Don’t dream it’s over; or, A return to cataloguing

A double rainbow joins the National Book Church™ and the National Science Temple™ against a backdrop of dark clouds, November 2018. Photograph by the author

A handful of Wednesdays ago I quit my job at a mildly prestigious library that shall remain nameless, after just over three years of employment. I wore my favourite cataloguing-themed t-shirt to work, bought one last book with my staff discount at the bookshop, and treated myself to a final helping of bain-marie slop at the cafeteria across the road. It still doesn’t seem entirely real that I’ve left. I still had so much to do.

The last six months have been the happiest and most fruitful of my entire career. I’ve absolutely loved being a systems librarian. I’ve had great fun crafting Access queries, running Perl scripts, devising Excel macros and more, while running complex data reports and conducting bulk data edits for business areas. I learned a heck of a lot about how data and systems work together (or not). But more than anything I’ve really loved my team. They’ve been wonderful people to work with, and I wish them every success.

I was a little surprised by how much time I spent saying ‘thank you’ during my last week. I’m not sure I expected to feel quite so grateful at the conclusion of my time there, but I guess I had a lot of complex feelings about the whole thing. Besides, it turned out I had a lot of people to thank: my wonderful boss Julie, my colleagues Sue and Brad, my director Simon, my previous director Libby, my old boss Cherie, good people like Ros H and Ros C and Catherine.

I wanted to finish that job feeling like I achieved something of lasting value. Instead I settled for starting something that will outlive me and hopefully become standard practice. Sure, helping implement a new service desk ticketing system was useful from an internal workflow perspective, but it’s not quite what I went there to do. Instead I called a meeting with a bunch of managers (well, my boss called it for me) to highlight several pieces of egregiously and systematically racist metadata in our catalogue, mostly relating to Indigenous Australians. Some of the old subject headings hadn’t been updated to the current terminology, while other headings should never have been in our catalogue in the first place. I outlined how my team could remediate these problems, but some policy decisions needed to be made first—ideally by those attending the meeting.

It’s a shame this meeting wound up happening on my last day. But the looks on the faces of my Indigenous colleagues convinced me I was doing the right thing. These terms should have been nuked from the catalogue twenty years ago, but the next best time to do that is now. I kinda felt like this shouldn’t have been up to me, a systems librarian, telling a roomful of people who all outranked me how to fix a data problem. But it needed doing, and I was in a team that had the technical ability to make the necessary changes. I regret that I won’t be around to see them happen.

Shortly after this meeting my director Simon was regaling us with an anecdote about longitudinal datasets; he has a background in statistics and often compares library metadata to things like the HILDA survey. But the key difference is that while HILDA’s questions and expected answers have changed over time in a discrete fashion, making it easier to see where such changes have occurred, library metadata corpora are a total mismash of standards and backgrounds, with each MARC field potentially having been added at a different time, in a different socio-cultural context, for a different purpose. Metadata librarians are grappling with the ongoing impact of data composition and recording choices made decades ago. We have virtually no version control (though it has been suggested) and little holistic understanding of our metadata’s temporal attributes. It makes retrospective #critcat efforts and other reparative description activities a lot harder, but it also hinders our ability to truly understand our descriptive past.

I was pleased to end my time there on a constructive note. But like I said, I have a lot of complex feelings about the last three years. I started out being one of those bright-eyed and bushy-tailed new professionals who didn’t have a whole lot else going for her, wanting to prove her passion and devotion to her dream job by working herself to death, thinking that maybe her job would start to love her back. Please don’t do what I did. I might not have realised at the time how harmful this mindset is, but I also did not realise that I deserved better from an employer. Whenever I think about my time there—barring the last six months—I can’t shake these feelings of deep unhappiness. I feel like I was thrown in the deep end right at the start and spent years desperately trying not to drown. I started thinking nobody would care if I did drown. I was lucky that the restructure threw me a life raft, but the damage was done.

Happily, I have much more to look forward to now. After a pandemic-induced false start I’m finally fulfilling a lifelong dream to move to Victoria, to be closer to friends and forests, and to take up a role as a metadata team leader at a regional academic library. Professionally I feel like I’m returning to my metadata spiritual home, and I like having the word ‘quality’ in my new job title. The ‘team leader’ part is slightly intimidating though—I have no supervisory experience whatsoever (and they know that) but it’s something I’m very keen to do right. Everyone I’ve met there so far has been really lovely. I can’t wait to start next week.

I’m glad to be ending this rather turbulent chapter of my life and beginning a new, hopefully calmer one. I took this job for many reasons, but I keep coming back to the potential I sensed in it. There’s so much possibility here. It’s very exciting.

