Upgrade StorageNew Zealand Law Style Guide
Hello,
I am trying to develop a citation style for the New Zealand Law Style Guide. I have researched extensively to try and find a reference management programme that can deal with the complexities of legal citations with little luck. Zotero seems to offer more flexibility than other programmes to create styles. I have been reading the forums on the development of the Canadian legal citation and think New Zealand needs something similar. I wonder if the authors of the Canadian style can offer me any advice? I'll detail a few of the difficulties I have identified so far.
The forum for the Canadian style can be found here: http://forums.zotero.org/discussion/9489/style-request-canadian-uniform-legal-citation-mcgill-guide/
The New Zealand Law Style Guide is very similar to the Canadian legal style and it is available online at http://www.lawfoundation.org.nz/style-guide/index.html . Like the Canadian style, there is a much wider array of resources than most styles allow for:
Types of resources that Zotero recognises:
Books, Journal Articles, Legal Encyclopaediae, Laws of New Zealand (a specific encyclopedia), Unpublished papers, Internet Materials, Newspaper Articles, Interviews, Press Releases, Speeches, Letters and Emails, Reports
Types of resources that Zotero doesn't recognise:
Treaties, United Nations Materials, Official Documents of the WTO, Official Documents of the GATT, Hansard (Parliamentary Reports), Appendix to the House of Representatives, Submissions to Select Committees, Standing Orders, Cabinet Documents, Cabinet Manual, New Zealand Gazette, Law Commission Reports.
CASES:
Cases are difficult to code in CSL. We need to be able to provide a jurisdiction or authority for the reference. Cases can be referenced in many different ways depending on the jurisdiction or court. Without being able to search the value of a field I see few ways to output different formatting depending on the type of case.
Examples:
Reported case where report series is organised by year (and typically has no volume number) = 1
Reported case where report series is organised by volume number = 2, 3
Neutral citation = 4
Unreported cases - file number citation = 5
Canadian case with neutral citation = 6
Scottish reported case if report series is organised by year = 7
United States Cases = 8
Maori Land Court and Maori Appellant Court decisions = 9
International Court of Justice and Permanent Court of International Justice = 10
European Community Cases = 11
1. Case-name [year] volume report-series starting-page.
2. Case-name (year) volume report-series starting-page.
3. Case-name (year) volume report-series starting-page (jurisdiction).
4. Case-name [year] court-identifier judgment-number.
5. Case-name court-abbreviation registry file-number day-month-year.
6. Case-name year court-identifier judgment-number
7. Case-name year report-series starting-page
8. Case-name volume report-series starting page (jurisdiction year).
9. Case-name – block-number (year) minute-book-reference (citation).
10. Case-name (parties) (phase) [year] volume publication starting-page.
11. Case-number Case-name [year] report-series page.
What I can do:
Format the case field to cite the generic reported case style. If the volume field is blank, use square brackets. If the volume field is filled, use round brackets.
For all other cases, put the whole citation in one field (not ideal).
Put all other items that Zotero doesn't recognise in the generic "document" field (also not ideal).
Further Problems:
1. The NZLSG uses minimal punctuation. This means stripping periods from all names, journals etc. Currently when you strip periods, a space is automatically inserted.
This works fine for journals such as NZ.L.Rev (which becomes NZ L Rev).
It doesn't work, however for journals such as V.U.W.L.R (which becomes V U W L R). The NZ style removes all spaces from this type of reference (VUWLR). The same problem occurs when you strip periods in name. The NZLSG requires that there are no spaces between initials (e.g. HLA Hart, or AV Dicey).
One way to fix the problem for journals would be to have a plugin for journal abbreviations. The NZLSG has an inflexible list of the correct abbreviations to be used for all NZ journals. I'm not aware of any plugin that exists at this stage.
This wouldn't fix the problem for name. Is it possible to remove the space between initials? The only way I seem to be able to do it is to use "initialize-with" which seems to delete the unwanted spaces, but converts all first names to initials.
If anyone has any more information on writing citation styles for legal references, it would be great to hear from you.
