Introduction
The index to International Law Reports (ILR) has always been regarded as an intrinsic part of the project. Indexers and indexing styles may have changed over the years but it is remarkable the extent to which the first index (1929) carries the marks of each that has followed. ILR indexes have always been detailed and precise, as concerned with concepts as with names, mean with author citations, and focused on points of legal significance rather than interesting facts not essential to an understanding of the legal material. The index to Volume 138, now in the press, will follow this pattern. In addition to the index, the individual volumes have also always included tables of cases and (since Volume 25) a table of treaties.
From Annual Digest to International Law Reports
The International Law Reports began life in 1929Footnote 1 as the Annual Digest of Public International Law Cases, in 1940Footnote 2 was renamed Annual Digest and Reports of Public International Law Cases, and in 1956 became simply International Law Reports. The then editor, Professor Hersch Lauterpacht, wrote in his preface to Volume 7
“The change indicates that a further step has been taken in the direction of accomplishing the object which was inherent in the purpose of the Digest from its inception, namely the publication of International Law Reports proper. This object has been gradually realized with the appearance of each successive volume. Whenever possible, the paraphrasing of decisions has given place to a literal reproduction of the relevant parts of judgments and awards; at the same time the length of verbatim extracts has been considerably increased.
…
On the whole the editorial policy has been to replace digests by full or abridged reports suitably edited and annotated. The result is that in most cases the volume is no longer an index to sources, but that its contents form a primary source of decisions either in their original form or in translation. At the same time the wishes of those users of the Digest who seek for a short statement of the underlying legal rule are to some extent met by the headings, the sub-titles and by the expedient of inserting the operative part of the decision at the beginning of the report or digest. But full or abridged verbatim reports fulfil an obvious purpose which overrides any passing inconvenience. In addition to their other uses in the field of international law, they show how – by apparently divergent mental processes and in spite of differences in judicial method, approach and terminology – tribunals and judges of various countries reach identical results, demonstrating in a significant manner the essential unity of the law of nations”.
Always underlying the commitment to such a series of reports was the service that they would provide in the realisation of Article 38 of the Statute first of the Permanent International Court of Justice and then of the International Court of Justice (judicial decisions as evidence of customary international law and the “general principles of international law recognized by civilized nations”).
The other major change along the way has been in the arrangement of the cases. From Volumes 1–74, this was by classified content, replaced in Volume 75 by a Digest of Cases by subject heading, a change not without implications for the index.
Tables of cases and treaties
Cases
The cases have always been listed both alphabetically and by jurisdiction. In addition, Volumes 1–24 included a table of cases cited, a practice discontinued for the simple reason that “enquiry showed that the use made of this table does not really justify its preparation”. But, more useful probably than simply a list of cases cited, the subject index increasingly includes references to cases of particular importance in a given context and may also on occasion (e.g. Chorzów Factory), where the case name has become shorthand for a legal principle, include it as a main heading.
Cases by jurisdiction
The table of cases by jurisdiction raises few problems. The citations simply follow the official version at each stage of the proceedings.
Cases in alphabetical orderFootnote 3
The alphabetical table, particularly in its consolidated form, is more challenging. ILR policy is to enter cases under all common variants of their title. Where the cases do not have a title as such in their original version, e.g. in jurisdictions where the practice is to use case numbers rather than titles, the case name may have been “invented” by the ILR editors or by others. These “invented” names, inevitably, may take a number of guises, the post-WWII war crimes trials being full of examples such as List (Hostages Trial). The answer is to double post and so this is also included in the table as Hostages Trial (Re List and others).
For purposes of the alphabetical table, case titles have often been simplified. Where, as for example with ICJ cases, it would be misleading to include the full formal title including the introductory “Case concerning”, the cases are listed under the key element[s], e.g. “Aerial Incident of 27 July 1955”. Phrases such as “Government of”, “Kingdom of”, “Republic of” are omitted unless needed to distinguish between countries or between the same country at different periods or where the case[s] themselves make it clear that it is important to retain the phrase. For international tribunals, “Liberia”, “Iran” etc. will usually suffice. In proceedings in its own courts involving the State, the formula adopted is “State (France)”, “Public Prosecutor (France)” etc. so that it is clear in the alphabetical table which jurisdiction is involved. “Secretary of State for Home Affairs” and its variants is “Home Secretary”, without an indication of jurisdiction since this title is used only in the United Kingdom. “Canton of”, “City of”, “Commonwealth of”, “Commune of”, “Dominion of”, “Government of”, “Kingdom of”, “Municipality of”, “People of”, “Principality of”, “Republic of”, “Town of”, “Union of”, are listed alphabetically according to the place concerned eg, India (Union of). For the sake of consistency English is normally used in case titles e.g. “Administration des Douanes” is rendered as “Customs Administration”.
