UPDATE: À la Recherche des Travaux Préparatoires :

An Approach to Researching the Drafting History of International Agreements


By Jonathan Pratter


Jonathan Pratter has been the Foreign and International Law Librarian at Tarlton Law Library, Jamail Center for Legal Research of the University of Texas at Austin since 1985.  He holds a law degree from the University of Nebraska and a Masters in Library and Information Science from the University of Illinois.  When not working, teaching, or writing as a law librarian, he can be found studying a new foreign language or taking a long walk.


Published April/May 2015

(Previous updates on May/June 2008 and November/December 2012)

See the Archive Version!




Table of Contents

A. What Are Travaux Préparatoires and Why Look for Them?

B. The Five Models of Publication of Travaux Preparatoires

1. Model I – Unavailable (nonexistent or inaccessible)

2. Model II – Commercial or quasi-commercial compilation and publication

3. Model III – Collected under one title

4. Model IV – Treaty-specific conference records

5. Model V – Integrated with standard international organization documents

C. Conclusion

D. 2015 Addendum



A.  What Are Travaux Préparatoires and Why Look for Them?


There are at least two good reasons why one would go in search of the travaux préparatoires to an international agreement (and/or ask for the assistance of a law librarian in doing so).  Before we go into those reasons, what exactly are travaux préparatoires ?


The phrase is of course French and translates literally as “preparatory works.”  Synonymous phrases in English are “negotiating history” or “drafting history.”  It is better to avoid using the phrase “legislative history” as a synonym.  While they bear similarities, treaty interpretation differs significantly from statutory construction (especially as the latter is done in the U.S.).  As well, there is another use for the phrase “legislative history” – as a synonym for the ratification history of an international agreement as that takes place in domestic law.


Two definitions from leading texts are:


An omnibus expression which is used rather loosely to indicate all the documents, such as memoranda, minutes of conferences, and drafts of the treaty under negotiation, for the purpose of interpreting the treaty. [1]


[T]he record of the negotiations preceding the conclusion of a treaty, the minutes of the plenary meetings and of committees of the Conference which adopted a treaty, and so on …. [2]


Two further requirements, unstated but implied in these definitions, are made explicit in the award in the Young Loan Arbitration: [3]  


It must first be stressed that the term [travaux préparatoires] must normally be restricted to material set down in writing – and thereby actually available at a later date. … A further prerequisite if material is to be considered as a component of travaux préparatoires is that it was actually accessible and known to all the original parties. [4]


The first reason for seeking out travaux préparatoires can be called the interpretive reason.  There is doubt or disagreement about the meaning of an international agreement.  Those charged with interpreting the agreement -- it could be a court, or an arbitral tribunal, or anybody who is interested in the meaning of the agreement, including scholars -- will want to consult the travaux préparatoires for insight into the “common intentions and agreed definitions” [5] of the negotiators.


The Vienna Convention on the Law of Treaties has a specific rule:


Recourse may be had to supplementary means of interpretation, including the preparatory work of the treaty and the circumstances of its conclusion, in order to confirm the meaning resulting from the application of article 31, or to determine the meaning when the interpretation according to article 31:

a)     leaves the meaning ambiguous or obscure; or

b)     leads to a result which is manifestly absurd or unreasonable. [6]


So the Vienna Convention assigns a limited, supplementary role to travaux préparatoires .  They can be consulted to confirm the meaning, to overcome an ambiguity, or to avoid an absurdity in the plain text. [7]   On the other hand, recognized treatises have ascribed a larger role to travaux préparatoires .  Jennings and Watts point out that


“[T]he International Court of Justice and its predecessor have frequently affirmed the usefulness of recourse to travaux préparatoires . … [W]here a treaty has been negotiated with thorough preparation and full deliberation, and an efficient and complete record … has been kept, the value of the travaux préparatoires may be great …. [8]


Daillier and Pellet note that “an evolution can be traced for some years now tending to accord greater weight to travaux préparatoires.” [9]


While 114 states are parties to the Vienna Convention, the United States is not. [10]   (Neither is France.)  In fact, courts in the United States take a more liberal view of the use of travaux préparatoires than the Vienna Convention does.  A recent case in the United States Court of Appeals for the Ninth Circuit, Gonzalez v. Gutierrez, demonstrates this. [11]  The international agreement before the court was the Hague Convention on the Civil Aspects of International Child Abduction.  The Gonzalez court said: “Although in interpreting a treaty we begin with the text, we may look beyond the written words to other factors for interpretive guidance.  Appropriate sources to consult include the purposes of the treaty, its drafting history , and its post-ratification understanding.” [12]   The United States Supreme Court has expressed itself even more strongly.  “ Because a treaty ratified by the United States is not only the law of this land … but also an agreement among sovereign powers, we have traditionally considered as aids to its interpretation the negotiating and drafting history (travaux préparatoires) ….” [13]   There is nothing in these quotations about a merely secondary or supplementary role for travaux préparatoires .


