Legal Citation Style – a Wicked Problem in Legal Information
The lack of uptake for citation management software programs, such as Zotero, EndNote, and RefWorks, by even tech savvy legal practitioners and scholars has puzzled me for some time. The absence of these programs or similar solutions is particularly surprising when one considers the large number of vendor supplied and internally customized labour saving solutions law firms implement in the interest of repeatedly saving small amounts of time and the institution wide licences many universities have implemented to encourage their use. As I started exploring the reasons for this absence, I found that there are many issues surrounding it, and that they combine to create a wicked problem – one that is not easily solved and continues to become more complex the more detail is known about it.
Citation management software programs generally provide the ability to import citation and sometimes full text data in an appropriate format downloaded from research tools, with some research tools having the ability to insert information directly without the steps of exporting from the originating research tools and importing into the citation software. This set of citations can then be used in conjunction with a word processing program to automatically insert and format citations in documents. These databases can be treated as document, topic, or matter specific; or for someone with ongoing specific interests, one collection of resources can be developed over time.
By automating citation management and formatting, these software packages save considerable time in research and composition, but there are many issues surrounding their adoption by the legal community. The first issue is the fact that few people know of its existence and the amount of time it saves. In a prior position where I managed large numbers of citations on a regular basis, I estimated that I saved so much time that the software paid for itself every time I used it.
However, in looking to implement the software, I found that the main research services in law, including CanLII, Westlaw Canada, QuickLaw, and others, do not provide materials in a format that would readily allow for importing citation information into the databases, and that they would require manual data entry or copying and pasting each citation element separately. Legal information is generally presented in a semi-structured format, which means that there are generally labelled fields signifying relevant dates, docket numbers, judges, court, etc., but they are not tagged consistently. This format facilitates human access in improved readability, but limits the use of automated access to the data because the relevant fields are not presented in a consistent format the programs can recognize. To make the use of citation management software feasible, data will need to be presented in a machine readable format, such as RIS, and ideally be available for download with documents from multiple sources within a research platform in a single file, rather than requiring each document or set of search results to be downloaded individually. At this time each of the research platforms provides some aspects of these requirements, but none of them make all of the elements for importing machine readable documents available.
The lack of a suitable document export format would not be a complete limit to someone determined to use this software, as it is possible to take files exported from many of the services and convert them into a suitable format using a programming software such as Sublime Text and regular expressions, but this is a rather difficult process unless someone is quite comfortable with computer coding.
The next problem with using citation management software is the fact that the Canadian Guide to Uniform Legal Citation prescribes a level of detail for citations that is not supported by any extant software package I am aware of. Not only are the software packages not able to address the nuances of legal materials, but there are types of legal materials, such as looseleafs, which were not considered by the developers of the standards for data format and citation type. This level of customization means that there is not a citation management software product on the market that would be a workable solution. Either a software company will need to develop a custom product for law or a sufficient number of dedicated volunteers will need to develop from scratch or customize an open source alternative, such as Zotero. It seems unlikely that the legal market in Canada is sufficiently large to make the development of a custom citation management software package viable for a software company, as they are not generally very expensive.
There are multiple possible starting points to resolve this issue, including rationalizing legal citation style, improving output formats and functionality for legal research tools, and exploring options for software development. I believe the most logical place to start is for publishers to develop output styles for their electronic services that are compatible with the software, as a citation style file that will handle the majority of Canadian primary and secondary legal materials can be created, and this would only require manual editing of those materials that fall outside the style. This is a problem that should be solved: it is in the interests of all members of the legal community who write regularly to have access to these time saving software packages, as this is a place where immediate efficiency improvements could be made with little expense or change in work style.
I think this is a great example of an area that has been largely ignored by information science researchers, and also one where there is tremendous potential for collaboration between information and computer science researchers, legal practitioners and vendors. There are likely numerous reasons to explain why there hasn’t been much research on law-specific information, while health and medical information, along with its researchers and practitioners, have been the subject of countless information science studies.
As we are all aware, legal information is fundamentally different than other kinds of information, both in content and the ways in which formatting and linkages between documents play critical roles in providing context and sense-making. I’ll go out on a limb and guess (without empirical evidence at my fingertips to back it up) that the ways in which we seek, classify and organize that information are different as well. I question whether the current tools available (and I’ll add Mendeley to the list) are suitable for some of the domain-specific information tasks of the legal profession. It may be more than simply convincing publishers to add metadata so that existing citation management tools can properly grab citations; it may require a new system that is uniquely our own.
Zotero is great at capturing bibliographic information for many kinds of non-legal documents, but it’s not particularly robust in its annotation or linking capabilities, both of which are so important in law. I think there is an incredible opportunity to develop a system that can capture bibliographic information at the paragraph/section level, but also allow for document storage, annotation and ways to clearly link and show relationships between specified legislation, case law and secondary sources, and all with a few simple clicks of a mouse, of course.
Perhaps a good topic for a PhD dissertation…
Since publication of this I have had this pointed out to me, which aims to solve some of these problems.
There is also this book also by Frank Bennett, which aims to solve these problems: Citations, Out of the Box: Adapting Zotero for legal and multilingual research.
Frank Bennett has done an amazing amount of work in “citation-friendly metadata” for law, developing Bluebook, McGill and other styles for Zotero over the last couple of years. Earlier this year, RefWorks issued a release of an output style for McGill 6th. Thomson/Carswell, I understand, plans to issue an update to Endnote that will incorporate the forthcoming McGill 7th output style.
In my experience and to the extent of my capabilities with them, Zotero and RefWorks work as well as can be expected for secondary legal materials—and only for secondary materials. I agree with Sarah and Kristin that the value of citation tools and reference managers can be so much better exploited for primary legal materials drawn from databases or open-web sources, including CanLII, so they are cited as what they are.
I look forward to that PhD dissertation. And perhaps it can include a chapter on the adjacent question of the continued necessity of the McGill style…
Unfortunately for both librarians and the legal community this a topic which having begun as a mole hill metastasizes into a mountain.
There are no fantastic “savings” to had by driving some of these bibliographic utilities into a quest for a complete hand in glove product. Law or the legal fight turns in part on the use of precisely identified parts of a whole wielded in full throated combat. Hard to believe this will change over the course of the next few lifetimes. But minor issues such as this sure drives drivel.