Treaties
The introduction of the table of treaties was explained as follows in the Preface to Volume 25: “The frequency with which the decisions of international and municipal courts involve points of treaty interpretation is increasing; and it has been felt, therefore, that it would be helpful to incorporate this piece of technical apparatus which will enable the reader readily to ascertain whether any particular section of a treaty has been the subject of judicial consideration of mention”.
The rubric to the table of treaties always includes the caveat that “[i]t has not been possible to draw a helpful distinction between treaties judicially considered and treaties which are merely cited”. Does the table of treaties still serve a useful purpose? I think it does, not so much, perhaps, for the more frequently referenced treaties where it might be more profitable to resort to the subject heading in the index, but certainly for the more recondite treaties. These are unlikely to appear by name in the index, and I find myself often resorting to the table of treaties to see if a particular treaty has appeared previously. But treaties remain notoriously difficult to track down even in this age of Google, and an additional service offered by the ILR Table is to provide details as accurate as possible of treaty title, date of signature and places where the text may be found including, wherever possible, a reference to an English language text and, increasingly, a reference to a readily available collection of documents which students are likely to have on their own shelves. Such references are updated continuously.
One problem with any long-running table of treaties is, of course, the matter of revision and amendment. The most troublesome example of this, perhaps, is the EEC-European Communities-European Union conundrum, although the European Convention on Human Rights is not without its problems. Easy enough, perhaps, on a single case basis, even a single volume basis, but the moment it comes to recording citations over time, it is difficult to find a user-friendly way of showing the evolution. There are tricks, yes: but tricks aren't always enough.
Consolidation
Indexes to individual volumes can only be part of the story – it is with their accumulation into a single coherent index that patterns can be seen and the true extent of the jurisprudence on a given topic, superficially esoteric or novel, be fully appreciated. There have been several consolidated indexes to the ILR: Volumes 1–10 (published 1947), Volumes 1–35 (1969), Volumes 36–45 (1973), Volumes 36–80 (1990), and Volumes 36–125 (2004). Work is in hand on a consolidated index to Volumes 1–150. It was with the 1990 consolidated index that my own involvement with the indexing of the ILR began. I was at the time assistant director at the Research Centre for International Law in Cambridge, now the Lauterpacht Centre, and had found myself a year or two previously indexing the Iran-US Claims Tribunal Reports. I came to this as a complete novice in indexing terms, found that I enjoyed it and had a talent for the task and, foolishly perhaps, offered to help a colleague working on the 1990 ILR consolidation.
Doing it by hand
This was the late eighties, when, hard though it may be to believe, the computer as far as the individual user was concerned was still in its infancy. The PC was a novelty. The first dedicated indexing software had been introduced in 1982, but I was not aware of it and most indexers still used index cards and shoeboxes. The original assumption when work began on the Volumes 36-80 consolidation was that this was just a question of merging and resorting existing material. A sophisticated, mainframe, computer programme had been devised and the relevant data fed in with great care. But different indexing styles over the years, the changing nature of the subject matter and the idiosyncrasies of individual indexers, quite apart from the many trivial variations seized on by the computer as an excuse to scatter related entries far and wide, meant the material was simply not homogeneous enough for this approach. The only result was an amorphous mess beyond the control of man or machine.
With great reluctance we reverted to the old fashioned way of doing it – in other words I sat down at my desk and patiently went through all the indexes one by one, building them gradually into a single, consolidated version. This was above all a consolidation. In other words, revisiting the text was kept to a minimumFootnote 4 and the temptation to re-index resisted. Headings needed to be synchronised and I made an occasional check of the source material when it looked as if the index entry might be an error. But that was it. I used my PC as a glorified typewriter with all the wonderful tools like cutting and pasting the new word processing software offered, but that was it. All the sorting I did manually. and believe me that is no easy task. You think you know your alphabet? I dedicated the last hour of each working day to backing up my files, all forty of them, to the University mainframe, always asking myself when tempted not to do this whether I would mind coming in the next morning and finding my work had vanished.
