Category Archives: Uncategorized

The following resources can help you learn more about Resource Description and Access:

Functional Requirements for Bibliographic Records, Rev. Ed. 2011.

Functional Requirements for Authority Data, 2009.

Library of Congress RDA page and Library of Congress RDA Training Materials. These are the best Web sites for information on RDA.

Maxwell, Robert L. Maxwell’s Handbook for RDA. Chicago: ALA, 2014. The best book on RDA.

Oliver, Chris. Introducing RDA: A Guide to the Basics. Chicago: ALA, 2010. An excellent primer, especially on theoretical questions.

RDA Record Examples. Excellent examples of RDA MARC records for a variety of formats.

Stanford—RDA Questions and Answers. Very detailed.

Planet Cataloging. A great collection of cataloging-related blog posts.

 

RDA mandates many changes to authority records, because, according to the Functional Requirements for Authority Data, there is a great deal more information about persons and corporate bodies that needs to be recorded. Fortunately, most of the changes are additions to authority records, and the main structure of the authority record remains unchanged from its current configuration. The following is a brief review of the new authority MARC fields. It may be useful, in considering these changes, to look at sample RDA name authority records at the bottom of this page.  All the information on determining this information is in RDA. Chapter 9 of RDA provides instruction on “identifying persons,” and chapter 11 on “identifying corporate bodies.”

Full documentation on all these fields—including their many subfield codes—can be found at http://www.loc.gov/marc/authority/ecadhome.html. For most of the fields, there are “common” subfields for recording the source of the information in the field: $u is for a URI from which information is taken; $v is for any other source.

Also note that some of these fields are meant to be used only for work or expression authority records, which are not in common use now.

046 field: Special coded dates (e.g., birth and death date for a person, dates of a conference, etc.) Subfields used to specify the meaning of the date.

336 field: Content type. Just like the 336 field in bibliographic records. (Used for authority records for works and expressions only.)

368 field: Other corporate body attributes.  Used only for corporate body authority record. (E.g., “$a Television station”).

370 field: Associated place, with subfields for place of birth, death, residence, etc. LC gives following example for Ernest Hemingway, showing places of birth, death, and residence: $aOak Park, Ill. $bKetchum, Idaho $eOak Park, Ill. $eToronto, Ont. $eChicago, Ill. $eParis, France $eKey West, Fla. $eCuba $eKetchum, Idaho.

371 field: Address. Contact information for a person or corporate body. Many subfields.

372 field: Field of activity. For person or corporate body. Apparently, there isn’t a clear distinction between this field and 374. Stay tuned for changes.

373 field: Associated group. For person or corporate body. Formerly called Affiliation.

374 field: Occupation. Used only with personal name. Occupation is what a person is paid to do. Start and end dates may be given.

375 field: Gender. Used only with personal name. May be most useful for names not specific to one gender or foreign names.

376 field: Family information. E.g., type of family ($a Dynasty) or prominent family member. Used only with family name authority records.

377 field: Associated language. For language used by person/corporate body in the works it produces.

378 field: Fuller form of personal name. Used for personal name when, for example, preferred form is an abbreviation.

380 field: Form of work. May be used to differentiate works with same title. Used for authority records for works and expressions only.

381 field: Other distinguishing characteristics. Used for authority records for works and expressions only.

382 field: Medium of performance. Used only for music.

383 field: Numeric designation of musical work. Used only for music.

384 field: Key. Used only for music.

A major change from AACR2 is the introduction of three MARC fields: content type, media type, and carrier type, which go in fields 336, 337, and 338, respectively. These replace the information that in AACR2 was contained in the general material designator (GMD), which went in the “$h” subfield of MARC field 245. The GMD was considered imprecise and will no longer be in use in RDA records. A description of the three new fields follows.

Content type (field 336) is “a categorization reflecting the fundamental form of communication in which the content is expressed and the human sense through which it is intended to be perceived” (RDA 6.9.1.1). Got that? It is most useful to look at the list of content type terms here: http://www.loc.gov/standards/valuelist/rdacontent.html. (The terms are also listed in the RDA Toolkit.) These include terms such as “performed music,” “spoken word,” “text,” and “two-dimensional moving image.” This is a closed list, and catalogers should not use terms that are not on the list.