There is another reason for consulting travaux préparatoires that has little to do with interpretation as a matter of law.  We can call this other reason the genetic reason.  There may be absolutely no doubt about the meaning of the treaty text; it is clear to every reader, even to a lawyer.  Yet, we may take great interest in how the text of the agreement evolved into its final form.  In other words, the evolution of the text has intrinsic historical interest.  The examples are of course limitless.  One brief illustration will have to serve.


The very first right specified in the International Covenant on Civil and Political Rights is in Article 1(1), first sentence:  “All peoples have the right of self-determination.”  The researcher will learn from the travaux préparatoires that the proposal to include peoples’ right of self-determination in the Covenant sparked a sharp controversy.  Was this a political principle or a legal right?  Three contending schools of thought formed.  The debate covered a span of years in the drafting of the Covenant.  A working party to deal with the question was convened.  It made its report, recommending the inclusion of the right of self-determination.  In the end, we know what happened, but if it were not for consulting the travaux préparatoires the researcher would have no inkling of the struggle that preceded the adoption of Article 1(1), first sentence of the Covenant. [14]


B.        The Five Models of Publication of Travaux Preparatoires


The dual significance of travaux préparatoires having been established, the question naturally arises (especially for law librarians), “How do you find these things?”  The answer to this question is multifaceted.  It turns out that travaux préparatoires come in a variety of guises.  The search for travaux préparatoires can range from quite easy at one end of the scale to impossible at the other end.  The World Wide Web has made a remarkable difference, but the Web is not a universal antidote to the difficulties of tracking down travaux préparatoires .


I have produced a set of models of publication of travaux préparatoires that helps to clarify the situation for anyone who is about to embark on a search for them:


Model I – Unavailable (nonexistent or inaccessible)

Model II – Commercial or quasi-commercial compilation and publication

Model III – Collected under one title

Model IV – Treaty-specific conference records

Model V – Integrated with standard international organization documents [15]


First, however, I want to mention two kinds of resources that can immeasurably facilitate research in this area.  The first is a guide to the travaux préparatoires that tracks the agreement article-by-article, with references to the relevant places in the travaux préparatoires where each article (or the text that preceded it) is discussed.


This kind of guide is immensely useful.  Travaux préparatoires are usually published in the chronological order of their production and do not correlate well with the order of the final text of the agreement.  They can amount to a large and complex body of documentation.  If somebody takes the trouble to analyze the travaux préparatoires by collating them with the final text of the agreement, an invaluable research resource is created.  Given the amount of effort required to produce such guides, not many of them exist.  However, the researcher should always check first before launching into her own hunt.


There are few enough of these guides that they deserve to be noted here:



The second kind of resource is an article-by-article commentary on a particular agreement.  In the nature of things such a commentary refers to the travaux préparatoires , either for interpretive guidance or historical context.  Two outstanding recent examples are Schwenzer’s commentary on the CISG [18] and Schreuer’s commentary on the ICSID Convention. [19]


This is also the place to note that Yale University Law Library now produces a page on its website titled Collected Travaux Préparatoires . The object of this valuable page is to bring together references and links to collected travaux préparatoires , as they are found either in hard copy or online.  This resource is a recommended starting point for research.


In a final preliminary remark I note that the UN has started to post on the World Wide Web the travaux préparatoires for a limited number of international agreements sponsored by the UN.  Two good examples are the Arms Trade Treaty [20] and the United Nations Convention against Transnational Organized Crime. [21]   Also worthy of note in this connection is the [UN] Audiovisual Library of International Law, and in particular the component called Historical Archives. [22]  This has information on a variety of international instruments, international agreements in the main, sponsored by the UN.  For each instrument there is a “Documents” link that takes the researcher to a list of “Selected preparatory documents” in chronological order.  Some are available in full text.  An examination of these lists of preparatory documents shows that these collections can be fitted into either Model IV or Model V below.  I mention this resource here because it is available only online.


1.         Model I – Unavailable (nonexistent or inaccessible)

I start with the model of unavailability not because I want to make the process of research seem terribly difficult or frustrating, but rather because I want to inject a note of realism and common sense into the endeavor.  If it is understood that unavailability is one model of (non)access to travaux préparatoires , it will save the law librarian or other researcher the disappointment of unreasonable expectations.


Many, many bilateral agreements are entered into without a great deal of formal negotiation.  What formal negotiation does take place is not recorded, and any record that does exist is probably in the form of notes taken by the negotiators, and certainly is not intended for publication.  The obvious conclusion is that in the case of many bilateral agreements, the researcher should not expect to find travaux préparatoires .


A significant multilateral agreement that has presented much difficulty is the North American Free Trade Agreement (NAFTA).  For a decade after it went into force all documents relating to the drafting history of NAFTA were inaccessible.  That included both the various draft texts and the minutes of the meetings of the various negotiating committees.  It is not that they formally were declared confidential; you just couldn’t see them.  Then, finally, the NAFTA Free Trade Commission in the Joint Statement on its July 2004 meeting in San Antonio had this to say:


“We are committed to transparency in trade negotiations.  The negotiating texts of the NAFTA are documents of historical value and we recognize the level of public interest in them.  We asked our officials to compile the NAFTA negotiating texts, bearing in mind the time necessary to complete this.  We began the process with Chapter 11 [Investment] and are pleased to announce that Chapter 11 texts will be available through our websites.”


