Institute for Comparative Public Law and International Law (later to become the Max Planck Institute for Comparative Public Law and International Law, MPIL), described the main concept as follows:

“It is […] the task of science to ascertain the bases and basic principles of international law as a legal order and to draw up a system, informed by the observation of international practice and which is a material system, corresponding to the character of a legal order, specifically an order for the coexistence and interaction of states. Thus, the theory of international law must be extracted from experience in its practice and thus practice must be evaluated on the basis of the findings made in this manner. This is only possible based on a comprehensive overview of international practice. To create this overview is the purpose of the publication of the Fontes Juris Gentium.[2]

As international law is not only the legal basis of decisions of international courts, but national courts also have to apply international law, the Fontes Iuris Gentium project was designed not only to analyse the case law of international courts, but also to examine the positions taken by national courts on questions concerning international law. Accordingly, the Fontes series on jurisdiction (Series A) was divided into two sections: the first section (Fontes Iuris Gentium, Series A, Sectio 1) was devoted to international jurisdiction, i.e. the Permanent Court of Arbitration (PCA) and the Permanent Court of International Justice (PCIJ) as well as its successor, the International Court of Justice (ICJ). The second section of this first series (Fontes Iuris Gentium, Series A, Sectio 2) dealt with questions of international law in decisions of the highest courts of the most important states. In addition to the series on international jurisdiction, a second series was created, containing the political and legal principles derived from the exchange of diplomatic notes between the most important states (Series B). Planned, but never realised, were a third series on opinions and decisions of international bodies that do not have the character of international courts (Series C), and a fourth series (Series D), which was to contain the treaty clauses of the most important treaties concluded since the beginning of the 19th century.[3]  That this highly ambitious project could only be realised through the long‑term collaboration of several scholars was obvious. It was therefore evidently in the right hands at a research institution such as the MPIL and could arguably only be accomplished here.

System and structure of the Fontes 

My personal experience with the work on the Fontes only relates to Series A, Sectio 1, i.e. the processing of the decisions of the ICJ, starting with the publication of Volume 6, which contains the decisions from 1959-1975, until the termination of the project in 2005.

When I came to the MPIL as a research fellow in 1970, it was a special situation in two regards, although I only realised that later. Firstly, the employment of women in the scientific field as such was unusual at the time and can be seen as a “pilot project” for the later increase in the recruitment of female colleagues. Secondly, and much more importantly, my curriculum vitae differed from that of other research fellows. After my first degree I had initially chosen the career of a certified interpreter and had worked as a simultaneous translator at the EEC, which at the time had only six member states. This rather uncreative occupation did not strike me as satisfying in the long term, however, and so I went on to study law at the University of Heidelberg. The combination of language skills and a law degree was obviously the decisive factor in my recruitment – especially with regard to the Fontes project, which, due to the increasing number of cases brought before the ICJ, required more intensive work. Fortunately, however, there was still time to complete a doctoral thesis.[4]  In addition to my regular duties as a research fellow at the Institute, I remained “the constant” of the Fontes team, the leadership of which I took over in 1985, when, on initiative of Rudolf Bernhardt, the project was continued in the form of the World Court Digest. The other three to four members of the working group were each only involved on a temporary basis.

As already mentioned, my work on the Fontes began with the two parts of volume 6 of the Fontes Series A, Sectio 1, which (together with Volume 7, covering the ten-year period from 1976 to 1985) were the last volumes to appear under the original title and in the original form of the Fontes Iuris Gentium. The documentation of the ICJ’s later case law was then continued under the title World Court Digest, albeit with some substantial changes.

But, first of all, to the general structure of the Fontes series: As the title suggests, it is a collection of sources. This means that only excerpts from the original texts of the court’s decisions have been reproduced. This of course also includes advisory opinions. As the decisions are each formulated in English and French, both language versions were included, the French version on the left side of the book and the English version on the right side. Excerpts from the authentic text of the decisions were marked with an asterisk (*). Declarations as well as separate and dissenting opinions attached to the decisions were also analysed and initially included in smaller print; here, too, the authentic language was indicated. Later, during the transition to the World Court Digest, the inclusion of the separate opinions and declarations was called into question fundamentally. Fortunately, however, it was decided to continue with this practice after all, as these texts are extremely informative with regard to the state of the legal debate within the Court and, in particular, the development of the Court’s legal opinion.