Media type (field 337) is “a categorization reflecting the general type of intermediation device required to view, play, run, etc., the content of a resource” (RDA 3.2.1.1). (Note that content type is in chapter 6 of RDA because it describes expressions; media type is in chapter 3 because it describes manifestations.) Basically, media type tells you what you need to do to access the content of the resource. If you have only to cast your eyes upon it, it is “unmediated.” Otherwise, it may be “audio,” “computer,” “video,” etc. The list of 10 prescribed terms can be found here: http://www.loc.gov/standards/valuelist/rdamedia.html. Again, this is a closed list.

Carrier type (field 338) is “a categorization reflecting the format of the storage medium and housing of a carrier in combination with the type of intermediation device required to view, play, run, etc., the content of a resource” (RDA 3.3.1.1). Chris Oliver adds that “[c]arrier types are closely correlated with media types and can be considered as the next level of granularity for the media types” (53). For example, a resource with media type “audio” might have carrier type of “audio disc” or “audiocassette.” A resource with media type “video” might have carrier type “videocassette” or “videodisc.” The (long) list of possible carrier types can be found here: http://www.loc.gov/standards/valuelist/rdacarrier.html. Again, it’s a closed list.

All three fields are repeatable. For example, a CD may be able to be played in an audio device as well as a computer, so it might be given two media types: “audio” and “computer.”

The following are some examples of how these three fields will look for different resources. A basic book would be coded in this way:

336 ## text $2 rdacontent
337 ## unmediated $2 rdamedia
338 ## volume $2 rdacarrier

The “$2” subfields are mandatory. A DVD would look like this:

336 ## two-dimensional moving image $2 rdacontent
337 ## video $2 rdamedia
338 ## videodisc $2 rdacarrier

An audiobook on CD would look like this:

336 ## spoken word $2 rdacontent
337 ## audio $2 rdamedia
338 ## audio disc $2 rdacarrier

And a Web site would look like this:

336 ## text $2 rdacontent
337 ## computer $2 rdamedia
338 ## online resource $2 rdacarrier

These fields are not display fields. An ILS will have to design a way for the information contained in the 336-338 fields to map to a display in the OPAC. The display may be very similar to the GMD we have today and may thus mean a relatively small change for catalog users. However, the new fields may have important effects on the faceted searching (the “buttons”) provided by the OPAC, if these searches use information from the GMD. Information on content and carrier, etc., in the fixed fields will not change.

A very important feature of RDA is the provision of relationship designators. These are terms added to entities in MARC records to specify the relationships between all the entities defined by FRBR/RDA. They replace the relator codes that were part of AACR2 but were infrequently used. Because relationships play such an important role in RDA, the relationship designators do as well.

Relationship designators that go after a name entity go in subfield $e (except when attached to the 111 and 711 [meeting names] fields, in which case they go in subfield $j). When the relationship designator records the relationship between works, expressions, manifestations, or items, it precedes the entity in the MARC field and goes in subfield “$i”.

Relationship designators are commonly attached to people and specify a relationship between a person and a resource. The terms to be used in this case are found in appendix I. For example, the record of a book written by Ed Smith, edited by Linda Jones and Frank Simmons, translated by Albert Wang, and illustrated by Leslie Brown might have the following fields:

100 1# Smith, Ed, $d 1967- $e author.
245 10 This old book / $c by Ed Smith
700 1# Wang, Albert, $e translator.
700 1# Brown, Leslie, $d 1981-, $e illustrator.
700 1# Jones, Linda $d 1965- $e editor.
700 1# Simmons, Frank $d 1945- $e editor.

All these personal name entries now specifically “relate” to the title in the 245 field. An example of relationship designators given for a film might be as follows:

700 1# Hecht, Ben, $d 1893-1964, $e screenwriter, $e director, $e producer.
710 2# Miramax Films, $e production company.
710 2# Miramax Home Entertainment (Firm), $e production company.

There are a wide variety of relationship designators. Some can give information about the physical book itself (this example is from LC).

700 1# Morrison, Jarrett $q (Jarrett Stephen), $d 1973- $e wood-engraver.
700 1# Rimmer, Jim, $e compositor.
710 2# Bowler Press, $e printer.
710 2# Press and Letterfoundry of Michael & Winifred Bixler, $e compositor.
710 2# Wookey Hole Mill, $e papermaker.