Sure enough, you can now find the NAFTA Chapter 11 negotiating texts on the websites of the Canadian Ministry of Foreign Affairs and International Trade [23] and of the U.S. Trade Representative . [24]   Now there are only 21 chapters of NAFTA left to go.


Two cases of confidentiality deserve mention.  For 35 years the travaux préparatoires of the Treaty of Rome establishing the European Economic Community were declared to be inaccessible. [25]   Then, in 1994 they were declared to be open.  They are now available in the Historical Archives of the European Union in Florence, Italy.


In the interest of thoroughness I note that a hardcopy resource has recently come to light:  S. Neri and H. Sperl, Traité Instituant la Communauté Économique Européenne: Travaux Préparatoires: Déclarations Interprétives des Six Gouvernements: Documents Parlementaires (Luxembourg: CJ(?), 1960).  The difficulty is that this work is only about 450 pages in extent, which raises a doubt about its completeness.  Moreover, according to OCLC WorldCat only five libraries in the world hold this work, and four of them are associated with the EU!


The travaux préparatoires of the European Convention on Human Rights were published in eight volumes between 1975 and 1985. [26]  In addition the Library of the European Court of Human Rights, [27] through its catalogue, makes available the texts of discrete segments of the travaux préparatoires by article of the Convention. These are PDFs of the mimeograph originals, so that the hardcopy collection just mentioned should probably be preferred.


However, the travaux préparatoires relating to all of the other international agreements sponsored by the Council of Europe, and there are 215 of these, remain confidential.  In 1965 the Consultative Assembly recommended to the Committee of Ministers that it should authorize the publication of the travaux préparatoires of Council of Europe conventions. [28]   The Council of Ministers refused, but agreed to put in place an alternative. [29]  This is the publication of an official explanatory report that accompanies each convention sponsored by the Council of Europe.  Not all conventions or protocols have an explanatory report, but most do.


These are easy to find.  At the main Council of Europe conventions website , the researcher needs only to click on “Full list,” and then on the name of the convention of interest.  The researcher then clicks on the link to the Explanatory Report, which comes up in full in printer-friendly HTML format.  In hardcopy the texts of the Council of Europe conventions together with their Explanatory Reports are available from the Council of Europe Bookshop Online .


2.         Model II – Commercial or quasi-commercial compilation and publication

Brill/Martinus Nijhoff, a leading commercial publisher of mainly international law, began in 2008 a series titled The Travaux Préparatoires of Multilateral Treaties .  The first volume in the series is a work titled The Slavery Conventions: the Travaux Préparatoires of the 1928 League of Nations Convention and the 1956 United Nations Convention .  Volume two in the series, also published in 2008, is The Genocide Convention: the Travaux Préparatoires .  Unfortunately, no additions to the series have appeared since then.  Given the amount of time and effort required to produce such works, it is understandable that the pace of publication is slow.  Nevertheless, one would have thought that the publisher had a set of projects in the pipeline before undertaking such a series. [30]


Therefore, it is not surprising that much of earlier Model II publication was quasi-commercial in the sense that, while the product was done as a specially compiled and titled work, the publisher was not strictly speaking a commercial one.  A good example is the book The Travaux Préparatoires of the International Convention for the Unification of Certain Rules of Law with respect to Collision between Vessels 23 September 1910 … (1997).  It is clear that this work was specially compiled and distinctively titled, but the publisher is the Comité Maritime International (CMI), the organization that sponsored the drafting and adoption of the conventions.  Another example is The Collected Travaux Préparatoires of the 1951 Geneva Convention Relating to the Status of Refugees (1990).  Here the publisher is the Dutch Refugee Council together with the European Legal Network on Asylum.


The advantage of Model II publication for research is obvious.  The phrase “travaux préparatoires” almost invariably appears in the title of the work, which means, obviously, that the phrase also appears in the cataloging record created by libraries for the work.  In contrast, this convenient characteristic of Model II publication is almost invariably not the case in the other models of publication.


3.         Model III – Collected under one title

The Hague Conference on Private International Law is the best example of Model III.  The first Hague conference was convened in 1893.  In 1951, the Hague Conference became a permanent intergovernmental organization, founded on its Statute, which is an international agreement.  The Conference has 77 member states, but over 140 nations from all parts of the world are parties to one or more Hague conventions.


The purpose of the Hague Conference is “to work for the progressive unification of the rules of private international law.” [31]   The principal method for achieving this goal is the drafting and negotiation of multilateral international agreements.  To date the Hague Conference has sponsored 39 conventions on a variety of subjects in private international law.


The working method of the Hague Conference can be summarized as follows.  The Permanent Bureau (secretariat) undertakes preparatory work.  Then preliminary drafts of a convention are drawn up by special commissions made up of governmental experts.  The draft convention is then considered, debated, perhaps amended, and adopted at a plenary session of the Hague Conference, which convenes every four years.


