skip to content

Must Contextual Description Be Bound To Records Description?

I've been struggling with the fact that (American) archival practice seems to bind contextual description (i.e., description of records creators) to records description. Much of these thoughts have been stirring in my head as a result of my class at Rare Book School. If we take a relatively hardline approach, e.g. the kind suggested by Chris Hurley ("contextual data should be developed independently of the perceived uses to which it will be put", 1, see also 2), it makes total sense to separate them entirely. In fact, it starts making me mad that the <bioghist> tag exists at all in EAD. Contextual description requires that it be written from a standpoint relative to that of the creator it describes. I guess what I keep getting hung up on is if there could be a relevant case that really merits this direct intellectual binding. I therefore appeal to you, humble readers, to provide me with your counsel. Do you think there are any such cases, and if so, why?


  1. Chris Hurley, "Ambient Functions - Abandoned Children to Zoos," Archivaria 40 (Fall 1995): 21—39. Availalable from
  2. Chris Hurley, "Problems with Provenance," Archives and Manuscripts 23, no. 2 (November 1995): 234-259. Available from



  • 💬 Jonathan Rochkind at August 1, 2009, 01:33 UTC:

    Is this different in European archival practice? In my vague memories of the archives class I took in library school (which, much like every other class I took involving library metadata, involved me asking "Wait, why the heck do you do THAT", and professionals getting exasperated with me -- a fine preparation for my professional career!), anyway, in my memory, I thought we got that from European tradition in the first place, and they do it even more than us?

    Ah, but I bet Australian/New Zealand does it better. They seem to do everything related to archival description for the contemporary world a couple decades ahead of us.

  • 💬 Jonathan Rochkind at August 1, 2009, 01:45 UTC:

    Oh, and PS:, I remember being captivated by the writings of David Bearman in that archives class, he seemed to get it about doing archival metadata right. I asked the professor (a professional archivist) "What happened to him and his ideas, anyway?". He said "Eh, he got frustrated that everyone just ignored him, and went and did something else."

    Some day, that too can be you and me! heh.

    And no, I don't see any reason for contextual metadata to be in a metadata record for a particular archival record set, for the main reason that that contextual metadata is gonna often end up duplicated across multiple archival record sets, aint' it?

  • 💬 M.A. Matienzo at August 1, 2009, 03:17 UTC:

    I guess something else I'm picking up on is this sort of thing at least should make archivists more interested in FRBR. I'll be the first to admit that FRBR Group 1 entities tend to get ontologically sticky when you consider a fonds and its parts – either something being at once a Work, Expression, Manifestation and Item (e.g. an otherwise unpublished letter), and something quite different, like a hand-annotated version of a screenplay for a movie that's been remade several times. However, Group 2 and Group 3 entities in part help to establish context, and the essential role of relationships between those entities fits nicely with the work being done, for example, on EAC-CPF.

    Australian and New Zealand libraries and archives do often seem to be worlds ahead of us. I've asked a few colleagues from that part of the world why (informally) they think this is the case - people like Tim Sherratt at the National Archives of Australia, and Virginia Gow of the DigitalNZ project. The overall opinion seems to be that it's in part due to their relatively recent history as independent nations, as well as the relatively recent history of national libraries and archives in those countries. To me, this also suggests the possibilities of the professional practice being less entrenched than, say, in the United States.

    As far as David Bearman goes, he's still causing a ruckus, albeit not directly in the archives world anymore; he's still active with his consulting firm, Archives and Museum Informatics, most notable in part for organizing the Museums and the Web conferences these days.

  • 💬 Jonathan Rochkind at August 1, 2009, 03:48 UTC:

    Whole/part relationships still haven't been adequately modelled by FRBR type stuff, and that's often what (I think?) you need with traditional archival description.

