I realize this has become a discussion on the value or not of dates, as opposed to a general discussion of variant access points.
As to the first issue raised by Ben, not only do dates not change individually, but they don't change in relationship to other people's dates, so people can be put into some sort of context. That means being able to tell if someone came of age in the Twenties or the New Deal or one of the World Wars, not just the broader case of the 13th versus 19th centuries. It means that they would sort chronologically in an old fashioned browse search.
As to the second, as a historian I am not sympathetic to people who question the value of dates, but I am sympathetic to people who don't have the time or the means to deal with dates in the context of the flood of modern authorship.
The value of dates depends entirely on the type of works and the type of use. For people looking for things that are popular or canonical, they are usually superfluous. These are known products like Ben & Jerry's ice cream or the "Fast and Furious" franchise. People go looking with blinkers on and a narrow focus, and in such cases one-click shopping is both desirable and probably obtainable. People will get the names to look for because they are circulating in current media, or because they get them in things like course reading lists or discussion group reading lists. At the same time, it is worth remembering that what is current fades away. Who but specialists know the names of the ordinary pop stars and celebrities and the movers and shakers of the Edwardian Age, not to mention earlier periods? They have slipped as far out of popular consciousness as once-popular products like Hostetter's bitters or Uneeda biscuits or bombazines. So a lot of things can be expected to fall out of timely, generalized use into the specialized and obscure. And to the extent that the past is the proverbial "another country," it requires more road maps and travel guides. Strictly speaking, someone looking for the "Principia" doesn't really need Isaac Newton's dates, but people looking for the Isaac Newton who was a 19th century New York naval architect and steamboat operator, or the near-contemporary Boston editor and promoter of the same name, do. And they don't want the "Principia" or the history of science or even a published work as such.
With such research, especially like the kind of work I have spent my career doing in primary sources, one typically finds a name somewhere in an old letter or newspaper or obscure pamphlet and it becomes a matter of finding more about the person, not simply what they have published, which may be nothing at all, but perhaps their motive for taking one stand or another, or personal relationships with persons and events already discovered. That kind of use takes in a search for other obscure sources, something that digitization has made much easier by its ability to search inside the work, but also more complex when dealing with common or similar names. It is not a blinkered search, because one is often not entirely certain of what one is looking for, (e.g., the name as first found may be slightly misspelled), and it is definitely not a matter of one-click shopping, even with sources that can be trolled digitially. It is not a matter of looking for something pre-packaged, because at some point the search leads away from the published to the unpublished or, in the case of small notices in a periodical, from the wrapper to the fine print. In such searching, every little clue helps. Dates in particular are practically a necessity to filter out totally irrelevant results when searching for people in Google, Ancestry and similar data bases.
One problem with using dates on an as-needed-to-disambiguate-only basis goes something like this. The first "Samuel G. Wright" to be discovered goes in without dates. Any subsequent "Samuel G. Wrights" receive dates. Now it is impossible to tell at a glance their chronological ranking (e.g., father/son), or even the century of the undated person. Adding the date in the first instance would eliminate this. For a real example, look at "Dey, Anthony." The undated "Anthony Dey" was not an author but a New York City Federalist and entrepreneur associate of Hamilton. It would most likely be the second type of researcher who would be looking for him and not the others. That Anthony Dey's interests and activities were sufficiently varied that it would be hard to hang a succinct occupational qualifier on him, short of following the Clint Eastwood example and calling him "member of X," "associate of Y," and so forth. Maybe when everything is up and running, the 3xx fields can do that, but someone has to take time to fill in the data first. In fact, the dates are probably the easiest data to obtain and the quickest to enter.
In archival description, the point has ALWAYS been to identify the person, and less to collocate the works, which are usually unpulished letters or non-literary things. It calls for all relevant all data ABOUT the person, not the simplest level of disambiguation.