What kind of documentation is produced during the preparation of a Hague convention?  We can identify at least the following kinds:


}    In one category, the documents of the preliminary work, which include such items as:

o    A report drawn up by the Permanent Bureau on the advisability of proceeding to draft a convention on the subject in question;

o    A checklist drawn up by the Permanent Bureau of provisions that might be included in the convention;

o    The preliminary conclusions of the special commission appointed to draft the convention;

o    The preliminary draft convention adopted by the special commission;

o    An explanatory commentary on the draft convention drawn up by one of the expert members (the rapporteur) of the special commission.


}    In another category, the so-called “Acts” of the session of the Hague Conference that considered the convention, including:

o    Working documents of the commission appointed by the plenary to consider the draft convention, including proposals to amend the draft text;

o    The minutes of the meetings of this commission;

o    Working documents of the plenary session;

o    The minutes of the plenary session.


}    In a third category, the final text of the convention as adopted and the official explanatory report drawn up by a recognized expert delegate, usually the person who had previously been appointed as rapporteur during the drafting phase.


All of this documentation is gathered up and published in a series known popularly as “ Actes et Documents .”  If these volumes are cataloged under one title and classed together, as in my view they most certainly should be, the full title will be Actes et Documents de la … Session .  The corporate author will of course be the Hague Conference on Private International Law.


Now, the Hague Conference maintains an excellent website. [32]  The question is, what effect has the website had on the availability of Hague Conference travaux préparatoires ?  The answer is that there has been a highly beneficial, but not quite revolutionary, impact.  The reason is that the full text of the Actes et Documents set, as such, has not been put on the website.  Rather, various extracts or off-prints or analogous texts have been made available.  Let us examine this phenomenon in some detail.


The Actes et Documents set is listed on the website for each session, but the links for each volume do not go to the full text.  Rather, each volume is offered for sale in hardcopy.  The situation for the rapporteurs’ explanatory reports is better.  In all, 30 explanatory reports are listed.  Of these, 19 are available in full text.  Most of these are PDF off-prints from Actes et Documents .  Fortunately, several key Hague conventions are covered:  Evidence, Child Abduction, Trusts, Intercountry Adoption, Child Protection, Protection of Adults, and Choice of Court Agreements.


As for preliminary documents, such as those listed above in the first category, for some recent conventions there is a good selection, but for most conventions, there is nothing at all.  And also for a few recent conventions there is a link to the final act of the Conference session that adopted the conventions.  In sum, researchers should make full use of the website when researching the travaux préparatoires to Hague Conference conventions, but any library with a substantial collection of international legal materials should definitely have the hardcopy Actes et Documents.


The other example of Model III is the International Institute for the Unification of Private Law (UNIDROIT).  UNIDROIT also is an intergovernmental organization whose purpose is to study the needs and to propose methods for modernizing, harmonizing and coordinating private law, in particular commercial law.  UNIDROIT’s main working method is to draft legal instruments, including international agreements, that harmonize the law in particular fields.  UNIDROIT emphasizes that its focus is on rules of substantive law, not rules of private international law, which is what distinguishes UNIDROIT from the Hague Conference.


Much of UNIDROIT’s official work is published in an annual series under one title.  Beginning in 2004, the French parallel title ( Actes et Documents ) was dropped, and the series is now known by its cover title in English, UNIDROIT Proceedings and Papers .  This series publishes official documents of UNIDROIT, including the final texts of adopted uniform laws or international agreements, as well as preliminary study documents and draft instruments that qualify for the description of travaux préparatoires .  The UNIDROIT series is not as complete as the Actes et Documents of the Hague Conference, because it does not contain the verbatim record of meetings.  The UNIDROIT website tells the researcher that the Proceedings and Papers publication is available on the website.  Sadly, the link is dead.


4.         Model IV – Treaty-specific conference records

This is both the most common model and the most difficult to research.  When a significant multilateral agreement is to be negotiated, a sponsoring body – it could be a state, say Switzerland; or it could be an international organization, say the United Nations – convenes an international conference to which as many states as possible will be invited to send representatives.  The conference elects a president and organizes itself into working committees, usually including a drafting committee.  The conference works through and prepares the text of the agreement, which is then adopted in the plenary session of the conference in a document called the “final act.” [33]  In the course of its work, the conference produces such documents as procès verbaux (verbatim records of discussion in committee and plenary sessions), summary records of discussion, and working drafts of the agreement.  These documents are gathered together and published as the record, that is, the travaux préparatoires , of the conference.


From the perspective of research, the difficulty of working with Model IV is that there is no necessary link between the name of an international agreement and the travaux préparatoires of the conference that produced it.  Indeed, it may be difficult to determine exactly what the correct name of an international agreement is.  The cataloging record of the conference publication does not have to, and frequently does not, contain a tracing or added entry for the name of the agreement.  Therefore, a search using the name of the agreement, which is usually what the researcher has in hand, often will not turn up the cataloging record for the travaux préparatoires . [34]


An example illustrates both the difficulty and the method.  The famous international agreements negotiated in 1949 for the protection of war victims, are called almost universally in the legal literature by their popular name, the Geneva Conventions. [35]  This is the information that the researcher has to start with.


