SUBJECT HEADINGS
	Alva T. Stone Florida State University atstone@law.fsu.edu

In February 1999 the Library of Congress began coding form subdivisions with the USMARC code $v, a subfield code for subject headings approved some years ago by MARBI. Marie Whited, as AALL's official representative to the ALA ALCTS/CCS Subject Analysis Committee (SAC), reported on this as an upcoming development in the last issue of TSLL. Now it is a reality, and judging from recent comments posted to AUTOCAT, libraries everywhere that use LC cataloging copy have decided to also implement the $v subfield code. Let's take a little time to examine how the new code should be used, and answer some questions that have come up. First, it might be useful to read a definition of form data, as described by a SAC Subcommittee, and presented at the "Educational Forum on LCSH and Subfield v," held last January during the ALA Midwinter meeting:

"Form data are those terms and phrases that designate specific kinds or genres of materials. Materials designated with these terms or phrases may be determined by an examination of:

• their physical character (e.g., videocassettes, photographs, maps, broadsides);
• the particular type of data that they contain (e.g., bibliographies, questionnaires, statistics);
• the arrangement of information within them (e.g., diaries, outlines, indexes);
• the style, technique, purpose, or intended audience (e.g., drama, romances, cartoons, commercials, popular works);
• or a combination of the above (e.g., scores).

A single term may be modified by other terms, in which case the whole phrase is considered to be form data (e.g., aerial photographs, French dictionaries, conversation and phrase books, wind ensemble suites, telephone directories, vellum bound books, science fiction)."

This is a generalized definition, not restricted to LCSH terms, and in fact, it is applicable to form headings as well as form subdivisions. Which brings us to our first question: Is LC going to be using the USMARC 655 tag, instead of 650, for form headings now? (An example of a form heading is "Law reports, digests, etc.") No, not at this point. Only subject subdivisions are affected by new coding right now.

Apart from using the criteria shown above, are there other ways in which we can determine if a subdivision should be coded as form? Yes, you can use two of the same tools you've always used for verifying subject headings. One way is to check in your online version of the LC Subject Authorities. Here you will find new records for subject subdivisions, in 18X fields, with a 680-tagged field explaining how the subdivision is used. If it is a form subdivision, the tag and code will be 185 $v; but if it is considered a topical subdivision, it will be coded as 180 $x. Here is an example of one of the new records, for a form subdivision:

010    sh 99001405
040    IEN $b eng $c DLC
005    19990225091848.7
073    H 1140 $z lcsh
185    $v Commercial treaties
480    $x Commerce $v Treaties
585    $v Treaties $w g
680    $i Use as a form subdivision under names of
          countries, etc., for collections of commercial
          treaties of those places.

Another way to determine if your subdivision is form or topical is to look it up in the Subject Cataloging Manual: Subject Headings (SCM:SH). The Library of Congress has updated the instruction sheets for free-floating subdivisions and for subdivisions controlled by pattern headings. In this manual, subdivisions that once were coded $x (topical) which now should be coded $v (form) have been indicated by a diamond-mark, , to the left of the subdivision. However, by the time you read this, the SCM:SH 1999 Update no. 1 (spring 1999) should have arrived. The diamond-mark designation will no longer be used; instead, you will see the actual subfield code for all listed subdivisions. Here is an example, from SCM:SH 1154.5, for subdivisions used under Legal Topics headings:

$v Cases
$x Codification
$x Compliance costs (May Subd Geog)
$x Criminal provisions
$v Digests
$v Forms
$x Interpretation and construction
$x Language
$x Legal research
$x Legislative history
$v Popular works
$x Research (May Subd Geog)
$x Trial practice

(The 11th ed. of Free Floating Subdivisions: An Alphabetical Index is also being printed with the subfield codes added.) A simple analysis of the above subdivisions should convince us that the form ($v-coded) subdivisions represent what the work is, while the topical ones are used to show what the work is about. But wait, you might say, doesn't "--Legislative history" represent a particular kind or genre of material (i.e., compilations of texts of acts, public hearings, committee reports, etc.)? Most of the time, yes, the subdivision is used that way. However, LC has announced that, since the subdivision contains the word "history," they wish to be consistent with the practice being used for "--History" and for "--History and criticism," which are both being coded as $x topical subdivisions. The SCM:SH 1999 Update no. 1 subsequently shows the "--Legislative history" subdivision with a $x subfield code.

Form subdivisions always come at the end of the LCSH string, right? Not necessarily. There are some reasonable constructions where a topical subdivision must follow the form subdivision, to make the heading unambiguous. A common example is a dictionary (form) which is in a language other than that of the cataloging agency, e.g.:

Law $v Dictionaries $x German.

Both SAC and MARBI addressed this and other special situations when a separate coding for form data was initially proposed. Another question often asked was: what about those times when a term most often used as a form is actually the topic of the work being cataloged? (An example might be: "Law schools -- Statistics" for a discussion of statistics about law schools, not a compilation of the actual statistics.) The guideline we are to use is: code the subdivision for its function, not for its wording or its position in the string. LC will be creating two separate authority records for subdivisions that can often be used either way, one coded 180 $x and the other coded 185 $v. There are other instances where the "coding follows function." In the next example, the second heading clearly deals with treaties as a topic, while the work represented by the first heading actually contains the texts of the treaties:

Indians of North America $v Treaties
Indians of North America $x Treaties $x History

Hmm ... very interesting, but ... does this mean that "--Law and legislation" under topics, and "--Legal status, laws, etc." under classes of persons or ethnic groups will be coded $x when the work is about the law, but $v when the work contains the text(s) of the law? No. At the moment, LC has decided that these two subdivisions should be coded as $x (topical), which is thought to be the predominant characteristic of works for which the subdivisions have been used. It is possible that LC will consider adding an additional subdivision for form (perhaps "--Sources," or "--Texts," or ???) to such headings, when the work consists of or has substantial texts of laws, but LC would like to hear what the AALL suggests on this matter. (Please consult your reference librarians, and be prepared to discuss this at the TS-SIS cataloging committee and roundtable meetings this summer in Washington. Or, ask the chair of the Cataloging & Classification Committee if there will be a Working Group to study these issues, and volunteer to be a member!)