I don't know how many of you have created or are familiar with a typical archival BIB record, but a reasonably-sized collection has anywhere from 100 to 300 added entries which actually constitute an INDEX to the collection, including the names of principal correspondents and entities recorded or discussed, plus topical and geographical headings for subject matter. While every document is a discrete "work," the description is at an aggregate level, and because the arrangement is by provenance and original order, and not by an artificial collocation, the added entries can be as far-ranging as the principal creator's interests and activities. Such an index is needed, even with full on-line finding aids, because it has to encompass content WITHIN files rather than simple file titles. Indeed, many things with a chronological arrangement, such as letterbooks, account books or minute books, have no title that will describe contents without such an index. In the BIB display, the added entries form a mini-browse list linking collections, but they are also an explication of the named entities and clues to their possible relationships with one another. With dates, people file in proper generational order. Many of these people are relatively obscure, so whatever information we can provide to both reference staff and patrons is a potential advantage. Sacrificing this information to comply the the needs of others constitutes a kind of instirtutional dumbing-down and trading that road map to "another country" for a contemporary Google map with its paid flaggings of contemporary bars, restaurants and lodgings. With dynastic families, relationships are particularly important, and identically-named people are more common. That is why we have always added dates in our local catalog when available, and since we have official records, they are more likely to be available. As I have mentioned before, we also have relationships with closely associated individuals that change with death, which is why we also close life span dates. Even though it might seem that that has nothing to do with retrieval, it often does affect access and copyright, neither of which is an issue with published materials.
As an illustration of why one-click shopping is not practical with manuscript research, our catalog contains 1,695 instances of "E.I. du Pont de Nemours &Company," without considering topical or organizational subdivisions, 806 instances of "Du Pont de Nemours, Pierre Samuel," the only family member to be famous as a writer and printer, 106 entries for "Du Pont, Eleuthere Irenee" the founder of the company, and 190 entries for "Du Pont, Pierre S. (Pierre Samuel)," who published little but founded our organization.
Perhaps the pressure for one-click shopping for current materials has created an illusion that all ambiguities can be removed and one-click shopping be made universal. As John Marr notes, who is going to disambiguate all the legacy data already in OCLC? How did we get by all these years? The rules practically mandate the creation of undifferentiated or ambigous corporate names. Who is going to create authority records for all the names currently in OCLC without them? We are guilty of that, because we simply don't have the time to create NARs when a single new collection has a hundred or more new names, much less go back and create them for the tens of thousands of names we needed before being able to contribute to NACO. There are plenty of times when we have names that are the same as the one authorized for another person. However, the fields of activity are sufficiently different that only one would reside in our catalog. They would only jostle in a supposedly universal catalog.
I have written before that I think that while it might be possible to create a universal catalog for artists, broadly defined, and their creations, a universal catalog of entities, including entities represented by only one work or whose works are not widely disseminated, is entirely different and beyond the competence and resources of this community. I am sort of curious on what basis and with what knowledge OCLC intends to "authorize" unauthorized names from records harvested from us and others. In our case, we consider the form of many such names provisional. Since we operate in a limited spectrum of subject matter, we are always discoverning new things, as when obituaries are published within the communities we monitor or researchers share or publish what they have discovered in our materials. I find it troubling when soneone who knows little or nothing of the matter, runs away with our data and sets it in cement. Is this work being done by algorithms? I found an instance where an old professor of mine, who was a patrician esthete who collected fine art, has been mashed-up by WorldCat Identities with a similarly named member of the Black Panther Party. Might not dates have made the mistake more obvious?
Yours truly
Chris Baer
Assistant Curator of Manuscripts
Hagley Museum and Library
-----Original Message-----
From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Benjamin A Abrahamse
Sent: Wednesday, September 09, 2015 2:54 PM
To: [log in to unmask]
Subject: Re: [PCCLIST] Instructions for variant access points
I think when they are being used to disambiguate, biographical dates have one advantage over other possible elements, namely that they don't change. People aren't born more than once--except metaphorially, in some religious traditions. So long as catalogers are forming unique strings to serve as identifiers it is probably a better element than others for purposes of qualification.
However, I'm sympathetic to the voices on this list that question the value of dates in identifying people even when their name is unique in the file. I think there are grounds to question the widsom of LC-PCC PS for 9.19.1.3, which instructs catalogers to include dates (if known) in access points regardless of whether they are needed to resolve a conflict.
--Ben
Benjamin Abrahamse
Cataloging Coordinator
Acquisitions and Discovery Enhancement
MIT Libraries
617-253-7137
-----Original Message-----
From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of CHRISTOPHER WALKER
Sent: Wednesday, September 09, 2015 2:41 PM
To: [log in to unmask]
Subject: Re: [PCCLIST] Instructions for variant access points
Colleagues,
The usefulness of dates depends, of course, on the name, the dates, and the amount of conflict to clarify.