A search on OCLC WorldCat for keywords “Geneva Conventions” returns almost 7000 records.  A title-keyword search returns over 1700.  Knowing that the conventions were negotiated in 1949, we can limit the search by year.  This brings up the formal name of the negotiating conference as that appears in the cataloging for early publications of the text of the conventions: Diplomatic Conference for the Establishment of International Conventions for the Protection of Victims of War (1949: Geneva, Switzerland).  A search on the terms in the name of the conference quickly reveals the record for the travaux préparatoires under their formal title, Final Record of the Diplomatic Conference of Geneva of 1949 .  This method of searching for Model IV documents may not be ideal, but it does work.  The caveat that I would enter here is that for best results the researcher should be a law librarian. [36]


The U.N. has a website titled United Nations Diplomatic Conferences .    Here are found links to the full texts in PDF of diplomatic conferences for the negotiation of a dozen significant international agreements sponsored by the U.N. between 1958 and 1998.


There is a complicating factor that deserves mention.  It can happen that the record of the negotiating conference does not represent the entirety of the travaux préparatoires.   A good example is the Vienna Convention on the Law of Treaties.  The travaux préparatoires of the conference have indeed been published. [37]   However, in the decade preceding the decision to convene a conference (1955-1966), the International Law Commission carried out an immense amount of preparatory work on the law of treaties, which led directly to the General Assembly’s decision to convene the conference.  Fortunately, this work is thoroughly documented on the website of the International Law Commission .  This pattern obtains for several agreements for which the International Law Commission did preparatory work.  The Commission’s website has complete information.  (See the links under “Researching the work of the Commission,” in particular the “Research Guide.”)


5.         Model V – Integrated with standard international organization documents

Working with Model V requires knowledge of the international organization in question and its documentation.  Each international organization is unique and has to be taken on its own terms.  Of course, we are concerned here with international organizations that carry out law-making functions through the adoption of international agreements.  There are several of these.  We take the International Labour Organization (ILO) as an example, partly because the ILO, through its outstanding website, has revolutionized research into its travaux préparatoires .


Founded in 1919, the ILO became the first specialized agency of the UN in 1946.  The structure of the ILO has three main components.  The International Labour Conference is composed of the member states of the ILO.  The Conference meets in Geneva in June of each year.  The Governing Body is the executive council of the ILO.  One of its significant functions for our purposes is to propose the agenda of subjects to be considered by the Conference.  The International Labour Office is the permanent secretariat of the ILO in Geneva.


One of the major functions of the ILO is standards-setting on a broad array of questions in the general field of labor and labor rights.  One of the chief mechanisms of the ILO’s standards-setting activity is the adoption of international agreements, called conventions.  As of 2012, the ILO has adopted 189 conventions dealing with all aspects of labor standards.  One of the most significant recent conventions, which we take as an example, is the Worst Forms of Child Labour Convention of 1999. [38]


Once the International Labour Conference accepts a standards-setting issue for its agenda, it asks the International Labour Office to prepare reports for submission to the next session of the Conference.  The Conference then assigns consideration of the agenda item to a committee of the Conference.  The committee makes its own report, which is considered by the plenary of the Conference.  If the proposal for enacting a standard in the form of a convention is approved, the plenary session of the Conference adopts the text of the proposed convention.  It should be noted here that the consideration of proposed conventions usually takes place during two consecutive sessions of the Conference. [39]


In hardcopy, the work of the International Labour Conference is published in the annual serial, Record of Proceedings .  On the other hand, the reports that the International Labour Office submits to the Conference are published as discrete bibliographic entities.  Each has a title and even an ISBN.  A header on the title page referring to the session of the Conference, along with a reference to the agenda item, indicate that the document is a report made to the Conference.


I want to propose that today the ideal method for researching the travaux préparatoires of ILO conventions online.  The ILO has an outstanding website with all kinds of information.  One of the central components of the website is the database of ILO-sponsored labor standards, called NORMLEX . Of course, a major element of NORMLEX is the collection of ILO conventions . The full text of every ILO convention is easily accessible, along with links to ratification information.  However, there is no link from the text of a particular convention to the relevant travaux préparatoires , even though they are available elsewhere on the ILO website.


To get to the travaux préparatoires you have to go to the link on the ILO homepage to the International Labour Conference.  With the knowledge that the Worst Forms of Child Labour Convention was adopted in 1999, the researcher selects the link for the 87 th Session, which took place in June 1999.  There the researcher finds links to a full range of sessional documents.  For our purposes, the significant documents are:


}    “Reports Submitted to the Conference” (by the International Labour Office).  These are organized by agenda item with an indication of the subject, so it is easy to find the reports titled “Child Labour.”


}    “Reports of Conference Committees and discussion in Plenary.”  These too are organized with an indication of the subject.


Knowing that proposed conventions are discussed at two consecutive sessions of the Conference, the researcher then goes to the link for the 86 th Session (1998) and repeats the process. [40]   Here I have to enter a caveat.  This kind of research using the ILO website is possible only as far back as the 85 th Session.  However, from 1997 going forward, the website has revolutionized research on the travaux préparatoires of ILO conventions.