Relationships can also exist between works, expressions, manifestations, and items. The terms used to describe these relationships are found in Appendix J and include terms such as “abridgement of,” “remake of,” “adapted as,” “basis for libretto,” “concordance,” “supplement to,” and many, many others. The resource that is in relationship to the resource that is in the 1XX/245 fields goes in the 7xx field as an authorized access point with the relationship designator preceding it. (Some of this same information is found in the note fields in AACR2; notes are of course still allowed in RDA) Let’s consider a record for a movie of Pride and Prejudice. In AACR2, our record might include these fields:

500 $a Based on the novel by Jane Austen.
700 1# $a Austen, Jane, $d 1775‐1817. $t Pride and prejudice

In RDA we might find this:

700 1# $i motion picture adaptation of (work) $a Austen, Jane, $d 1775‐ 1817. $t Pride and prejudice

Here we see the relationship designator followed by the authorized access point for Pride and Prejudice. The relationship designators between WEMI can work reciprocally as well, so a record for the book Pride and Prejudice might have an added entry with “adapted as a motion picture” followed by the name of the film. Note the “$i” subfield for the relationship designator.

Sometimes, a relationship between two WEMI entities does not need a relationship designator to specify a relationship:

100 1# $a Shakespeare, William, $d 1564-1616.
245 10 $a Hamlet ; $b King Lear / $c William Shakespeare.
700 12 $a Shakespeare, William, $d 1564-1616. $t Hamlet.
700 12 $a Shakespeare, William, $d 1564-1616. $t King Lear.

Here, the second indicator “2” in the 700 fields indicates that this is an “analytical entry” – i.e., the resource in the 245 field “contains” the resources in the 700 fields.

The other types of relationships for which relationship designators may be used are those between people, families, and corporate bodies (Appendix K), which would be likely to go in name authority records, and relationships between subjects, which have not yet been set by RDA.

It should be noted that LC only requires one relationship designator: “illustrator” when used for an expression intended to children. But it is likely that we will see many more relationship designators than this.

The following are some of the smaller changes we can expect from the change from AACR2 to RDA.

The fixed fields. RDA MARC records will usually have an “i” in the “Desc” (Descriptive Cataloging Form, Leader/18) field. The “i” indicates that the record follows ISBD conventions. An RDA record will not have an “a” in the “Desc” field.

The 040 field. RDA MARC records will note “rda” under the “$e” subfield of field 040.

No more “rule of three”. AACR2’s “rule of three” stipulated that if there were more than three authors for any resource, only the first would be noted in the 245 field, followed by “[et al.], and that the title would become the authorized access point (i.e., there would be no 1xx field).  RDA does not retain this rule. All authors are listed in the 245 field, and the author with primary responsibility (or the first-named author) goes in the 1xx field, with additional authors in the 7xx field. The following example was created by Adam Schiff at the University of Washington libraries, about a book titled Managing Bird Damage to Fruit and Other Horticultural Crops coauthored by John Tracy, Mary Bomford, Quentin Hart, Glen Saunders, and Ron Sinclair. AACR2 would code the book in this way:

245 00  $a Managing bird damage to fruit and other horticultural crops / John Tracey … [et al.].

700 1_  $a Tracey, John Paul.

RDA says to do it this way this:

100 1_  $a Tracey, John Paul, $e author.

245 10  $a Managing bird damage to fruit and other horticultural crops / John Tracey, Mary Bomford, Quentin Hart, Glen Saunders, Ron Sinclair.

700 1_ $a Bomford, Mary, $e author.

700 1_ $a Hart, Quentin, $e author.

700 1_ $a Saunders, Glen, $e author.

700 1_ $a Sinclair, Ron, $e author.

Note that the 245 field in the RDA example has a first indicator of “1,” meaning that a title added entry will be made. We will discuss the relationship designators, in the $e subfield, later.

Uniform titles. The uniform title, which goes in the 240 field (or 130 field if it is the authorized access point), is now called the “preferred title.” There are several small changes to its use under RDA. When one work is published simultaneously in two languages, the preferred title is that of the first book received, not of the book published in the United States. This is in keeping with the international character of RDA.

In addition, when a new edition of a resource is made with revised content and a new title proper, the “preferred title” in the 240 will remain that of the first edition.

Another small change is that if two different languages of a work are represented within the same resource (for example, a dual-language translation), you should skip the 240 field (i.e., no more “Polyglot” or “French & Spanish” in the “$l” subfield) and put the preferred titles, with the “$l” language subfield, in the 700 fields for each language expression. We will see this infrequently.

Perhaps more important, “Selections” can no longer be used alone in a preferred title; it needs to be used with “Works.” A compilation of three or more works by Mark Twain, for example, would be coded in this way.

240 10 $a Works. $k Selections. $f 2001

245 10 $a Two novels and favorite essays / $c Mark Twain.

AACR2 would have left out the “Works.” However, “Poems. Selections” would still be appropriate in RDA.

