Readers might remember that one strand of decolonising our collections in response to Russia’s war against Ukraine, as outlined in an earlier blog post, was about classification.  As explained then and long known by readers using our open-shelf collections, large parts of the UL’s classification system still strongly reflect the times and attitudes of empire.  There’s a lot of work to be done here just to tease all the various threads out.

Taking the focus back to Ukraine specifically, I have taken a preliminary look at the Ukrainian component in the “Russian literature” classes – 756 and 757.  These classes, meant to contain Russophone literature only, was in practice also the destination for Ukrainophone literature too until the introduction in 2011 of a separate class (758:6) for the latter.  There was always a different classmark for “Other Slavonic” (758:8) for languages without their own classmark, but unfortunately Ukrainian appears to have been placed standardly in Russian for decades.

Today’s initial work has been to work out what at least roughly what amount of books it is that we might potentially move, reclassify, and re-label.  Here are the initial results.

• 756 contains 252 titles in or translated from Ukrainian
• 757 contains 190 titles in or translated from Ukrainian

So far, so relatively straightforward, if still representing quite a lot of work (I think it would be a challenge to deal with one book in 10 minutes, given all the things that would need to happen, so those figures alone would mean 2 weeks full-time as a minimum).  What is missing here, though?

Firstly, I said “titles” deliberately, rather than books, because 1 title might be for a single book or for a multi-volume set.  When I ran the same search for separate item records (which would find extra volumes), the results were:

• 756 Ukrainian = 372 items
• 757 Ukrainian = 396 items

Secondly, we must not forget the books that a search like this using our library management system can only work with the metadata it contains – and we know that the metadata is imperfect.  For a start, my search depended on each title’s catalogue record having the language of the book represented accurately.  What results did I get when I ran the same search but swapped the language from Ukrainian to “undetermined”?

• 756 has 249 titles with language as “undetermined”
• 757 has 364 titles with language as “undetermined”

Most of these will probably be Russian, but it’s still a lot to explore.  Plus I also know from experience that there will inevitably be some Ukrainian titles that have been incorrectly coded by project cataloguers in the past without sufficient language expertise as Russian – which leads me to realise that a proper tackling of the 756 and 757 sections to identify Ukrainian material held there would involve going through the tens of thousands of books in these classes at the shelf.

I should also explain that most of the “undetermined” results are for another UL legacy problem – volumes in multi-volume sets that have their own record with either their volume classmark or the dreaded “Record in non-Roman script” label appearing as the record’s title.

These records are legacy fallout from decades before, when two separate systems came together: the main catalogue system (standard bibliographic records) and the borrowing system (individual item records).  Most of the time, they are annoying for the reader but not catastrophic (hence staff chipping away at them when they come across them in their standard work rather than these records being the focus of a huge project), since there is usually a “proper” record for the set as well as the individual volume records, so the reader would still be able to find what they are looking for in pretty much the standard way.

What next?  Some metadata action but also a fair bit of talking.  We can at least make quite a few corrections to the metadata in the meantime, trying to tie up separate volume records for Ukrainian titles with their set records.  I’ll start conversations with various groups about the work to be done and any related work (the open shelves have long struggled with overflows across the UL, so there are likely to be other projects in the pipeline that might have some impact on these classes too).  There is also broader work to be done on the vocabulary of the UL’s classification schemes.  We should consider a shift from eg “Russian” to “Russophone” and should start to address issues like the fact that the literature classes for languages with European roots is called “Western literature”…

Decolonisation work can be challenging but is always interesting – and pulling one thread inevitably pulls out many, many more.

Mel Bach