C.         Conclusion

Having worked through all five models, the observant reader will have noted a small difficulty.  When the United Nations General Assembly convenes an international conference for the purpose of negotiating an international agreement, the travaux préparatoires produced by the conference are normally published as stand-alone bibliographic entities, which puts them in Model IV.  However, at the same time these documents are issued in the A/CONF./[number] series of United Nations documents, which puts them in Model V.  This dissolves the distinction between Models IV and V.  In response to this small difficulty, I do not claim that these models represent airtight categories.  I do maintain that my five models of travaux préparatoires represent a useful working framework in which to approach the task of researching them.



D.        2015 Addendum

There is more than one way to interpret an international agreement. [41]  The Vienna Convention on the Law of Treaties imports several interpretive methods, and not always in a harmonious way.  The primary approach is usually labelled objective (Art. 31(1) – “the ordinary meaning to be given to the terms of the treaty in their context”).  But Art. 31(1) also uses the phrase employed classically to express the teleological approach to interpretation (“in the light of its object and purpose”).  The search for travaux préparatoires under Article 32 partakes of what is called the subjective, the historical, or the intentional mode of interpretation.  The idea here is to divine the intentions of the drafters at the time of the conclusion of the agreement, which of course may be far in the past.  But the Vienna Convention also adopts elements of the evolutionary approach, which takes account of the change over time in the meaning of the terms of the agreement (Art. 31(3)(b) – “Any subsequent practice between the parties regarding the interpretation of the treaty or the application of its provisions”). [42]


Does this mishmash of interpretive approaches deserve criticism?  Not in my view.  Authorized commentators accept without question that interpretation relies on a basket of possibilities for establishing the meaning of a text in international law. [43]  A moment’s reflection shows why this is so.  The point of interpretation is to gain understanding of the meaning of a text.  Understanding by human beings, and this includes those who work in the field of public international law, implies the working of a complex and subtle process through which one can arrive at a stronger grasp of the sense of a text.  Are we prepared to tell those who seek understanding that they are required to limit themselves to the use of only a limited number of tools normally employed by anyone who wishes to strengthen her grip on a difficult set of words and concepts?  I think not.


The literature on treaty interpretation does quite well at identifying the catalog of interpretive techniques and at locating the search for travaux préparatoires in the domain of subjective/historical/intentional interpretation.  Unfortunately, however, the literature comes up short on the question of why we should engage in that kind of interpretation at all.  To put it in crudely American terms, why should today’s interpreter care at all about what the original framers thought?  And in fact, we can find cases in which international instances display utter disregard for the historical approach.  The European Court of Justice simply never refers to the travaux préparatoires of the Union treaties when deciding cases arising under them. [44]


Public international law has a healthy mechanism for remedying gaps in the law, whether in primary sources or secondary commentary.  It refers to analogous figures in national law and brings them into the public international legal domain suitably modified for use there.  The question of historical/subjective/intentional interpretation arises all the time in national law, and there the whys and wherefores of it have received attention.  In the United States the analogous issue of how to use legislative history as a way of gaining insight into the drafters’ intentions has generated a good deal more heat than light.  Therefore, we will pass over the acrimonious spat in that jurisdiction and settle on the more enlightening approach to historical/subjective/intentional interpretation we find in Germany.


Vogenauer, in his work on the comparative interpretation of statutes, [45] says that in consulting what he (following the courts) calls “official materials” courts have a two-fold purpose. [46]  First, they want to inquire into what he calls the legislator’s Normvorstellungen .  Now, this is an unusual word.  It occurs neither in standard nor in legal dictionaries.  Fortunately, the word Vorstellung is well-defined.  It means idea, notion or concept, and can also mean mind or imagination.  Vogenauer says it means the subjective ideas of the drafters about the meaning of the provision in question.  Vogenauer points out that the courts have drawn a parallel between the drafters’ “concrete” ideas and the “will of the legislator.”  Second, the courts wish to determine the Ziele und Zwecke (aims and objects) of the legislator.  This manifests a direct correlation between subjective intent and the teleological approach to interpretation.  It leads to a search for the legislator’s Wertungen (appraisals of value) and Abwägungen (weighing up of considerations in the balance).  Vogenauer then quotes the Bundesgerichtshof (Supreme Court in Civil and Criminal Cases) in a case from the middle of the last century:


… which economic and social conditions the drafters had in view [ vor Augen hatten ], from which legal position [ Rechtszustand ] one started and which reform efforts [ Reformbestrebungen ] the legislator wished to take account of … -- in other words: what clashes of interests the legislator wished to conciliate [ ausgleichen ] … -- and so finally to which constellations of cases [ Fallgestaltungen ] the legislation is to be applicable. [47]


We see then that the raison d’être of the subjective/historical/intentional approach receives a fuller treatment in national law than it does in public international law.  Nevertheless, it is quite clear to me that a good deal of what national law has to say about the matter translates without difficulty to the international plane.


