1. The author would like to thank Andrew Gifford and Tim Davies for all the information they kindly provided about the history of the Courtauld classification scheme. She would also like to express her thanks to Dr Julian Gilbey for his help and advice writing this article, and Derek Lee for his guidance. Finally, she would like to thank Antony Hopkins and all the staff at the Courtauld Institute of Art book library for their support.Google Scholar

2. Currier, Sarah, ‘Classification schemes in art libraries in the United Kingdom,’ Art libraries journal 27, no.1 (2003): 20. Around four fifths of the survey respondents were from the UK, with the other fifth replying from Australia, New Zealand and France.Google Scholar

3. Some libraries discussed in the body of the text are included in the statistics for standard schemes but are known to use a modified or simplified version of that scheme, such as the Courtauld Institute of Art and the Paul Mellon Centre for Studies in British Art (Currier, ‘Classification schemes’: 18-22).Google Scholar

4. Ferrari, Roberto C., ‘The art of classification: alternate classification systems in art libraries,’ Cataloging and classification quarterly 28, no. 2 (1999): 74.Google Scholar

5. Currier, ‘Classification schemes’: 18.Google Scholar

6. Oxford English dictionary, 2nd ed., s.v. ‘adapt, v.,’ http://www.oed.com:80/Entry/2110 Google Scholar

7. There are many reasons why the paper focuses on this particular aspect of adaptation. First, examining the original adaptation process would be difficult as no contextual documentation survives from this era. Second, examining the original adaptation outside of the Zeitgeist which produced it is of questionable usefulness. Third, Ferrari provides a detailed account of the process of creating adapted (‘alternate’) schemes for art libraries; therefore, this paper aims to complement his research by focussing on the next stage of the process (Ferrari, ‘Art of classification’: 74).Google Scholar

8. Library of Congress, Classification: class N:fine arts, 3rd ed. (Washington: Government Printing Office, 1922). From here onwards referred to in the text as ‘LCC (1922)’.Google Scholar

9. Davies, Tim, email to author, 27 May 2011; Gifford, Andrew, conversation with author, 18 May 2011; Gifford, Andrew, conversation with author, 25 May 2011.Google Scholar

10. Though there was no official ‘master copy’, it seems that the copy used by one of the deputy librarians served as a pseudo-master copy. Certainly, the other volumes were less meticulously updated and discrepancies and omissions between the three copies of the scheme were found (Davies, email). Different coloured repair tape was used to reinforce the joints of each of the three copies. Therefore, when describing a specific copy of CCS (1932), the colour of the tape will be used to identify the exact volume.Google Scholar

11. Davies suggests that it was probably the later half of 1999, but was definitely in the years 1999 to 2000 (Davies, email). One member of staff kept a pre-2008 copy of CCS, which is assumed to be a surviving copy of the 1999 version. This copy has been used as a source for research into the various formats of the CCS.Google Scholar

12. The permanent collections section of the Z schedule has a different history. These were transcribed earlier than the rest of the scheme, in 1990 to 1991 (Davies, email). Again, the computer file did not survive. However, in this case the printouts were scanned and OCR technology was used to create an editable Word file. This work was carried out by Sarah Hall (UCL placement student) in January 2009, under the supervision of the systems and services librarian and the senior cataloguer.Google Scholar

13. A long-term classification project commenced in 2008. The first phase of the project was to improve the scheme’s layout, update the geographic tables, provide instructions for number-building and identify obsolete or omitted classifications. It is through work on this project that many of the transcription errors described in the last part of this paper have been discovered.Google Scholar

14. Courtauld Institute of Art, Courtauld classification scheme, 1932 version; Courtauld Institute of Art, Courtauld classification scheme, 1999 version; Courtauld Institute of Art, Courtauld classification scheme, 2008 version. From this point onwards the various versions of the CCS will be referred to in the main text as CCS (1932), etc.Google Scholar

15. The annotations in CCS (1932) referring to photography appear to date from the creation of the scheme.Google Scholar

16. This was preferred to treating photography as an art form in its own right, creating a separate subclass akin to the NH photography schedule. See Lynda Bunting and Ani Matosian, ‘NH Classification Schedule for Artistic Photography,’ Revised ed. 2004, http://www.arlisna.org/resources/onlinepubs/NHschedule.pdf for more details about the NH schedule, and also see Bunting and Matosian’s article concerning photography classification and the NH schedule in this issue of Art libraries journal on pages 44–49.Google Scholar

17. As a concession, photography is divided between items concerned generally with photography (three-letter suffix for main entry) and those about a specific photographer (six-letter suffix, first three letters for surname of photographer and second three for main entry).Google Scholar

18. Unlike photography, installation art was not included in the original version of the CCS as the art form was ‘created’ much later. The precise date that installation art was included in the scheme has not been found in extant documentation; however Gifford recalls that it was discussed and decided between 1988 and 1990 (Gifford, conversation, 18 May 2011).Google Scholar

19. Library of Congress. Cataloging Policy and Support Office, Library of Congress classification: N: fine arts, (Washington D.C.: Library of Congress, 1996). From here onwards referred to in the text as ‘LCC (1996)’.Google Scholar

20. Interestingly, the three different annotated copies of the scheme (CCS (1932)) give slightly different descriptions of this classmark. Some give suggestions of examples of subject areas covered by the new classmark, such as semiology and psycho-analysis; alternatively or in addition, others give examples of theorists whose works would be found at this classmark, such as Freud, Marx and Barthes. One of the annotations specifically mentions that this classmark should be used for works ‘which don’t fit into the category of theory of art (Z7420 [taste] and Z7430 [theory])’ (CCS (1932), green copy). Also, it is clear from the typographical layout, variations in writing implement and handwriting, that the classmark was not part of the original adaptation and was instead added at a later date. A memorandum preserved within one of the annotated volumes documents the creation of this new classmark. (CCS (1932), memorandum ‘Z7439’). Though the year is not mentioned directly on the memorandum, contextual evidence suggests that it dates from 1986 to 1989 (Gifford, conversation, 18 May 2011).Google Scholar

21. LCC (1996) has been chosen for comparison with LCC (1922) in this paper, as this is the most recent edition held by the Courtauld. It is unlikely that the latest edition (2010) of LCC has influenced the CCS, as the Courtauld does not hold it.Google Scholar

22. Geographic tables assign numbers for specific countries; the countries are further divided into time periods, specific towns/cities and artists. The table value is then added to the number in the main schedule to form a classmark. There are four tables in the CCS which have different levels of detail. For simplicity, all examples in this section of the paper come from table four, which is the most detailed table and is used to create classmarks for history, general art, architecture, sculpture, painting and graphic arts.Google Scholar

23. Encyclopaedia Britannica, 2011, s.v. ‘Bukhara’, http://www.britannica.com/EBchecked/topic/83938/Bukhara. In the CCS, it is transliterated as ‘Bokhara’.Google Scholar

24. The illegibility of the annotated volumes is given as the reason why CCS (1999) was created (Davies, email).Google Scholar

25. All the errors mentioned in this paper appear in both CCS (1999) and CCS (2008). It is assumed that the errors occurred when the scheme was originally transcribed in 1999 and were copied across to the 2008 scheme; however, without documentary evidence concerning this transcription this is only an inference.Google Scholar

26. It is interesting to note that the typographical problems with CCS (1999) and CCS (2008) were so severe that CCS (1932) is now being used as the primary source for the Courtauld’s reclassification project.Google Scholar

27. As the CCS does not consistently assign classmarks for ‘other’, it cannot be assumed that an ‘other’ at the end of a section refers to the whole of that section.Google Scholar