There are various other small, but tricky, changes to how preferred titles are used in RDA, all to be found in RDA 6.2.2.8-6.2.2.10. The main thrust is that RDA more frequently uses conventional collective titles (e.g., “short stories,” “speeches,” “poems”) and is much more likely to give 7xx name-title added entries for all parts of a compilation.

Bias toward transcription. RDA’s mantra is “take what you see.” In most cases, the information from the preferred source should be transcribed. So if the title page reads “Third revised edition,” AACR2 would have this:

250 ## 3rd  rev. ed.

But RDA would have this:

250 ## Third revised edition

However, if the title page of the book reads “3^rd rev. ed.,” RDA instructs us to record it as is. Truncation of the names of publishers, also recommended in AACR2, is no longer used in RDA.

In general, RDA stipulates that catalogers not correct incorrect information but add a note about it. For example, obvious misspellings in the title are corrected in the 246 field. So, where AACR2 would have this:

245 ## Teusday’s [i.e.] Tuesday’s tasks

or perhaps

245   Teusday’s [sic] tasks

RDA would have this:

245 ## Teusday’s tasks

246 ## Corrected title: $a Tuesday’s tasks

However, errors in titles of serials should be corrected in the 245 field.

Abbreviations (Latin and otherwise). RDA does not use any Latin words or abbreviations. These include (but are not limited to) AACR2 standbys such as s.l. (sine loco), s.n. (sine nomine), sic, ca., and et al. That is, where AACR2 would have [s.l.], RDA prescribes [Place of publication not identified]; [s.n.] would be [publisher not identified]. However, RDA and LC strongly encourage catalogers to supply a probable date and place of publication, taken from outside the resource if necessary.

The lack of abbreviations means a different “look” to the catalog record. Be prepared to see names listed in this way:

Smith, John, approximately 1837-1893

Non-Latin abbreviations are also mostly prohibited. Use “born” instead of “b.”; “illustrations” instead of “ill.”; “century” instead of “cent.,” etc. For example, where AACR2 would have this:

300 ## $a 86, [21] p. :$b ill., 1 folded map ;$c 24 cm.

RDA would have:

300 ## $a 86 pages, 21 unnumbered pages :$b illustrations, 1 folded map ;$c 24 cm

Note the change in how unnumbered pages are listed.

Allowed abbreviations include “cm” (without a period) and abbreviations for time such as “min.” and “sec.” (useful in the 300 field for a DVD/CD). Remember that abbreviations are permitted when the preferred source of information has them.

Preferred sources of information. AACR’s “chief source” of information is now the RDA “preferred sources.” RDA permits catalogers to seek information about a resource—e.g., its title or date—from a broader range of sources than AACR2 does. RDA 2.2 lists the preferred sources of information for different types of resources. If the information cannot be found in the preferred sources, then RDA (2.2.4) permits catalogers to seek information outside the resource itself. Information taken from outside the resource should be enclosed in brackets.

The 264 field. In MARC21, the 260 field has traditionally been used to code information on publication and distribution. Recently, LC has recommended that the 264 field take over the duties of the 260 field. The 264 field is meant to satisfy RDA’s requirement that bibliographic records distinguish information about publication from that about copyright, production, manufacture, and distribution. The second indicators for the 264 field are as follows:

0 – Production

1 – Publication

2 – Distribution

3 – Manufacture

4 – Copyright notice date

RDA—and LC—have a complex set of rules for when to use the distribution, manufacture, and production information. Very clear instructions can be found in LC’s “RDA: Module 1” in the list of LC training materials (see “Resources” section below). Also see RDA 2.7-2.11 for detailed instructions. In short, distribution information should be given when publication information is not available, manufacture information should be given when publication and distribution information are not available, and production information should be given for unpublished resources. Publication and copyright dates should always be given. Because 264 is a repeatable field, there will be many records that look like this (example is taken from LC):

264 #1 North Vancouver, BC :$b The Bowler Press,|c 2008.

264 #4 $c ©2008

Note that the second indicator 4 means that the date is a copyright date. Also, the copyright symbol (i.e., ©) should be used instead of “c.” If the resource lists only a copyright date, and it is reasonable to assume that the publication date is the same, the publication date should be put in brackets. Information on the source of publication information can go in the 588 field.

Do not privilege “home country.” Another change is that catalogers should record the first listed place of publication where more than one are listed. There is no instruction to use a place found in the catalog’s “home country.” RDA has an international focus.