What if anything does all this have to do with the eminently practical process of searching out travaux préparatoires ?  First, let it be observed that the substance of treaty interpretation, and in particular the subjective/historical/intentional approach, could not take place without good research first having been done.  But is the converse also true, i.e., is a knowledge of the substance of interpretation essential to doing good research?  Let’s ask a more general question: Is a knowledge of the substance of public international law of value to one who is doing research, including what I will call documentary research, on public international law?  The question could not more clearly answer itself – Yes, just as sure as God made little green apples, or as we could say in German, so sicher wie das Amen in der Kirche .



[1] Lord McNair, The Law of Treaties 410 (1961).

[2] Robert Jennings and Arthur Watts, 1 Oppenheim’s International Law 1277 (9 th ed. 1992).

[3] 59 I.L.R. 495 (1980).

[4] Id. at 544.

[5] Id. at 545.

[6] Vienna Convention on the Law of Treaties, May 23, 1969, art. 32, 1155 U.N.T.S. 331 (emphasis added).  Article 31(1) of the Convention, which contains the general rule of interpretation, reads:  “A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.”

[7] A professor at the Michigan Law School has maintained recently that far from assigning a limited role to travaux préparatoires , the drafters of the Vienna Convention intended that travaux préparatoires would work as an integral part of the interpretive search.  Julian Davis Mortenson, The Travaux of Travaux : Is the Vienna Convention Hostile to Drafting History?, 107 Am. J. Int’l L. 780 (2013).  The paradoxical difficulty with this thesis is that it runs head-on into the plain text of Article 32 of the Convention.

[8] Sir Robert Jennings, Sir Arthur Watts, Oppenheim’s International Law 1277 (9 th ed. 1992).

[9] Patrick Daillier, Allain Pellet, Droit International Public 286 (8 th ed. 2009) (citing several cases in the International Court of Justice).

[10] This fact has not prevented U.S. courts from citing the Vienna Convention in a variety of contexts.  A WestlawNext search in U.S. federal cases returns 131 cases that have cited the Convention.  Chubb & Son, Inc. v. Asiana Airlines, 214 F.3d 301 (2d Cir. 2000), has a good discussion of the court’s view of the Vienna Convention’s place.  The court said in conclusion that “[w]e therefore treat the Vienna Convention as an authoritative guide to the customary international law of treaties.”  See also, Maria Frankowska, “The Vienna Convention on the Law of Treaties before the United States Courts,” 28 Va. J. Int’l L. 281 (1988); Evan Criddle, The Vienna Convention on the Law of Treaties in U.S. Treaty Interpretation, 44 Va. J. Int’l L. 431 (2004).

[11] 311 F.3d 942 (9 th Cir. 2002).

[12] Gonzalez at 948 (emphasis added).

[13] Zicherman v. Korean Air Lines Co., 516 U.S. 217, 226 (1996) (emphasis added).

[14] Naturally, I did not develop this example from original research.  Rather, I consulted Bossuyt’s unsurpassable guide to the travaux préparatoires of the Covenant, mentioned below.

[15] Lomio, Spang-Hanssen and Wilson see fit to attack my entire project in the following terms: “However, only model III (treaty-specific conference records) refers to the fact that legally binding international instruments are always done through a conference between states of international organizations representatives.  This means that although travaux preparatoires material under models I, II, & IV can be found through one of the parties’ archives or websites, these models do not illustrate how international instruments are made or how the travaux preparatoires come into being or are produced – thus they cannot be relied upon.”  J. Paul Lomio, Henrik Spang-Hanssen & George D. Wilson, Legal Research Methods in a Modern World 356, n. 34 (3 rd ed. 2011).  Mssrs. Lomio, Spang-Hanssen and Wilson have simply missed the point – this guide is about the publication of travaux préparatoires and how to find them .  I discuss “how international instruments are made or how the travaux préparatoires come into being or are produced” for contextual and illustrative purposes.  I hope they will come to see the difference between these two things and will correct this egregious misapprehension in a future edition of their book.

[16] This source is a monograph, but it contains detailed references to the travaux préparatoires on several aspects of the Covenant.

[17] Granted, the Declaration is not an international agreement .  How much does that matter in this context?

[18] Ingeborg Schwenzer, Commentary on the UN Convention on the International Sale of Goods (CISG) (3 rd ed. 2010).

[19] Christoph Schreuer, The ICSID Convention: A Commentary on the Convention on the Settlement of Investment Disputes between States and Nationals of Other States (2 nd ed. 2009).

[23] http://www.international.gc.ca/trade-agreements-accords-commerciaux/topics-domaines/disp-diff/trilateral_neg.aspx?lang=eng .  As in the case of the USTR’s collection of these texts (see just below), the Canadian site is an archive, meaning that updating with more documents for other NAFTA chapters is not going to happen.

[24] https://ustr.gov/archive/Trade_Agreements/Regional/NAFTA/NAFTA_Chapter_11_Trilateral_Negotiating_Draft_Texts/Section_Index.html .  I much regret to note that this page is part of the USTR’s archive site, which means that it is not going to be updated.  This simply proves the validity of Model I—Unavailability.

[25] See the extract of the procès verbal of the meeting of the Committee of Permanent Representatives, April 3, 1959 (on file with the author).