- J
I am trying to develop a citation style for the New Zealand Law Style Guide. I have researched extensively to try and find a reference management programme that can deal with the complexities of legal citations with little luck. Zotero seems to offer more flexibility than other programmes to create styles. I have been reading the forums on the development of the Canadian legal citation and think New Zealand needs something similar. I wonder if the authors of the Canadian style can offer me any advice? I'll detail a few of the difficulties I have identified so far.
The forum for the Canadian style can be found here: http://forums.zotero.org/discussion/9489/style-request-canadian-uniform-legal-citation-mcgill-guide/
The New Zealand Law Style Guide is very similar to the Canadian legal style and it is available online at http://www.lawfoundation.org.nz/style-guide/index.html . Like the Canadian style, there is a much wider array of resources than most styles allow for:
Types of resources that Zotero recognises:
Books, Journal Articles, Legal Encyclopaediae, Laws of New Zealand (a specific encyclopedia), Unpublished papers, Internet Materials, Newspaper Articles, Interviews, Press Releases, Speeches, Letters and Emails, Reports
Types of resources that Zotero doesn't recognise:
Treaties, United Nations Materials, Official Documents of the WTO, Official Documents of the GATT, Hansard (Parliamentary Reports), Appendix to the House of Representatives, Submissions to Select Committees, Standing Orders, Cabinet Documents, Cabinet Manual, New Zealand Gazette, Law Commission Reports.
CASES:
Cases are difficult to code in CSL. We need to be able to provide a jurisdiction or authority for the reference. Cases can be referenced in many different ways depending on the jurisdiction or court. Without being able to search the value of a field I see few ways to output different formatting depending on the type of case.
Examples:
Reported case where report series is organised by year (and typically has no volume number) = 1
Reported case where report series is organised by volume number = 2, 3
Neutral citation = 4
Unreported cases - file number citation = 5
Canadian case with neutral citation = 6
Scottish reported case if report series is organised by year = 7
United States Cases = 8
Maori Land Court and Maori Appellant Court decisions = 9
International Court of Justice and Permanent Court of International Justice = 10
European Community Cases = 11
1. Case-name [year] volume report-series starting-page.
2. Case-name (year) volume report-series starting-page.
3. Case-name (year) volume report-series starting-page (jurisdiction).
4. Case-name [year] court-identifier judgment-number.
5. Case-name court-abbreviation registry file-number day-month-year.
6. Case-name year court-identifier judgment-number
7. Case-name year report-series starting-page
8. Case-name volume report-series starting page (jurisdiction year).
9. Case-name – block-number (year) minute-book-reference (citation).
10. Case-name (parties) (phase) [year] volume publication starting-page.
11. Case-number Case-name [year] report-series page.
What I can do:
Format the case field to cite the generic reported case style. If the volume field is blank, use square brackets. If the volume field is filled, use round brackets.
For all other cases, put the whole citation in one field (not ideal).
Put all other items that Zotero doesn't recognise in the generic "document" field (also not ideal).
Further Problems:
1. The NZLSG uses minimal punctuation. This means stripping periods from all names, journals etc. Currently when you strip periods, a space is automatically inserted.
This works fine for journals such as NZ.L.Rev (which becomes NZ L Rev).
It doesn't work, however for journals such as V.U.W.L.R (which becomes V U W L R). The NZ style removes all spaces from this type of reference (VUWLR). The same problem occurs when you strip periods in name. The NZLSG requires that there are no spaces between initials (e.g. HLA Hart, or AV Dicey).
One way to fix the problem for journals would be to have a plugin for journal abbreviations. The NZLSG has an inflexible list of the correct abbreviations to be used for all NZ journals. I'm not aware of any plugin that exists at this stage.
This wouldn't fix the problem for name. Is it possible to remove the space between initials? The only way I seem to be able to do it is to use "initialize-with" which seems to delete the unwanted spaces, but converts all first names to initials.
If anyone has any more information on writing citation styles for legal references, it would be great to hear from you.