Current indexing practice
Harnessing the technology
As I came towards the end of this task of consolidation, I also took on the indexing of the individual volumes, starting at Volume 76. As it happens, this may have been one of the most difficult volumes of all to index. It was dedicated to the ICJ Nicaragua case, challenged for difficulty only by the long run of Tin Council cases which appeared in the succeeding volumes. I struggled…
It was also at this point that I began using indexing software, something which transformed the procedural aspects of indexing. One of the major advantages from the ILR point of view was the possibility it offered of preparing the index to an individual volume within the existing consolidation and then extracting it, with cross-references added as appropriate. One consequence of this is that individual indexes may sometimes seem curiously structured or oddly balanced. In a volume dealing solely with “Pinochet” cases, or war crimes trials, it would normally be right to make a number of contextual assumptions. But each individual volume of the ILR needs to be seen as part of a whole, a whole which the indexing process seeks to reflect.
Consistency
Working on the basis of the consolidated index makes it possible to retain a degree of control over the consistency of headings, a task which is helped beyond measure by a number of tools today's indexing software offers, each upgrade offering yet more sophistication and I no longer have an excuse for distributing the same material indiscriminately between alternative headings.
Of course, a passion for form and tidiness must not be taken too far. Some ideas may be close but not identical, or the identity insufficiently certain, perhaps because the precise meaning of the original text is unclear, to justify amalgamation. Slight variations of usage between English-speaking jurisdictions, changing usage over time, and problems of translation can make consistency something of a mirage. “Competence” and “jurisdiction”, “interim measures” and “provisional measures”, “sovereign immunity” and “State immunity” probably require a choice to be made between them, and what about “ambassadors”, “envoys” and the like? Do they deserve their own entries or should they be subsumed in entries picking up the language of the Vienna Convention on Diplomatic Relations (VCDR), now the received language of the courts.
Changing usage over time presents the serial indexer with a particular problem. Is it helpful to keep the original, anachronistic terms or should their modern version be adopted? One wants to keep the flavour of the past but on the whole it probably provides the best service to today's user to adopt modern terminology where there has been a seamless transition over the years, and simply cross-reference from earlier forms. With codifying conventions such as the VCDR, I limit entries under the convention itself to procedural matters related to the convention such as ratification. Subject matter goes under the subject matter heading and I include under that heading material pre-dating the convention which clearly falls within the provisions of the convention, with the warning that there is a degree of anachronism here. And I include the convention article number, as also the article number in legislation, because I have discovered over the years that this is by far and the best way of identifying precisely the issue at stake.
Translation problems
Translated material poses its own problems, particularly where the translator is not a legal specialist or where the concepts under discussion do not have an English equivalent. It may need a leap of the indexer's imagination, if the translator has failed in his task to decide exactly where, in an English index, to lodge a particular item. The problem is at its most glaring in dealing with legislation. “penal” and “criminal” are clearly interchangeable – and alphabetically a long way apart. The “Law on the Organisation of the Judiciary” may also be the “Constitution of the Courts Act”, the “Courts Act” or the “Judicial Organisation Act”, to name but some of the options. Often, the only way of determining whether items are one and the same is by reference to their title or abbreviation in the original language, in the given example, the acts turned out all to be “OJ”, but in many cases this vital information is missing, either omitted altogether, or the proffered abbreviation being derived from the English translation. I try to retain the original language abbreviation and to achieve as much identity of translation as possible. This is a counsel of unattainable perfection, but the practice of bringing all legislation together under “law of”, of pushing “act” or “law” to the end, and giving an indication, wherever possible, of the substance of a law known only by its number, has helped to eradicate some of the previous scattering, and reduced the field of search to a manageable few inches of sub-heading rather than what may be the pages of a whole country entry.
Cross references
“[O]rder in some respects not so good, that a man may think a place is missing, when it is only put in another place”: Samuel Pepys on Newman's Concordance, (Diary, 8 June 1663).
By and large, things in an index are not in the wrong place. It is simply that they may not be where a particular user expects to find them. Some of the reasons why this should be so have been outlined above. The problem can largely be resolved by generous cross-referencing. The hard pruning of the original ILR index headings to give the consolidation coherent shape has meant many of the old headings are no longer used. However a high proportion of them are retained, with a cross reference to where material once listed under them may now be found. My rule is quite simple. If anyone seeks an item in the “wrong” place, add that “wrong place” with a cross reference.
