5.3 Modelling of Bibliographic Identities

“The modelling of bibliographic identities or personas in FRBR-LRM makes use of the nomen entity and the has appellation relationship. The appellation relationship is one-to-many and holds between any entity and the various nomens used for that entity. In particular, persons (defined as: an individual human being) generally have multiple nomens; the use of each nomen may be governed by many factors, including the preference for certain nomens in specific contexts.” p. 61


4 thoughts on “5.3 Modelling of Bibliographic Identities

  1. Pingback: World-wide review of the FRBR-Library Reference Model, a consolidation of the FRBR, FRAD and FRSAD conceptual models | FRBR Open Comments

  2. With identifiers in place, the names given to things have less importance. In this model, there is a single identity with potentially multiple names. That is similar to the authority control model with one authoritative name and zero or more non-authoritative names.

    Note that there is a huge flaw in this model because non-authoritative names are not unique within the universe, and therefore are many-to-many, which isn’t addressed here. This model seems to assume that all entities have been assigned identifiers, but doesn’t address if/how names and identifiers from various sources are brought together. cf. VIAF matching.

    This model implies a single, authoritative identifier for entities. That is not a reasonable assumption since the bibliographic world is not a single community, and different communities have employed different identifiers. In addition, there doesn’t seem to be any room in the model for ambiguity, and yet in the reality of the bibliographic world there is lots of ambiguity. Already, VIAF reflects this. It creates a single VIAF identifier as its internal ID, but all of the many identifiers from different library systems are still valid, and in other systems where VIAF IDs are used (e.g. Wikipedia), the VIAF ID becomes just yet another identifier for something whose definition is essentially crowd-sourced.


  3. I’m not quite convinced by the treatment of pseudonyms in FRBR-LRM. Let’s take as an example a person (P) who writes both under his or her real name and a pseudonym (let’s say that work X was written under the real name, and works Y and Z under the pseudonym). We’ll assume that there is a preferred version of the real name (NRa) and a variant form of the real name (NRb) as well as a preferred version of the pseudonym (NPa) and a variant form of the pseudonym (NP2b). Let’s also imagine that the pseudonym has its own, fictitious biography.

    In RDA, this is modelled as two persons:
    P1 has as attributes the names NRa and NRb.
    P2 has as attributes the names NPa and NPb.
    The biographic information for the real person is recorded in the description of P1, the fictitious biographic information is recorded in the description of P2. There is a relationship between P1 and P2. In our implementations, there are two authority records – one for P1 and one for P2.

    Now, in FRBR-LRM, the entity person ist restricted to real human beings (p. 19). Therefore, we can have only one person entity (P). Another difference is that the four names are no longer attributes, but entities in their own right (“nomens”). So we do not have two entities, as in RDA (P1 and P2), but five entities (P and the four name entities). There are relationships from the person to the “nomens”, like this:

    P — has appellation –> NRa
    P — has appellation –> NRb
    P — has appellation –> NPa
    P — has appellation –> NPb

    However, it is still possible to make use of the concept of separate bibliographic identities. This is explained on p. 61f. of the draft: “a bibliographic identity is a cluster of nomens used by a person in the same bibliographically significant context or contexts”. In our example, there are two such clusters:

    Cluster 1:
    NRa, NRb (context of use: work X)

    Cluster 2:
    NPa, NPb (context of use: works Y and Z)

    As the concept of separate bibliographic identities is explicitly mentioned in FRBR-LRM, and a solution given how to model it, I believe that FRBR-LRM doesn’t aim at abolishing this principle.

    I also assume that, in the real-life implementation, we would not have to change our data model. Rather, I think we could still use the two existing authority records in a FRBR-LRM framework, although the theory behind them would have changed. The record which formerly represented P1 would, in future, represent both P and the nomens NRa and NRb (i.e. cluster 1). The former record P2 would, in future, represent the nomens NPa and NPb (i.e. cluster 2).

    But if this analysis is correct, there is still a snag in the model: There doesn’t seem to be any place where we could satisfactorily record the fictitious biographic information associated with the pseudonym.

    As mentioned before, FRBR-LRM views a bibliographic identity simply as a cluster of nomens. But I think this is not enough: There may also be a fictitious date and place of birth, a fictitious profession a.s.o. But these cannot be attributes of the nomens in cluster 2 or relationships to them. It seems to me that the only way to record this kind of information would be to define additional attributes and relationships for P, e.g.

    P has real profession: [real profession]
    P has fictitious profession: [fictitious profession]
    P — has real associaton with –> [real place of birth]
    P — has fictitious associaton with –> [fictitious place of birth]

    This looks rather odd to me…


    • This seems to deny the real utility and advantages of pseudonyms. There are political reasons, market reasons, and reasons that are quite useful to readers. For the latter, persons who write in different genres use pseudonyms to indicate which genre type is being authored (Charles Dodgson/Lewis Carroll being an extreme of this, from mathematical texts to children’s stories, but one can also look to writers like Lauran Paine who had something 80 pseudonyms and wrote in a number of genres). Users are looking for the author whose works they enjoy, and if that is a persona instead of a person they don’t mind, AFAIK. As for the market, many publishers refuse to have more than one “new” book out by a single author, probably avoiding diluting the market for that author’s work, so authors who an turn out more than a single book in a short time write under different names, sometimes for different publishers. The political reasons, of course, are that some people do not dare write under their own names. For that we have both political dissidents, as well as folks like Carolyn Heilbrun who wrote mysteries as Amanda Cross so as not to endanger her academic career.

      It always seemed to me that the choice made by later cataloging rules to catalog works under the name on the title page was a practical and user-friendly decision. Practical because it allowed catalogers to do their work without having to dig deeply into research on “real identities,” and user-friendly because there is no reason to expect users to know an author by anything other than the name they see on the book itself.

      It would be interesting to hear the motivation for this decision in the LRM, as well as the strategy for presenting users with the “correct” combination of author names and work titles.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s