- J
Welcome to the family. :)
The legal styles are indeed a particularly thorny problem, but some coordination is beginning to emerge in our response to the underlying issues. I'll write a bit later about some of the details, but for a start I'll just flag that I have a scheme in mind for coping with jurisdictional variations in cite forms, based on the URL:LEX (link is to current draft of the specification). This is still a draft standard, but (crucially) it implements a concept of rule-making authority that extends beyond national governments to include international organizations. If the CSL processor is given the ability to discriminate on the basis of the issuing authority, we should be able to cope with the variations without any real difficulty.
I've tried using initialize-with="" but it currently forces all names to be initialised.
Input names are of variable quality: some with full given names, and others with only initials provided. Initialization control is lost when the desired output is the full name but this is not available.
Input name variations:
John Fitzgerald Kennedy
J F Kennedy
J. F. Kennedy
JF Kennedy
J.F. Kennedy
I suggest the existing initialize-with property be replaced by two properties:
1. initialize-with="delimiter". Which describes the method of initialization. Sets the delimiter for initials.
2. initialize="true" or "false", defaults to true if initialize-with is set, otherwise false. When true, initialization is forced on all names irrespective of input style.
This retains the current meaning of initialize-with if used in isolation but allows for control over initials in incomplete name parts.
If it is of use to some styles, the initialize tag could have additional values to allow finer brained control, such as initializing only given names after the first name eg John F. Kennedy.
Is there anything I can do for now to create a wider variety of cases? Most domestic cases are cited using the same basic format and the "choose" feature allows the style to produces accurate citations for examples 1-5 above, depending on blank fields. I'm having a hard time with USA cases. They are cited completely differently but use all the same fields. Is putting the whole citation in the "case name" field all I can do?
More fine-grained programmatic control over name formatting probably won't be happen, since authors often have preferences concerning how their name is to be represented in its full form, and styles often provide that author preference should be followed in that case.
On the formatting of initials, would it be a correct description to say that spaces between lone initials are closed, while spaces between a lone initial and a full word or abbreviation are not?
I've been using the initialize-with="" function to remove unwanted spaces when I strip periods from names. The problem is that I don't want to use initials when a full name is given. If a full name is not available I need the initials to read without spaces.
What I meant about losing initialisation control is that it seems you can either use the full name or have initials without spaces, but not both depending on what form of the name you have been provided.
Example:
1. Kennedy, John = John Kennedy
2. Kennedy, J. F. = JF Kennedy Yes that would be the most consistent with the style guide. Dropping the middle initials would also be a possibility. I don't think this would work in every case. For example, N.Z. L. Rev (New Zealand Law Review) would be rendered NZL Rev instead of NZ L Rev. It might be cleaner to not insert a space when removing periods but leave any other spacing. This may also raise problems when used with other standard variables.
Other examples I can think of which would create problems would be F.C. L.H. Family Court Lower Hutt = FC LH and N.Z.L.C. P.P New Zealand Law Commission Preliminary Paper = NZLC PP.
- For names, strip-periods should aggressively collapse spaces between single initials; and
- In other contexts, strip-periods should remove periods, but leave explicit spaces in place.
I'm Jack Newberry, a real-life friend and admirer of caldwelljules. I don't study or practice law but I have an interest in computing and have been assisting Jules with this project.
You seem hesitant to suggest alterations to the CSL property tags. Changes to the interpretation of the tags are preferred. It seems likely that in trying to make legal citations work via CSL and citeproc this issue will arise often.
If you have a moment could you give us a description of the processes for changing these standards and how we can best assist.
Some factors I can see effecting the process include:
- Not creating too many too specialised CSL tags.
- Retaining backwards compatibility with existing CSL styles perhaps.
- Keeping tag names as self-explanatory as possible.
- Keeping the effects of tags consistent irrespective of what they're applied to.
- The ease of implementing any changes: ease of altering citeproc compared to CSL standard; and ease of coding the change.
Are there any important factors I have missed?