If there are few conflicts with the undifferentiated name, dates from widely-separated time periods may pretty clearly differentiate two or three persons, and suffice to produce a usable index in the file.
A person with “flourished” dates in the 13th century cannot very well be the author of a technical manual for typewriter repair, for instance.
Christopher H. Walker
Serials Cataloging Librarian
Penn State's representative to the CONSER Operations Committee Member at Large, ALCTS CRS Executive Committee 2013/2016
126 Paterno Library
The Pennsylvania State University
University Park, PA 16802-1812
(814) 865-4212
[log in to unmask]
----- Original Message -----
From: "Knop, Judy" <[log in to unmask]>
To: [log in to unmask]
Sent: Wednesday, September 9, 2015 8:43:51 AM
Subject: Re: [PCCLIST] Instructions for variant access points
I would agree that the use of dates is not a user helpful way of identifying authors. I don't believe the point was ever to necessarily identify a person, but rather to collocate a specific person's works and differentiate that person from others. For that purpose, dates are very useful, but they don't do well with identifying that person for patrons.
Judy Knop
ATLA NACO Funnel Coordinator
[log in to unmask]
-----Original Message-----
From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Dickerson, Eugene H
Sent: Tuesday, September 08, 2015 2:34 PM
To: [log in to unmask]
Subject: Re: [PCCLIST] Instructions for variant access points
Thanks for adding this additional comment.
Although there is a long history in cataloging practice of using birth and/or death dates to differentiate one identity from another (this device may work OK for the cataloger, and even that is debatable), but I'm not sure how helpful the dates are to the average person who encounters a name in a library catalog. For example, I'm looking in the NACO file for a Richard Smith who is Professor of Health System Economics at the London School of Hygiene and Tropical Medicine. (You can just image how many entries there are for Smith, Richard and all the various forms of that name.) At this stage, I don’t yet have any other information about this person, so all those dates aren't helpful to me in my search. In some ways, the occupational qualifier would be much more useful for the average user. While I'm not suggesting that we abandon using birth and/or death dates as qualifying data to an authorized access point, I think that we may be overestimating how useful that data are in distinguishing one identity from others in a database. A citation consisting of a name and a title plus other information is probably helpful for a user in identifying whether the person in question wrote on a field of the user's interest, but just a personal name and date(s) out of context is probably not as useful in the FRBR tasks of find, identify, and select, as we might hope. Although there is a policy statement urging us to add dates to the authorized access point when readily available, I'm not convinced that there is always a great return on the investment in doing this. Recording the dates in the authority record is useful when they are readily available as one means of identification, but I'm not sure that they always need to be included in an authorized access point. It's not likely that a lot of users are going to know birth and/or death dates associated with an identify unless the name being searched or retrieved is associated with a well-known person, especially for really common names.
Gene
Eugene Dickerson
Team Leader for Cataloging
Ralph J. Bunche Library
U.S. Department of State
Washington, DC
[log in to unmask]
(202) 647-2191 (voice)
No part of any article sent to you by the Bunche Library can be reproduced, distributed, or transmitted without prior written permission of the publisher. The exception are brief quotations. For a synopsis click here: http://diplopedia.state.gov/index.php?title=Copyright:_Synopsis_of_Important_Facts (Link not valid outside the Department of State.)
-----Original Message-----
From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Diana M. Brooking
Sent: Tuesday, September 08, 2015 1:19 PM
To: [log in to unmask]
Subject: [PCCLIST] FW: Instructions for variant access points
Yes, now that undifferentiated name headings are not possible, it is hard to find a qualifier sometimes to differentiate common names. That's why I worry about a proliferation of *different* qualifiers used in 4XXs-- I agree that then would make them unavailable for other 1XXs, wouldn't it?
And if all the possible occupations or whatever other possible qualifiers are present in 3XX fields in the authority record, and assuming some day those will be visible to users in some form (otherwise why are we creating all this data), that makes adding 4XXs with variant qualifiers seem really unnecessary.
Diana Brooking
Catalog/Germanics Librarian
University of Washington Libraries
Box 352900
Seattle WA 98195
206-685-0389
[log in to unmask]
-----Original Message-----
From: Program for Cooperative Cataloging [http://redirect.state.sbu/?url=mailto:[log in to unmask]] On Behalf Of Benjamin A Abrahamse
Sent: Friday, August 28, 2015 11:38 AM
To: [log in to unmask]
Subject: Re: Instructions for variant access points
Gene,
I kind of agree with you in principle, but there are times when qualifying by occupation is the only way to make an access point unique. Especially now that undifferentiated name headings have been deprecated.