The composition of all volumes follows a standardised structure: The first part of each volume is devoted to substantive law, with a uniform structure ranging from the foundations and basic questions of international law to specific areas such as treaty law, the law of state responsibility, international organs and war and neutrality. In this part in particular, the range of subject matters has significantly increased over the years, as international law now covers more and more areas, such as the law of the sea, aviation law, space law, human rights law, international commercial law, international environmental law, and international criminal law, to name just a few of the more recent aspects. The second part of each volume is devoted to the structure of international jurisdiction, i.e. the PCIJ and the ICJ, and has not been expanded in terms of subject matter. It covers the foundations of international jurisdiction as well as the areas of jurisdiction, proceedings, judgements, and advisory opinions.

As previously highlighted, the publication was originally entitled Fontes Iuris Gentium and subtitled Handbuch der Entscheidungen des (Ständigen) Internationalen Gerichtshofs (“Digest of the Decisions of the (Permanent) International Court of Justice”. This clarifying addition was needed, as decisions of international courts are not as such “sources” of international law, since courts do not make, but apply law. However, the boundaries are often blurred, and in international law in particular there was not only “less law” 100 years ago, but above all, there was less positive/written, law. The significance and value of the Fontes lay in particular in precisely identifying and documenting what an international court had found to constitute part of positive international law. This objective had to be kept in mind in the course of the work, as it was sometimes very tempting to keep an excerpt short and thus make the Court’s statement appear more concrete and definitive than it actually was. However, the scientific spirit and responsibility demonstrated by the team and its long-standing head Rudolf Bernhardt always ensured – we believed – the safeguarding of the “sources’” integrity and the reproduction of statements as comprehensively as necessary in order to accurately reflect their true content. This ambition also led to the inclusion of longer excerpts after the transition to the World Court Digest with the aim to reflect the question at stake, the point of view of the parties and the assessment of the Court in more detail.  However, in view of the fact that the number of cases and the volume of decisions had increased significantly over time – which documents a commendable acceptance of international jurisdiction – this made it necessary to make cuts elsewhere in the new edition. Specifically, bilingualism was abandoned and only the English texts of the decisions were now analyzed and reproduced – regardless of whether they were the authentic text version.  A summary of the cases in English was newly introduced, which was very helpful in view of the often quite complicated origins of the contentious cases and advisory opinions. In addition, a list of the judges and ad hoc judges of the period covered was compiled, so that the period dealt with could be easily overseen at a glance.

Editorial work in the pre-digital age: with scissors, paper, and glue

As Viktor Bruns had mentioned in the preface to the first volume of Series A, Sectio 1, only a team of scholars, usually four to five colleagues, could accomplish the work on the Fontes. It required, first of all, an intensive reading of the decisions and then, in no less intense discussions, the making of decisions as to which passages should be reproduced and at which point in the systematic overview. A certain continuity of at least part of the team was helpful here, so that experience with the culture of the Fontes as well as new input from new team members were ensured on a permanent basis. For colleagues, especially younger employees, this provided an opportunity to familiarise themselves intensively with the decisions of the “World Court”, which to this day are an essential reflection of the state of international law and its development. In this way, critical analysis of international law was excellently conveyed and practised.