Also, I understand you have the salient role in citeproc-js. What sort of role do you take with the CSL standard itself?
The resistance to adding attributes isn't really a matter of principle; there would also be an argument that attributes should behave consistently across different contexts (which would be an argument against the solution I'm ruminating about above). But from the discussion so far, it seems likely (to me) that we can nail these use cases with the proposed behavior. If so, avoiding adding to the complexity the specification, at least in the short term, seems like a good thing; but that's only my own opinion. I think you can sign up for the xbiblio-devel list via this link, if you'd like to join.
I think the cleanest solution is what has been proposed above: introduce separate control over whether initialization occurs and which suffix is used for initials. That would also give us more flexibility, e.g. it would allow "Charles R Darwin" to be recast as "Charles R. Darwin". I've opened a ticket:
https://github.com/citation-style-language/schema/issues/60
Moreover, it is my ambition that because of the wide use, that ultimately people don't have to worry about writing or editing their own styles: that they're simply easily available.
So, as I think you already realize, this puts a high burden on stability and interoperability, both of implementation code, but also of styles themselves. This is definitely not a Zotero-only project.
In terms of how you might deal with this, I'd suggest just being careful to clearly separate use cases and requirements from implementation assumptions.
http://www.lawfoundation.org.nz/style-guide/nzlsg_18.html#6.1.2
After a little discussion off-list, we have some tentative answers. As Rintze mentions, the NZ L Style Guide appears to permit names to be uniformly initialized, in which case initialize-with="" covers the use case.
For full names, it seems that (for styles we have considered while pondering this) the following input would be portable across styles: "Alan J.P. Taylor". In the NZ L Style Guide, you could produce the output "Alan JP Taylor" from this using the same construct as the McGill Guide:
<macro name="authors"><names variable="author">
<name/>
</names>
</macro>
...
<text macro="authors" strip-periods="true"/>
(In the Bluebook style used in the US, "Alan J.P. Taylor" would be correct, so the same input can just be rendered as an ordinary full name there.)
From all that, it appears that unless further wrinkles come to light, the needs of the style can be met with existing options, so long as attention is paid to curating names in the database to eliminate spaces between initials.
Can I clarify that "strip periods" no longer inserts a space when taking out the periods? I see above that you mention that this is the case with the latest CSL processor but I can't get it to do this on my computer.
It is true that using initials is consistent with the style guide but it is general university practice that full first names are preferred where available. I would have to check with the authors of the style guide on this one.
It would be preferable to have the flexibility to use Alan JP Taylor where available or AJP Taylor if the first name is not provided. This may already be the case with the current CSL processor, but like I said, my computer is still inserting spaces (A.J.P. Taylor = A J P Taylor).
Do you know if the change discussed here: http://forums.zotero.org/discussion/10313/ to have short-title as a variable will be included?
I am having the same problem when I want to use a short title for a case name. There are some cases with common names which use a short title in the citation.
E.g. Simpson v Attorney-General [1955] NZLR 211 [Baigent's Case].
Currently when I use the "title form=short" function, it inserts the full case name if there is no short title entered.
1. initialize-with="delimiter". Which describes the method of initialization. Sets the delimiter for initials.
2. initialize="true" or "false", defaults to true if initialize-with is set, otherwise false. When true, initialization is forced on all names irrespective of input style.In an off-forum discussion, Rintze convinced me of the virtues of your proposal to add an "initialize" attribute that can be set to false, as a way of normalizing the punctuation of initials in full names.
I've just completed the necessary coding in the processor, and will be checking it in shortly. We have a proposal covering the new attribute in the CSL tracker, and with luck it will be approved in the not too distant future.
Thank for your input; this is a valuable addition to the CSL quiver.
I seem to be unable to use "entry-encyclopedia" as a reference type. There has been some discussion on this forum http://forums.zotero.org/discussion/9812/encyclopediaarticle-dictionaryentry/ that entry-encyclopedia is still mapping to chapter. Can you confirm if this is the case? Otherwise there is something funny going on with the code that I am not aware of.