Indigenous names in authority records: the case of Jandamarra

UPDATE (20 June 2018): This has now been addressed! I noticed a couple of days ago that the NLA have updated this authority record, such that it now follows the AIATSIS example, “Jandamarra, approximately 1870-1897.” (The Libraries Australia heading and Trove duplicate issue both remain, but I understand they are managed by different areas within NLA.) While I was not directly informed of the NLA’s decision to update Jandamarra’s authority record, I am thrilled that they have done so. Thank you, NLA cataloguers, for making this necessary change.

A selection of books about Jandamarra. (Picture courtesy AIATSIS)

It’s all well and good for librarians to talk about decolonisation, but we need to put our money where our mouths are. Cataloguers are no exception—we decide how resources are described and accessed. We dictate the effectiveness of a search strategy. We alone have the power to name.1

Being the sort of person who browses library catalogues for fun, I wound up on a NLA record for a play about Jandamarra, the Bunuba resistance fighter. Except the subject headings in this record didn’t name him at all. Instead they named some bloke called ‘Pigeon’.

100 0# $a Pigeon, $d -1897
400 #0 $a Jandamarra


Apparently ‘Pigeon’ was a name given to Jandamarra by a white pastoralist.2 The 15 books held by the NLA with this subject heading overwhelmingly refer to a man named Jandamarra. His entry in the Australian Dictionary of Biography is under this name. The Canberra street named in his honour uses this name. Why doesn’t the NLA use this name?

The whole thing couldn’t be more colonial if it tried.3 A colonial institution (the library) referring to an Indigenous man by a colonial name (‘Pigeon’) and qualifying it with his year of death in a colonial calendar (1897). Jandamarra’s authority record represents his name and life as it was known to white people. How would the Bunuba describe him? Would they use the name ‘Jandamarra’ at all? What could a more culturally appropriate authority record look like? How might we disambiguate people without reference to colonial names, occupations or calendars?4

100 0# $a Jandamarra $c (Bunuba man)
400 #0 $a Pigeon, $d -1897

As a white cataloguer I firmly believe a change needs to be made. Such a change would have greater impact if it were also made in Libraries Australia, the national union catalogue (on which more later). The name ‘Pigeon’ is also used there, identically to its use in the NLA’s catalogue.

Let’s fix that

The issue then becomes: how might I make this change? More importantly, how might the community suggest a change? There is a way to suggest changes to name headings on the ANBD, but it’s very difficult to find—the Libraries Australia reftracker includes two options for ‘Propose a LCSH change’ and ‘Propose a new LCSH’ (where ‘LCSH’, apparently, includes all headings, name and subject alike). The form states that the info you provide goes straight to LC (that is, it’s not evaluated locally). It also immediately starts demanding my name, my NUC code, tells me to choose my 1XX heading, include 670 source citations, LC pattern or SCM memo, use for, broader term, related term??

I am a fluent MARC speaker and I know a 680 when I see one, but I have never dared fill out that form. I can’t see how an ordinary person would ever be able to suggest a formal change for Australian usage. Crowdsourcing initiatives like Violet Fox’s Cataloging Lab (which, for the record, I love), are necessarily US-centric and wouldn’t immediately address a local problem. Besides, the guidelines for establishing name authorities in the ANBD expressly state that Australian entities are exempt from the ‘let LC decide’ policy.

Besides, even if we were able to navigate the form and suggest a change, what would the change be? For guidance, I looked to AIATSIS’ catalogue. Sensibly, and in delightful accordance with RDA, they have opted to use 100 0# $a Jandamarra, $d approximately 1870-1897 as the preferred form. I figure if it’s good enough for AIATSIS, it’s good enough for me.

Wondering what other libraries used, I then looked at the Library of Congress’ NAF (Name Authority File) record. To my surprise, they used a different spelling:

100 0# $a Sandamarra, $d -1897
400 #0 $a Jandamarra, $d -1897
400 #0 $a Pigeon, $d -1897
400 #0 $a Tjandamara, $d -1897

This record was created in 1989 and revised in 2013. A search of both LC’s catalogue and WorldCat (via the Libraries Australia Z39.50 interface, in case that makes a difference) brought up no results with this spelling, so I couldn’t determine if a particular work was used as its basis. Usually these works would be recorded in a 670 field, but these had nothing.

It would not be beyond LC to update its heading to the more commonly-used spelling. Pleasingly, they have form in this area: in 2003, LC changed several dozen subject headings relating to $a Aboriginal Australians (or $a Australian aborigines, as they were then described) in consultation with the NLA.