That said, adding cross references just to accomodate mulitple occupations strikes me as curious. I should think it would be better practice either to try and find a more generic occupational term, or to figure out which of a person's various roles is most "identifiable".
Still, NACO seems to have decided that all "variant names" are acceptable if an individual cataloger deems them so.
--Ben
Benjamin Abrahamse
Cataloging Coordinator
Acquisitions and Discovery Enhancement
MIT Libraries
617-253-7137
From: Program for Cooperative Cataloging [http://redirect.state.sbu/?url=mailto:[log in to unmask]] On Behalf Of Gene Fieg
Sent: Friday, August 28, 2015 2:11 PM
To: [log in to unmask]
Subject: Re: [PCCLIST] Instructions for variant access points
I may be in a distinct minority here, but I should declare my views before I make any suggestions. I do not like using $c (function) as part of the name, especially in establishing a person in the NAF.
The name of the person is the name of the person without the qualification of function. For instance, Clint Eastwood; that is his name no matter what his function is. He may be any one of the following: Eastwood, Clint (Actor); Eastwood, Clint (Director); Eastwood, Clint (Producer); Eastwood, Clint (Mayor).
His name as in NAF should be what it is now: Eastwood, Clint, ?d 1930-
If a cataloger, wishes to add $c [function], that his/her business for the local library, but it should not be part of AAP. Function should not be part of it in the NAF. (As a background to this, we had some confusion in my field of cataloging of distinguishing between Biblical character, Biblical figure, and Biblical leader, etc when such descriptions followed a biblical name.
Such functions may, and I emphasize "may" here, be useful with given names, but not with surname, forename formulations.
Gene Fieg
Aug. 28, 2015
On Thu, Aug 27, 2015 at 7:16 PM, Adam L. Schiff <[log in to unmask] <http://redirect.state.sbu/?url=mailto:[log in to unmask]> > wrote:
A question recently came to me inquiring about an authority that we had created which had some possibly dubiously valid variant access points:
100 1 Francis, Matthew ?c (Dramatist)
370 ?c England ?c Great Britain ?f Greenwich (London, England) ?2 naf
372 Drama ?a Stage adaptations ?a Theater--Production and direction ?a Acting ?2 lcsh
373 Greenwich Theatre ?2 naf
374 Dramatists ?a Theatrical producers and directors ?a Television producers and directors ?a Actors ?2 lcsh
375 male
377 eng
400 1 Matthews, Francis ?w nne
400 1 Francis, Matthew ?c (Director)
400 1 Francis, Matthew ?c (Producer)
400 1 Francis, Matthew ?c (Actor)
400 1 Matthews, Francis ?c (Producer)
667 Formerly on undifferentiated name record: no2003109605
670 New tricks. Season three [VR], c2010: ?b container (Francis Matthews; producer)
670 Francis, Matthew (Dramatist). Jane Austen's Northanger Abbey, 1997: ?b title page (adapted for the stage by Matthew Francis) unnumbered page after title page (First presented at Greenwich Theatre on 4th July 1996; directed by Matthew Francis) production notes (our production at Greenwich)
670 Epic stories on stage website, June 24, 2015: ?b about (From 1990 to 1998, Matthew Francis was Artistic Director of the Greenwich Theatre. His work there included his own adaptations of Tale of Two Cities, David Copperfield, The Prisoner of Zenda, Northanger Abbey and Huckleberry Finn; has worked extensively in repertory theatres across the UK; he was Associate Director at the Chichester Festival Theatre; his credits as television producer include the BBC success Office Gossip and Gimme Gimme Gimme (the UK inspiration for Will & Grace). After working on the second series of the successful comedy My Dad's The Prime Minister, Matthew produced the third series of the BBC's most successful drama New Tricks; began his career as an actor - when, between theatre and TV appearances, he spent three happy months as front man for the Swedish pop group ABBA) ?u http://redirect.state.sbu/?url=http://www.epicstoriesonstage.com/about/about.htm
670 IMDb, June 24, 2015 ?b (Matthew Francis (I). Matthew Francis is a producer and actor, known for My Dad's the Prime Minister (2003), Gimme Gimme Gimme (1999) and Office Gossip (2001); and: Francis Matthews (II), producer) ?u http://redirect.state.sbu/?url=http://www.imdb.com/name/nm0290250/ <http://redirect.state.sbu/?url=http://www.imdb.com/name/nm0290250/> ?u http://redirect.state.sbu/?url=http://www.imdb.com/name/nm2699755/ <http://redirect.state.sbu/?url=http://www.imdb.com/name/nm2699755/>
670 Lipman, Maureen. I must collect myself, 2010, via Google books, June 24, 2015: ?b (Matthew Francis, former director and administrator of the Greenwich Theatre, also known to me as Francis Matthews)
670 Jane Austen's Northanger Abbey, 1997: ?b title page (adapted for the stage by Matthew Francis)
The question of course dealt with the 400s in the record that began with Francis, Matthew, which is the preferred form of the person’s name and is also found in the 100 field.