Dorothee Bender (1970s, photo: MPIL)

In addition to the “scientific team”, which was responsible for the decisions on the selection of excerpts and their position, the production of the individual volumes was of course largely dependent on the “technical team”, which usually consisted of just two or three members. The production of the manuscript in the “pre-digital age” may, from today’s perspective, seem quite adventurous. It deserves to be explicitly recognised here, as without the meticulous cooperation of the secretary’s offices, any volume would hardly ever have been published. Of course, it is not possible to name all the individual members of staff who deserve high appreciation for their efforts, but pars pro toto Dorothee Bender should be mentioned by name here, as she was primarily responsible for this work over decades in the non-digital age and managed it with immanent perfectionism. The process was as follows: Once the team had agreed on an excerpt and its categorisation in the system, this was indicated in the text of the decision and its categorisation was noted. This excerpt then had to be cut out of the text by hand with scissors and glued onto a sheet of paper; the corresponding heading of the excerpt and the page number in the original text then had to be inserted manually and in a clearly legible manner. If an excerpt was intended for several headings, it had to be typed out several times and then followed the same process. The decisions of the court looked very holey after this procedure.  The final manuscript of the Fontes ultimately consisted of the collected pages with the glued‑on excerpts. For the two parts of volume 6, for instance – which totalled over 830 pages ‑ this was an impressive compilation, which was of course still unbound and therefore had to be carefully guarded so that not a single page was lost or misplaced. We mourned the loss of this admittedly time-consuming but reliable way of producing manuscripts when the first manuscript written on a computer at the Institute, the World Court Digest 1986-1990, suddenly and mysteriously “disappeared” shortly before it was due to go to the publisher – and could only be partially restored. However, this mishap remained a one-off – and the overall production method could be modified in line with technological advances, which soon made it possible to provide a digital online version.[5]

Concluding remarks

The facilitation of access to the original texts of the decisions of the ICJ by digital innovations is undoubtedly to be commended, though in my opinion it cannot completely replace the Fontes Iuris Gentium or the World Court Digest. After all, these publications have made it possible to follow developments in international law in certain subject areas, as expressed in the decisions of the ICJ, continuously over decades. Especially in the more novel areas of international law, such a “historical” overview is extremely worthwhile and helpful. And, it may be noted in this context, that the editors have often realised that their temptation to shorten excerpts, and thus to make the statements on international law appear more categorical than they really were at the time, would often have anticipated the actual development of the law in many cases. The opportunity to trace the development of international law in individual areas in minute detail offered by Fontes only lasted until 2005; then, the entire project was discontinued, which, like all such decisions, is understandable but also has some regrettable consequences.

Translation from the German original: Sarah Gebel

[1] Georg Friedrich von Martens, Recueil des principaux traités d’alliance, de paix, de trêve, de neutralité, de commerce, de limites, d’échange etc. conclus par les puissances de l’Europe tant d’entre elles qu’avec du monde depuis 1761 jusqu’à présent, vol. 1, Göttingen: Jean Chretien Dieterich 1791; From 1819 to 1944 the edition was published by various editors under the title “Recueil Martens”.

[2] Viktor Bruns, Preface of vol. 1, Series A, Sectio 1, S. XI.

[3] Published in Series A, Sectio 1 are volumes 1 (1931), 3 (1935) and 4 (1964) on the Permanent Court of International Justice; volume 2 (1931) on the Permanent Court of Arbitration and the International Court of Justice; volumes 5 (1961), 6 (1978) and 7 (1990) in the “old” version under the title Fontes Iuris Gentium and, starting with the case law from 1986, volumes 1-4 under the title World Court Digest, which cover the case law up to and including 2005. In Series A, Sectio 2, Volume 1 (1931) contains the decisions of the German Reichsgericht (supreme criminal and civil court in the German Empire and Weimar Republic) from 1879 to 1929; Volume 2 (1960) contains the decisions of the Reichsgericht under consideration of the Staatsgerichtshof für das Deutsche Reich (constitutional court of the Weimar Republic), 1929-1945; Volumes 3 to 9 then cover case law of the highest German courts concerning international law up to and including 1985. Series B Sectio 1 comprises the diplomatic correspondence from 1856 to 1878 in Volume 1 and the diplomatic correspondence from 1871 to 1878 in Volume 2.

[4] Karin Oellers-Frahm, Die einstweilige Anordnung in der internationalen Gerichtsbarkeit, Contributions on Comparative Public Law and International Law vol. 66, Berlin: Springer 1975.

[5] This can be found at: <>.