 

The Bible. The major change in how the Bible is cataloged is that “O.T.” or “N.T.” will no longer be used to indicate whether a book is in the Old or New Testament. So we will have

Bible. Ezra.

Bible. Matthew.

If the resource is an entire Testament, it will be spelled out:

Bible. New Testament.

The Koran. Koran is now spelled Qur’an.

The 300 field. The 300 field is now repeatable. This may be useful for a book that has an accompanying CD.

The 775 and 776 fields. These fields are recommended by LC for coding reproductions. The 775 field is used when the reproduction is in the same carrier; the 776 field is used when it is in a different carrier. LC gives the following example:

240 10 Novels. $k Selections

245 10 André ; $b Leone Leoni / $c  George Sand.

776 08 $i Online version: $a Sand, George.  $s Novels. Selections. $t André  $d Paris : Honoré Champion Éditeur, 2011, ©2011 $w (OCoLC)744525681

“Online version” in the $i field is the relationship designator. See the section on “Relationship designators.”

Squaring the new RDA standard with the Machine-Readable Cataloging (MARC) encoding format is a source of much concern and confusion. RDA is not an encoding standard; it is a content standard. It makes no mention of MARC, nor did the creators of RDA have MARC “in mind” when they drew up RDA. Clearly put, RDA says a lot about what information to record, a little about how to record it, and nothing at all about where it goes or how to encode it. One problem created by this situation is that a typical MARC bibliographic record is a “mash-up” of FRBR group 1 entities (i.e., WEMI). It may contain in its fields data referring to the work, expression, manifestation and item of a particular resource. The powerpoints “MARC 21, FRBR, RDA,” prepared by the LC, and “MARC 21, RDA, and the FRBR and FRAD models,” prepared by OCLC, both noted in the resources section of this document, are good sources of information on how RDA maps into MARC.

Despite these resources, it is not always easy or straightforward to map RDA instructions to MARC fields. Consider a standard MARC bibliographic record. Some fields, such as 260—publication information—and 300—physical description—clearly refer to the manifestation of a work (as does the title proper in the 245 field). The creator in the 100 field is in relationship to the work, but a translator or illustrator listed in field 700 would be related to an expression. In the same way, the uniform title (now “preferred title”) in the 240 field is the work, but the language—the $l subfield of the 240 field—points to an expression. Understanding these distinctions can help you clarify the practical side of RDA’s theoretical foundation and also point you toward the relevant section of the RDA instructions.

In a fully RDA-compliant catalog, it is possible that there would be individual “records” for expressions and (more likely) works, thus separating the “mashed-up” information now in the standard bibliographic record. It remains to be seen whether these sorts of separate records will ever be widely adopted in MARC. For the moment, the RDA entities will be all mixed in the bibliographic and authority records of a standard catalog. The long-term future of MARC is (and has been for a long time) up in the air, but the reality is that we will be using MARC for some time to come, and RDA will have to work within it.

One big change from AACR2 to RDA is in the organization of the instructions. Where AACR2 was organized by type of material, RDA is organized according to the FRBR entities. RDA has 10 sections and an introduction, as follows:

0:            Introduction
Section 1: Recording Attributes of Manifestation & Item
Section 2: Recording Attributes of Work & Expression
Section 3: Recording Attributes of Person, Family, & Corporate Body
Section 4: Recording Attributes of Concept, Object, Event & Place
Section 5: Recording Primary Relationships Between Work, Expression, Manifestation, & Item
Section 6: Recording Relationships to Persons, Families, & Corporate Bodies
Section 7: Recording Relationships to Concepts, Objects, Events, & Places
Section 8: Recording Relationships between Works, Expressions, Manifestations, & Items
Section 9: Recording Relationships between Persons, Families, & Corporate Bodies
Section 10: Recording Relationships between Concepts, Objects, Events, & Places

Note that each section gives instructions on recording either the attributes of an entity or the relationships that exist between entities. Each section is divided into several chapters—there are 37 in all—although at this time, all of sections 4, 7, and 10 (those sections dealing with subjects) have not been written. Section 1 (on manifestations and items), for example, gives instructions on recording information unique to manifestations and items—e.g., edition, number, publication and production information, copyright, mode of issuance, etc., as well as information on the carrier (i.e., the format) and media (i.e., the device needed to play the resource) of the resource. Section 2 (on works and expressions) contains information on recording attributes of works and expressions, which may include title, date, content, and language.