What’s in Trove?

I then found myself browsing the Trove People and organisations zone, where authority records are given a new life as sources of biographical data. Like other parts of Trove, the P&O zone aggregates and incorporates data from a variety of sources. I was therefore surprised to find Jandamarra listed under this name, using data from AIATSIS and the Australian Dictionary of Biography; as established above, both sources used the most commonly-known spelling. Notably, this did not include data from Libraries Australia:

Record for Jandamarra within Trove’s People and organisations zone.

The great thing about Trove identity records is that they display the ‘Also known as’ data (or UFs, or non-preferred terms, or 4XX fields, or whatever). It’s really hard to get an ILS to display this info, especially in an easy-to-read format like Trove has done. I’m really pleased to see this data out in the open and not hidden down the back of the authority file sofa.

Now, what happens if I search the P&O zone for ‘Pigeon’?

Jandamarra appears twice, with a slightly different spelling

We see that Jandamarra (-1897) is the first result, but the fifth is for Jundumurra, Pigeon (?!), which features data from AIATSIS and Libraries Australia. (This particular LA record pulls its data from AIATSIS anyway, so strictly speaking this isn’t the NLA’s fault, but it’s still a dupe that LA and/or Trove would have to merge.)

Interestingly, the original authority record from the NLA (‘Pigeon’, remember him?) doesn’t appear to be represented in the P&I zone at all. I wonder if that was a conscious or unconscious decision?

For completeness, here’s the real AIATSIS name authority, which in my view is also the best one:

100 0# $a Jandamarra, $d approximately 1870-1897
400 #0 $a Jundamurra, $d approximately 1870-1897
400 #0 $a Sandamara, $d approximately 1870-1897
400 #0 $a Sandawara, $d approximately 1870-1897
400 #0 $a Tjandamara, $d approximately 1870-1897
400 #0 $a Tjangamarra, $d approximately 1870-1897
400 #0 $a Jandamura, $d approximately 1870-1897
400 #0 $a Pigeon, $d approximately 1870-1897
400 #0 $a Wonimarra, $d approximately 1870-1897

It turns out that Jandamarra has three (!) name authority records in Libraries Australia, one from the NLA and two from AIATSIS. Ordinarily I would consider this a major data integrity issue, and 100 10 $a Jundumurra, Pigeon is a bit of a problem, but for the moment I’m actually okay with the other two full-level records, because they help illustrate the differing approaches and mindsets from the two institutions. In time, I’d like to narrow that down, though.


In short, here’s what I would like to see happen so that Jandamarra is referred to by his rightful name in the ANBD, and in catalogues that use ANBD records:

1) Libraries Australia to modify their name authority record and establish the preferred form as $a Jandamarra, $d approximately 1870-1897, in accordance with that used by AIATSIS, and add non-preferred forms as appropriate. This change could then ripple across to the NLA’s catalogue, and other libraries that use Libraries Australia authorities would eventually follow suit. Maybe a little publicity around the change—after all, it’s being done for the right reasons.

2) Trove to merge the two identity records such that Jandamarra appears only once, that ‘Pigeon’ appears under the ‘Also known as’ list (so those who know him by that name are redirected accordingly), and that the sources of data encompass AIATSIS, Libraries Australia and the National Dictionary of Biography.

Such moves may seem small, but they would represent a sincere and concerted effort to decolonise the authority file. Cataloguers can, and should, restore the power to name to Indigenous communities, especially where colonial names have been used to describe Indigenous people and concepts. A name is not the cataloguer’s to take—it is the community’s to give.

  1. Olson, Hope A. (2001). The Power to Name: Representation in Library Catalogs. Signs: Journal of Women in Culture and Society, 26(3), 639-668. doi: 10.1086/495624 
  2. Pedersen, Howard (1990). Jandamarra (1870–1897), Australian Dictionary of Biography, National Centre of Biography, Australian National University. Accessed 20 May 2018. 
  3. For more on the cultural sensitivities around Indigenous subject headings, see Kam, D. V. (2007). Subject Headings for Aboriginals: The Power of Naming. Art Documentation: Bulletin of the Art Libraries Society Of North America, 26(2), 18-22. doi: 
  4. See also Frank Exner, Little Bear’s excellent treatise on Native American names in the world’s authority files: Exner, Frank, Little Bear (2008). North American Indian Personal Names in National Bibliographies. In Radical Cataloging: Essays from the Front. Jefferson, NC: McFarland and Co.