In Svenonius’s The Intellectual Foundation of Information Organization, chapter 6, work languages are discussed. Work language parts are defined as well as how to create work languages, and relationships between the parts. Problems arise with gaps in AACR’s work language relationship types, and when choosing controlled vocabulary and work identifiers for a work with no distinctive title and/or no author.
Work languages are used to identify and structure the intellectual (not the physical) attributes of information. Chief attributes are typically the title and author, named to use in ISBD or MaRC formats for collocation. To name the attributes, two type of expressions are used: 1) verbatim from the document using vocabulary that librarians didn’t choose (uncontrolled), or 2) librarian-assigned names using normalized (controlled) vocabulary.
When choosing controlled vocabulary, three steps are traditionally taken: 1) choose an authoritative form of name, 2) make the name distinctive, and 3) map the authoritative name to other authoritative names in similar works. AACR’s disambiguation rules frequently call for the element of birth and death dates for personal names, locations to corporate names, or authors to titles. Names with two or more attributes are called composite attributes. For mapping, use variant author names, a former name of a corporate body, or variant authors of titles as See also references. In terms of metadata, attribute names are called work identifiers (IDs). Subclasses of works (editions) are given IDs when needing to connect the edition to another manifestation of an edition or to a completely different work.
Relationships are formed between works, editions, and superworks through mapping. Work IDs are one of the paths (or devices) of relations between two works having the same attribute name. Other devices for relational paths are notes (descriptions), linking entries, See also references, codes, emblematic description (one item representing a set of items), hierarchical description, and analytic description. There are six different relationship types. The first is a membership relationship, where an entity is related to its entire work subset, usually using emblematic description. The second is inclusion relationships, which is between a subset and the main entity, typically described using hierarchical description. The third relationship is called an equivalence relationship, which is between two subsets of the same set. Fourth is an aggregation relationship, where a component of a subset is taken out and used as another subset. These are typically described using linking entries, See references, or analytic description to identify the removed section (the part) that is from the whole (the aggregate). The fifth relationship is Sequence, holding between some entities related by content, but separated in time, like sequels to a novel. The sixth one is commentary relationships between two or more bibliographic entities, like when one entity is a manual on, or review of another entity.
This chapter of Svenonius was much easier to understand with a few classes under my belt. There are way too many words meaning the same thing being used, which causes confusion. Attributes, meaning identifiers, or names, access points, and entries all being used interchangeably. I understand needing to use a thesaurus to change things up in a writing, but in a textbook for teaching, it takes a great deal of reading and rereading to understand that the author just means the same thing. Svenonius suggests the AACR definitions of relationships undergo a rigorous analysis as there are gaps and redundancies that are very obvious in the digital realms where everything has its place, and things not quite in place stick out like a sore thumb.