[26] Collected Edition of the “Travaux Préparatoires.”  The Hague: M. Nijhoff, 1975-1985.  The formal name of the convention is Convention for the Protection of Human Rights and Fundamental Freedoms.  The convention is frequently cited to 213 U.N.T.S. 221.  However, as the convention has been amended several times, the best place for a reliable current text of the convention is the Council of Europe’s conventions website .

[28] Recommendation 417 (1965) (on file with the author).

[29] Conclusion (65) 144 (on file with the author).

[30] On the Brill website one can find indications of a forthcoming work in the series: The Foundations of Modern International Law on Indigenous and Tribal Peoples: The Preparatory Documents of the Indigenous and Tribal Peoples Convention (vol. 1: Basic Policy and Land Rights) (ISBN 9789004289055).  The work is supposed to be due in July 2015.

[31] Statute of the Hague Conference on Private International Law, art. 1, Oct. 9, 1951, 15 U.S.T. 2228, 220 U.N.T.S. 12.

[33] This is obviously a much-abbreviated summary.  For a fuller account see Robert Jennings & Arthur Watts, Oppenheim’s International Law 1183-1187 (9 th ed. 1992) and “Conferences and Congresses, International” in the Max Planck Encyclopedia of Public International Law (2012 and online).

[34] To be fair to catalogers, I should say that when the travaux préparatoires include the text of the agreement, an astute cataloger will take note of that fact by making an added subject entry for the name of the agreement.  That will establish the link the researcher hopes for.

[35] A WestlawNext search on the phrase “Geneva Conventions” in the Law Reviews and Journals sector returns over 7900 documents.  Searches on other variations of the name return far fewer.

[36] This caveat is brought home by the observation that a Google search on the phrase “Geneva Conventions” has over 500,000 hits, rendering it almost useless.

[37] United Nations Conference on the Law of Treaties: Official Records.  New York: United Nations, 1969-1971.

[38] 2133 U.N.T.S. 163, 38 I.L.M. 1207 (1999).  The full name is Convention (C182) Concerning the Prohibition and Immediate Action for the Elimination of the Worst Forms of Child Labour.  However, the preamble to the convention says that it may be cited as the Worst Forms of Child Labour Convention.  Here we have another example of the name difficulty discussed above.

[39] For a fuller discussion of the structure and working methods of the ILO, see Ebere Okieke, Constitutional Law and Practice in the International Labour Organisation (1985).

[40] The ILO website also has a section devoted to its International Programme on the Elimination of Child Labour: IPEC .

[41] The standard work today is Richard K. Gardiner, Treaty Interpretation.  Oxford; New York: Oxford University Press, 2008.  This is unlikely to be improved upon until the second edition appears.

[42] For a comparative view on the position in France and the United Kingdom see Eirik Bjorge, The Vienna Rules on Treaty Interpretation before Domestic Courts, 131 L. Quarterly Rev. 78 (2015).  Unhappily, the author does not enter into the central question of the use of travaux préparatoires under art. 32 of the Convention.  He does mention the leading case in the House of Lords of Forthergill v. Monarch Airlines, [1981] A.C. 251, but in a different and more general connection.  He omits any mention of another more recent case in the House of Lords, Morris v. KLM Royal Dutch Airlines, [2002] 2 A.C. 628.  There Lord Hope of Craighead (at para. 79) had this to say: “It is legitimate to have regard to the travaux préparatoires in order to resolve ambiguities or obscurities [citing Forthergill ].  But caution is needed in the use of this material as the delegates may not have shared a common view.  An expression by one of them as to his own view is likely to be of little value if it was met simply by silence on the part of the other delegates.  It will only be helpful if, after proper analysis, the travaux clearly and indisputably point to a definite intention on the part of the delegates as to how the point at issue should be resolved.”

[43] Matthias Herdegen, Interpretation in International Law, Max Planck Encyclopedia of Public International Law (online).  Herdegen does a good job of concisely running through the catalog of interpretive techniques on offer in public international law.

[44] Anthony Arnull, The European Union and its Court of Justice 614, 615 (2 nd ed. 2006) (“This [recourse to travaux préparatoires ] is not a method which has in the past commended itself to the Court in cases concerning the interpretation of the Treaties themselves.” Arnull goes on to point out that “[i]ncreasing  pressure for transparency and the development of the Internet have now brought many travaux préparatoires concerning subsequent amendments to the Treaties themselves into the public domain. … This material is now routinely referred to in academic literature and will inevitably be drawn to the Court’s attention.  It seems inconceivable that the Court will be able to resist using it as an aid to the interpretation of Treaty provisions.”).  Gerard Conway, The Limits of Legal Reasoning and the European Court of Justice 255 (2012) (“The ECJ does not usually rely on travaux préparatoires or seek to unearth original understanding. … The Court is less willing still to refer to the travaux préparatoires of the Treaties.”).

[45] Stefan Vogenauer, 1 Die Auslegung von Gesetzen in England und auf dem Kontinent: Eine vergleichende Untersuching der Rechtsprechung und ihrer historischen Grundlagen (2001).

[46] Id. at 31, 32.

[47] BGH NJW 1967, 343, 348.