Implicit headings
When indexing a monograph, one can assume the subject-matter of the book as a main heading, and for most purposes start one or two levels down. This has also occasionally been the case with individual ILR indexes, where a volume has been largely devoted to a single case or to a limited range of issues, but in a consolidated index, this approach only makes sense if one happens to know that the volume under reference was a “single-issue” volume, as for instance with the South West Africa cases so I normally provide that missing main heading, unnecessary in the monograph but crucial in a multi-subject context.
Another trap it is also all too easy to fall into is to forget that detached from their context, it may not be obvious to which countries entries refer. There have been several examples over the years of particular pieces of legislation appearing as main headings, or, likewise, legal concepts, peculiar, perhaps, to the International Court of Justice, or the European Court of Human Rights, or to the United Kingdom or the United States, the reader being expected to supply the missing information. I try not to build in assumptions and, in particular, not to assume that the user is a British lawyer, but to make it all comprehensible regardless of the user's background and regardless of whether the individual volumes are available for elucidation.
Added value
It has come as a surprise to me as an indexer how much time has to be spent on research. Dictionaries in several languages, a good atlas, The Statesman's Yearbook, UN resolutions, texts of treaties and legislation, much now available, thank goodness, on the web, are all matter for frequent consultation, to resolve ambiguity or to supply essential information quite absent from the original text as, for example, when a UN resolution is identified only by number, or reference made to “the Constitution” when it may be vital to know which of half a dozen constitutions is under discussion. Most difficult of all has been to deal with countries which have changed their name, status, boundaries, form of government, sometimes several times. Where the only differences are in name and form of government, I use the most recent version with cross references from earlier ones. If the change has been more substantial, entries are made under all the variants according to which was in operation at the relevant time. Useful items of information, such as dates of treaties, are incorporated into the index.
Being selective
Being selective in the case of the ILR index is not so much about trying to save space as to serve the user best, concentrating on what is most likely to be helpful to him or her, so only points of international law as discussed by the court or tribunal are indexed. Facts and parties' arguments are indexed only in so far as this is essential for an understanding of the legal argument. The summaries to the cases are not indexed except to the extent that they contain relevant legal material not included in the report of the case itself. Points relating solely to issues of municipal law not relevant in the international law context are not indexed.
The same material is not indexed repeatedly. Where a case which has already appeared in the ILR is cited at length in a subsequent case, it will not be re-indexed, but simply subsumed in the indexing of the later case, with occasionally, where the case is particularly important, its name as a subheading.
What next?
The immediate task
As I said earlier, work is in hand on a consolidated index to Volumes 1–150, the indexing of new individual volumes proceeding simultaneously with the indexing of Volumes 1–35. Why not speed things up a bit by adopting the same approach with these early volumes as with Volumes 36–80? Well, I tried that and found that terminology and indexing styles between then and now had changed so much that it simply could not be done this way.
Why bother?
One of the exciting recent developments in the ILR context is their availability online at Justis (www.justis.com/ilr). Do we then still need an index? To make my own position clear, I am no Luddite as far as indexing and technology is concerned. I always have the most up-do-date version of my indexing software and am usually beta-testing the next. I Google endlessly and love the possibilities offered by the latest version of Adobe Acrobat which allows me, for example, to pull together all occurrences of the same word or phrase in a given text or series of texts. And I do not rule out the possibility of automatic indexing systems developing beyond their present rather primitive stage. But none of this goes to the heart of the matter. If you know precisely the phrase you are looking for, a free text search will find it, indeed will find every single instance of it, but totally without discrimination. I hope I have said enough in the preceding paragraphs to indicate that a human indexer, supported by sophisticated technology, will go far beyond that. Certainly indexers do index names and titles, even that, as I have explained, may require quite a lot of value-adding, but even more they are engaged in the task of indexing the concepts which underlie the words on the page, getting to the heart of the matter.
Zhag Qiyu, a leading Chinese indexer, has given the golden rules of indexing:
“Choose your terms well, with respect for what's in the document and for the needs of the user. Identify the relevant, exclude the superfluous, spot the unsaid, make the connections, order it all in such a way as to catch the user's attention and you will have achieved your goal: the creation of an index that will take users easily and directly to the information they are seeking from whatever the point at which they begin the search”.Footnote 5
That is certainly my aim in indexing the International Law Reports: and what makes it worth it. Others tell me they agree.
