Karen Coyle wrote:
|1. When creating a MARC record, the first LC subject heading on the
|record is supposed to be one-to-one with the single LC classification
|number assigned to the item. I don't know if catalogers still do that,
|but it was true at one time. That would presumably make the first LCSH
|field be "more important" than the others.
Here are some aspects of cataloging that may help to shed some light on
various things mentioned:
From LC's "Subject Cataloging Manual" (5th ed., 1996; revised pages,
2001):
SUBJECT HEADINGS
H80(1) - Assign the heading that represents the predominant topic of the
work as the first subject heading. If the predominant topic cannot be
represented by a single heading, assign as the first and second headings
the two headings that, taken together, express the predominant topic.
Although it is not significant which of the two is assigned first and
which second, if one of the two more closely approximates the class
number it is usually assigned first.
H80(2) - If a work has two equally important major topics, assign
heading(s) for the second of these topics immediately after the
heading(s) for the first, and before any headings for secondary topics.
H80(3) - Assign headings for secondary topics, as well as headings
required to complete standard arrays, in any order following the
heading(s) for the major topic(s).
H180(1) - Assign to the work being cataloged one or more subject
headings that best summarize the overall contents of the work and
provide access to its most important topics. -- LC PRACTICE: Assign
headings only for topics that comprise at least 20% of the work.
["Note: There are certain works to which the Library of Congress assigns
no subject headings because of their very general or amorphous nature,
for example, a general periodical or a collection of essays with no
discernable theme. In addition, it is Library of Congress practice not
to assign subject headings to texts of sacred works or to individual
works of belles lettres with no identifiable theme or specific form."]
H180(3) - The number of headings that are required varies with the work
being cataloged. Sometimes one heading is sufficient. Generally a
maximum of six is appropriate. In special situations more headings may
be required. If more than one heading is present, Library of Congress
catalogers assign them in order of predominance. See H80 for
instructions on order of headings. -- LC PRACTICE: Do not assign more
than ten headings to a work.
H180(4) - Assign headings that are as specific as the topic they cover.
Specificity is not a property of a given subject heading; instead, it is
a relative concept that reflects the relationship between a subject
heading and the work to which it is applied. For example, a seemingly
broad heading like "Psychology" is specific when it is assigned to an
introductory textbook on psychology. The method through which
specificity is achieved depends on the nature of the available headings.
In many cases, specificity can be achieved by assigning a basic heading
consisting of one word or a phrase; in other cases, specificity can be
achieved by subdividing a heading. ... Follow the hierarchical reference
structure built into the subject authority file to find as close a match
as possible between the topic of the work and headings that exist to
express that topic in the Library of Congress subject heading system.
Assign a heading that is broader or more general than the topic that it
is intended to cover only when it is not possible to establish a precise
heading, when an array of headings is needed to express the topic, or
when the assignment of a more general heading is called for by special
instructions in the "Subject Cataloging Manual".
H180(5) - A given heading, depending upon its place in a hierarchy, may
subsume several subtopics that are also representyed by headings in the
subject authority file. Assign to a work only the headings that most
closely correspond to the overall coverage of the work. Do not assign
headings that represent the subtopics normally considered to be included
in an assigned heading's coverage.
H180(6) - If a work discusses a general topic with emphasis on a
particular subtopic, or presents a principle and illustrates the
principle with a specific case or example, assign headings for both the
general topic or principle and for the subtopic or specific case or
example, provided that the treatment of the latter forms at least 20% of
the work.
H180(7) - If a heading exists ... that represents the two or three
topics discussed in a work, and that includes no other topics within its
scope, assign the one heading instead of two or three narrower headings.
H180(8) - ["Rule of three"] If a general topic includes in its scope
more than three subtopics, but the work being cataloged discusses only
two or three of these subtopics, assign the appropriate two or three
headings rather than the broader heading. If more than three of the
subtopics are discussed in the work, assign the broad heading instead,
unless the rule of four, described below, applies.
H180(9) - ["Rule of four"] In certain circumstances it may be
preferable to assign headings for four tubtopics of a broad concept. If
a heading covers a broad range and each subtopic forms only a small
portion of that whole range, assign the four subtopics instead. For
example, for a discussion of the works of four American literary
authors, a heading for each author may be assigned since the heading
"American literature--History and criticism" covers the works of all
American authors. -- LC PRACTICE: Do not exceed four subtopics under any
circumstances.
H187(1) - Establish a subject heading for a topic that represents a
discrete, identifiable concept when it is first encountered in a work
being cataloged, rather than after several works on the topic have been
published and cataloged. [my note: This is different from when I first
entered the profession 30 years ago. At that time the principle for
establishing new headings was "literary warrant", that is, there had to
be a certain number of works published (at least 3) on a given topic
before it was considered worthwhile to create a new subject heading.]
H187(3) - Headings are usually established to reflect current American
usage for a concept, but sometimes no consensus has yet developed among
the authorities in a given field as to the proper terminology for the
concept. When establishing a new heading in such a situation, conduct
authority research according to the principles given in H202 and then
make an intuitive judgment based on available evidence (in some cases
only the work being cataloged) by selecting elements that will allow the
heading to express what is intended and at the same time serve as a
retrieval term in the system. Provide UF references from any
significantly different terms that have been found to be used for the
same concept. If terminology for the concept later becomes well
established in another form, it may be necessary to change the heading
to the more appropriate form.