Indeed, RDA 9.19.2.1 does say “When constructing a variant access point to represent a person, use a variant name for the person (see 9.2.3 <http://redirect.state.sbu/?url=http://access.rdatoolkit.org/document.php?id=rdachp9&target=rda9-3021#rda9-3021> ) as the basis for the access point.” It also goes on to say that you may “Make additions to the name, if considered important for identification.” There is nothing in the current RDA chapter 9 instructions that would seem to allow using the preferred name of a person as the basis of a variant access point. And yet, these can fairly easily be found in RDA-coded NARs created by both LC and other PCC catalogers. Examples that I found quite quickly by some judicious searching in the NAF through OCLC:
100 1 Cabral, Ami´lcar, ?d 1924-1973
400 1 Cabral, Ami´lcar, ?d 1921-1973 ?w nne
100 1 Sa´enz, Jorge ?c (Author)
400 1 Sa´enz, Jorge ?c (Security specialist)
100 1 Franklin, Michael, ?d 1972-
400 1 Franklin, Michael ?c (Medical editor)
100 1 Bernard, James ?q (James T.), ?d 1977-
400 1 Bernard, James ?c (Author at PowerKids Press)
100 1 David, Michel ?c (Film producer and actor)
400 1 David, Michel ?c (Actor)
100 1 Pereira, Augusto ?c (Film producer)
400 1 Pereira, Augusto ?c (Actor)
100 1 Singer, David ?c (Film producer)
400 1 Singer, David ?c (Actor)
100 1 Carboni, Bruno ?c (Film editor)
400 1 Carboni, Bruno ?c (Film director)
If you look at instructions under works and expressions, RDA does allow us to use a preferred title as the basis for a variant access point. 6.27.4.1 says “Construct additional variant access points if considered important for access” as does 6.27.4.3. At 6.27.4.1 we have some examples that use the preferred title as the basis for a variant access point rather than a variant title:
Jeanne-Claude, 1935– . Wrapped Reichstag
Authorized access point for the work: Christo, 1935– . Wrapped Reichstag. A work of art created jointly by Christo and Jeanne-Claude. Variant access point considered important for subject access
Management series (Chicago, Ill.)
Authorized access point for the work: Management series (Ann Arbor, Mich.). Place of publication of series changed from Ann Arbor to Chicago
And 6.27.4.5 says that you can construct a variant access point for an expression by adding a variant of an addition used in constructing the authorized access point representing the expression. Example from the NAF:
046 ?k 1992
130 _0 Blade runner (Motion picture : ?s Director's cut)
381 Director's cut $a 1992 version
430 _0 Blade runner (Motion picture : ?s 1992 version)
430 _0 Blade runner (Motion picture). ?f 1992
I think that there may be times when variant access points for persons, families, and corporate bodies that are based on a preferred name rather than a variant name might be useful. Catalogers ought to be free to make variant access points of any kind if they will aid users, oughtn’t they? So I am wondering if there would be support for adding an instruction like what is in 6.27.4.1 to the variant access point instructions in chapters 9-11? That is, something like “Construct additional variant access points if considered important for access” with some examples.
Adam L. Schiff
Principal Cataloger
University of Washington Libraries
Box 352900
Seattle, WA 98195-2900
[log in to unmask] <http://redirect.state.sbu/?url=mailto:[log in to unmask]>
(206) 543-8409 <tel:%28206%29%20543-8409>
(206) 685-8782 <tel:%28206%29%20685-8782> fax
|