Sometimes it can be difficult to know where in RDA to turn for a particular instruction. What is the difference, for example, between sections 3 and 6? It’s something of a fine distinction. Section 3 has information on how to define and identify a person (or family or corporate body). How do we determine the preferred name? How do we deal with official titles or with popes and kings or names in non-Latin scripts? Section 6, on the other hand, gives information on how to determine who is the person or persons or body or bodies that are responsible for creating a particular work. The information in section 3 may be more useful for those creating or modifying name authority records, where section 6 may be more useful for those working with bibliographic records. Think of it this way: the person who in section 3 is a “person” is a “creator” in section 6 because section 6 concerns his or her relationship to a resource.

It is important to have some understanding of these distinctions in order to know where to look in the RDA instructions. I have found that chapter 2, “Identifying Manifestations and Items”; chapter 6, “Identifying Works and Expressions”; and chapter 19, “Persons, Families, and Corporate Bodies Associated With a Work” are particularly useful in working with bibliographic records.

Seymour Lubetsky wrote in 1969 that “the book…comes into being as a dichotomous product–as a material object or medium used to convey the intellectual work of an author.” This dichotomy between the physical book and the intellectual work is at the center of Resource Description and Access. From this insight has arisen a hierarchy of bibliographic entities (more on this word later) that attempts to “comprise[] the products of intellectual or artistic endeavor that are named or described in bibliographic records” (FRBR, 13). The entities that make up this hierarchy are called work, expression, manifestation, and item—or WEMI for short. Any resource may be considered at any stop along this hierarchy. The work is the intangible intellectual product; the expression is the physical realization—using alphanumeric symbols, for example—of the work; the manifestation is a particular embodiment of an expression; an item is a single, physical instance of a manifestation.

Let’s see how this works in practice. Take a well-known book, Pride and Prejudice by Jane Austen:

The work is Pride and Prejudice, as fully conceived in the mind of Jane Austen.

The expression is the realization of the work in, for example, text or spoken word. There are many ways a work can be expressed, and it can be difficult to determine what constitutes a separate expression, and what is a new work entirely; cataloger judgment is required. In general, translations, revisions, versions, illustrated editions, abridgements, and illustrated editions would be considered unique expressions of the same work. So a translation of Pride and Prejudice into Russian would be considered an expression of the work Pride and Prejudice. An illustrated version (in some cases) would be a new expression. If Austen had revised her novel, the revised version would be a separate expression as well. An annotated version of the work would also probably be considered a separate expression, as long as the annotations substantially change the expression (i.e., not just a couple of footnotes; if, however, the annotations are extensive and are considered the major part of the work, then it may constitute a separate—but related—work). An abridged version would probably be a separate expression, but a summary or adaptation (as, for example, a film version of Pride and Prejudice) would be its own separate work. Pride and Prejudice and Zombies would obviously be a separate work, though related to Pride and Prejudice by being a satire of it. Most works will have only one or two expressions—a text version and a spoken word version, say—but a great classic like Pride and Prejudice is likely to have many expressions.

An example of a manifestation is Pride and Prejudice, published by Penguin Books in 1989, with an ISBN number of 834935764-0. A single expression may have many manifestations, especially if (as with Pride and Prejudice) the book is frequently reprinted. Catalogs traditionally have separate records for separate manifestations.

The item is the book on the shelf, with a particular barcode, provenance, and circulation history.

Work, expression, manifestation, and item are all defined by Functional Requirements for Bibliographic Records as entities. FRBR defines a total of 11 entities, separated into three groups. Group 1 entities are those described above: work, expression, manifestation, and item. Group 2 entities are person, corporate body, and family. Group 3 entities are concept, object, event, and place. (These groups correspond to the traditional title-author-subject catalog divisions.)

Any one of these entities can have multiple attributes and can also have relationships with other entities. Another way to think of RDA is thus as a large entity-relationship model, composed of entities, their attributes, and the relationships between them. Let’s see what this means in practice. A work (entity) is created by (relationship) a person (entity) with particular birth and death dates (attributes). The work (entity) is also realized through (relationship) one or more expressions (entities), which in turn are embodied in (relationship) manifestations (entities), each of which is published by (relationship) a publisher (entity) located at a particular address (attribute). In addition, the work (entity) may be an adaptation of (relationship) another work (entity) and may have another expression (entity) as its supplement (relationship). And so forth. RDA’s goal is to record all these entities, attributes, and relationships. Visit www.rdatoolkit.org/background for many entity-relationship diagrams related to RDA. See “Relationship designators” below for more information on how RDA records relationships between entities.