ASSIGNING LC CLASSIFICATION NUMBERS TO SUBJECT HEADINGS
H365(1) - Assign a class number to a heading if: (a) the caption for the
number is identical or nearly identical in scope, meaning, and language
to the subject heading, or (b) the opric is explicitly mentioned in an
"Including" note under the caption for the number, or (c) the topic
belongs to a category for which it is standard LC classification policy
to classify works at a level that is broader than the subject headings
assigned. Do *NOT* assign: (a) a number representing a concept broader
than, and encompassing, a specific heading; (b) a number for a general
topic to a heading or heading-subdivision combination for a spcific
aspect of the topic, even if the specific aspect classes in the general
number; (c) a class number with a qualifier for a specific aspect of a
topic to a general heading, when the specific aspect is actually
expressed by a free-floating topical subdivision under the general
heading; (d) a class number in which a particular item (or the only
item) on a topic has been classed but whose caption does not directly
correspond to the subject heading; and (e) a "General specific" number
to any subject heading, even if the topic is specified in an including
note.
From the introduction to LCSH: "Approximately [only!!] 36 percent of
headings are followed by Library of Congress class numbers, which
generally represent the most common aspect of a subject. ... Class
numbers are added only where there is a close correspondence between the
subject heading and the provisions of the Library of Congress
classification schedules."
|2. ... I did hear that at around the same time that the MARC
|record became part of automated systems (and no one was trying to
|squeeze subject headings into the little margin at the top of cards)
|that the average number of subdivisions per heading AND the average
|number of headings per record went up measurably. So older works would
|have less to work with.
When I was in library school, catalogers-in-training were instructed to
typically assign no more than 3 subject headings to a work. I have seen
this particular practical "rule of 3" mentioned over and over throughout
the years up to the present, so I'm not so sure that the number of
headings per record has increased as dramatically as you suggest.
However, regarding subdivision practice, a major change occurred in 1974
that might have led to an increase. Prior to 1974, subdivisions
generally would be assigned only if found in LCSH under given subject
headings. In 1974, LC adopted a set of "free-floating" subdivisions (a
separate publication from LCSH) that could be applied to almost *any*
subject heading. That meant that catalogers could use subdivisions to
gain more specificity for topics.
|I'd like to see some playing around with the LC class numbers. I think
|this would be difficult, but the classification is in machine-readable
|form and there is text associated with the numbering system. The
|advantage is that you have a real hierarchy, or I should say "some real
|hierarchies" because there isn't really an overarching one. You also
|have facets for things like geography and time.
Regarding an "overarching" hierarchy, a standard joke about LC
Classification was that, when it was "invented" around 1897, they just
piled all the books on the floor in the middle of the library and
sort-of said (pulling together similar materials out of the pile),
"Let's put these in Class A, and these in Class B, and these in Class
C," etc. LC Classification is a truly "topical" alphanumeric
classification scheme; the Dewey Decimal Classification, on the other
hand, is *NOT* "topical" (the same topic can be found in different
areas, depending), but it's an arrangement by "discipline" (that is,
emphasis or aspects). (This "throws" a lot of public services
librarians, because they often complain about "why can't all the books
on this topic be in the same place?") For example, the topic "gold" can
be found in almost all of the 10 main Dewey classes: the use of or
mention of gold in the Bible might be in the 200's (religious aspects),
the etymology of the word "gold" might be in the 400's (language
aspects), the chemical element gold might be in the 500's (scientific
aspects), decorative gold ornaments and jewelry might be in the 700's
(artistic aspects), and the California Gold Rush might be in the 900's
(historical aspects).
|But what really gets me about the LC headings on records is exemplified
|by this: The 1948 edition of Norbert Wiener's book "Cybernetics" has
|only one subject heading: "Mathematical statistics." Since he had just
|invented the term "cybernetics" it didn't exist as a term. By 1965,
when
|the book was revised and reprinted, the only subject heading is:
|Cybernetics. Nothing links the two, at least not in the subject heading
|world. In this example, user tagging would probably be more useful.
|(Note, the books do get similar shelf numbers in some of the
|libraries I can see on my screen.)
Looking in LCSH at both headings, I would imagine that "Mathematical
statistics" no longer includes many of the kinds of things it might have
back in the days where at least 3 or more books on cybernetics had to be
published before it could get its own subject heading. Additionally,
cybernetics is related to information theory, a topic dear to the heart
of John von Neumann, too. But I'm quite sure that, many years ago,
before computers and systems were as important as they are now,
information theory used to be considered a branch of mathematics because
of its heavy emphasis on probability, statistics, and statistical
"noise". I don't know about LCC, but, in Dewey, the topics are in two
completely different places: cybernetics (and information theory) is in
003.5 (systems theory), and mathematical statistics is in 519
(probability and statistics).
But why link two terms that are no longer related? Perhaps definitions
in use have changed. Language always changes over time, which means
that vocabulary-dependent things have to be flexible enough to change
with it. I mean, it took what, 80 or 90 years, for LC to change
"aeroplane" to "airplane", and you're complaining about LC assigning
more exact subject terminology according to contemporary understanding??
Why is "user tagging" "more useful" when it's wrong? Just because it's
"tagging"? Why not assign arbitrary "wrong" subject headings just
because the book reminds me of something else? (The tagging principle
would be the same.) I didn't quote it above, but LC's instructions for
establishing new headings also include the concept of being as objective
and neutral as possible in the terminology chosen for a new subject
heading.
Just pointing out some more areas of consideration in NGC theory...
Harvey
--
===========================================
Harvey E. Hahn, Manager, Technical Services Department
Arlington Heights (Illinois) Memorial Library
Desk: 847/506-2644 -- FAX: 847/506-2650 -- E mailto:hhahn_at_ahml.info
Personal web pages: http://users.anet.com/~packrat
Received on Mon Feb 05 2007 - 19:32:44 EST