    But there's nothing odd with "something being at once a Work, Expression, Manifestation and Item" -- that is indeed true of any unique (rather than mass produced/copied) item, and there's nothing too tricky about it. I've suggested before ( it's useful to think of the FRBR Group 1 Entities as set relationships. So the workset consists of only one expression which consists of only one manifestation which consists of only one item. Maybe a bit of extra apparatus for a unique item, but no problem!

    But, hey, as soon as you digitize that unique unpublished letter, you've got another manifestation. And as soon as some scholar translates the letter, you've got another expression. The apparatus is there waiting for when it's needed, whether archival description practice will use it or not, somebody will.

    But the hand annotated screenplay gets into the trickiness of whole-part relationships again that FRBR isn't very good at, it needs to be fleshed out how to model it. Compare to a traditional published collection/anthology of individual items that may be previously or subsequently published independently. You've got multiple different works (each item)... published together in one single manifestation, and creating a new work representing the aggregate? That's how I'd choose to model it, but I'm not sure this is consensus. The annotated screenplay is actually quite similar, I'd call the annotations one work, and the screenplay another, coming together to form a manifestation of a new aggregate work (no matter that it's a work with only one expression and annotation... for now, until someone digitizes it or translates it!)

  • 💬 M.A. Matienzo at August 1, 2009, 04:12 UTC:

    I get what you're saying, and this is kind of what I was trying to allude to with noting that FRBR Group 1 entities considered in an archival context gets ontologically sticky. I really don't think part/whole relationships are an exact fit with the driving principle behind multi-level arrangement and description of a fonds. I think there is very important ontological distinction between, e.g., a multi-volume monograph, or a website and its components (as suggested in Barbara Tillett's What Is FRBR? [PDF]), versus a body of records created organically by a corporate body processed by an archivist. It just seems flat out wrong to me to say that a fonds, sous-fonds, series, etc. is one of the Group 1 entities — rather, only an item can every be an instance of one of those entities.

  • 💬 jordon at August 1, 2009, 06:07 UTC:

    I get what you're saying, but when I have multiple finding aids with the same records creator (i.e. let's say, an organization's archives, and I don't want to create one humongous finding aid), I copy and paste the same information in the bioghist note and then in certain circumstances I will modify the note based on the collection I'm describing--you know, emphasize some historical points, de-emphasize other stuff. In other words, the collection's contents have direct bearing on the contextual information.

    However, most of the time I do wish that at least part of this information--like, the boilerplate concerning the administrative history of the organization--could be managed elsewhere, and then somehow imported into/linked to finding aids, which is where I'm hoping EAC-CPF will come in. But that's coming from someone who has never even seen the schema, so I'm looking forward to checking it out in the coming weeks.

  • 💬 M.A. Matienzo at August 1, 2009, 06:53 UTC:

    Jordon, I guess that's my point - should we really be bothering to emphasize/de-emphasize that within the finding aid proper? I'm really starting to think they should be strictly separated.

    Your second point, though, is really what's behind the grand vision for EAC-CPF. The problem is (as I've written before) that EAD isn't namespace-enabled, meaning you can't arbitrarily pull in elements from other metadata schemas. EAC-CPF, however, has this baked into the schema itself. In part, this is a big reason why I (as well as others) have been calling for the revision of EAD.

  • 💬 David Bearman at August 18, 2009, 23:12 UTC:

    It's nice to see this discussion revived. You'll note Dick Lytle and I tried to get this segregatiuon introduced in 1985 ( "The Power of the Principle of Provenance", Archivaria, #21, p.14-27) See
    Later, I opposed EAD when it was being introduced largely for this reason (and the fact that it was in effect a kind of typographical rather than logical mark-up).
    The current wiki/web 2.0 environment makes the collective construction of the authority apparatus so much easier and argues for the right architecture. I hope you succeed.

  • 💬 Jane Stevenson at September 22, 2009, 14:06 UTC:

    What about archivists who like to tailor the specifically to the collection description? Would it be appropriate to have a small collection of a couple of letters that links to a full biographical history? I'm undecided on this, but my inclination is to favour (or favor!) the efficiencies that come from using one biographical history.