Authority File Comparison Rules (NACO Normalization)

Based on the report of the Task Group on Normalization of the PCC Standing Committee on Automation, approved by the PCC Policy Committee in November 2007 and further revised April 1, 2009.

When a new authority record is added to an authority file, each heading is compared against the headings and references already in the file to determine whether the new headings and references are allowable, that is, adequately differentiated from existing headings. The headings and references already in the file and the ones to be added are normalized before comparison so that only certain characters will be considered in the comparison.  The normalization rules that apply to characters are specified in Appendix A. The rules for heading comparison are given below.

Rules for Comparison

Note the following before applying the rules for comparison:

1. The following rules apply after characters have been subjected to the character normalizations described in Appendix A.
2. The rules apply to all name authority records that are part of the NACO program: names, corporate bodies, meetings, etc., titles/series, and name/titles.  Name authority records, however, are not compared against subject authority records.
3. The rules apply to LCSH subject authority records (008/11=a) that represent established headings (008/09=a) and subdivisions (008/09=d).  LCSH subject authority records, however, are not compared against name authority records. 
4. For the purposes of these rules, 1XX fields in LCSH subject authority records for untraced and traced references (008/09=b or c) are not treated as established headings but as see references, so they should be regarded as 4XX fields in the comparisons below. 
5. LC subject headings for children’s literature (008/11=b) or records from other subject heading systems are excluded from these comparisons.
6. Established heading linking entries (7XX fields) are not compared.
7. Except as noted in 6) below, the tags of the fields do not form part of this comparison, but subfield delimiters and codes are included.  Certain subfields (i.e., $i and $w in 4XX and 5XX) are eliminated before comparison.

1. An established heading (1XX field) must not compare the same as any other established heading, except: an established topical heading (150 field) in one subject authority record may compare the same as an established form/genre heading (155 field) in another subject authority record.
2. A see reference tracing (4XX field) must not compare the same as any established heading (1XX field) in any record.
3. A see reference tracing (4XX field) must not compare the same as any see also reference tracing (5XX field) in any record.
4. A see reference tracing (4XX field) must not compare the same as another see reference tracing (4XX field) in the same authority record.
5. A see reference tracing (4XX field) in one authority record may compare the same as a see reference (4XX field) in another authority record.
6. A see also reference tracing (5XX field) must have the same second and third tag characters and compare the same as an established heading in another authority record.  For example, a 500 field in one authority record must match a 100 field in another authority record.  This rule means that every 5XX field must have a matching 1XX field, but not all 1XX fields will have a matching 5XX field.

Notable Changes in this Revision

1. Subfield codes are now considered as part of the comparison.  This allows the same term in a 150$a and 180$x to occur without conflict because the subfield codes ‘a’ and ‘x’ will make the strings unique.
2. A see reference tracing (4XX field) may not compare the same as a see reference tracing (4XX field) in the same authority record; this was allowed under the earlier rules, although the practice wasn’t followed in LCSH records.

Appendix A

Note: The characters referred to below correspond to the MARC 21 Latin-based character sets, Basic Latin (ASCII) and Extended Latin (ANSEL).

Character Normalization Tables
General Characters
Character	Character	Comments
Leading blanks	Delete
Trailing blanks	Delete
Multiple blanks	Delete/Blank	Compress to a single blank
Lowercase letters	Uppercase equivalent
Uppercase letters	Retain
Numbers	Retain

Modifying Diacritics
Character	Character	Comments
Acute Breve Candrabindu Cedilla Circle above, angstrom Circle below Circumflex* Dot below Double acute Dot below Double tilde (first/second half) Double underscore Grave* Hacek High comma centered High comma off center Left hook Ligature (first/second half) Macron Pseudo question mark Right cedilla Right hook, ogonek Superior dot Tilde* Umlaut, diaeresis Underscore* Upadhmaniya	Delete	*Do not confuse with spacing character equivalents (see Other Special Characters below)

Back to Top

Translated Characters
Character	Character	Comments
Superscript numbers	Numbers	Convert to non-superscript equivalent
Subscript numbers	Numbers	Convert to non-subscript equivalent
Diagraph AE	AE	Upper and lower case
Diagraph OE	OE	Upper and lower case
D with crossbar	D	Upper and lower case
Eth	D	Upper and lower case
Lowercase Turkish i	I
Polish L	L	Upper and lower case
Script small L	L
O Hook	O	Upper and lower case
U Hook	U	Upper and lower case
Scandinavian O	O	Upper and lower case
Icelandic Thorn	TH	Upper and lower case
Eszett symbol	SS	Do not use in NACO records
Greek alpha	Uppercase Greek alpha (U + 0391)	Do not use in NACO 1XX fields
Greek beta	Uppercase Greek beta (U + 0392)	Do not use in NACO 1XX fields
Greek gamma	Uppercase Greek gamma (U + 0393)	Do not use in NACO 1XX fields

Back to Top

Punctuation
Character	Character	Comments
Exclamation mark	Blank
Quotation mark	Blank
Apostrophe	Delete
Opening parenthesis	Blank
Closing parenthesis	Blank
Hyphen, minus-	Blank
Opening square bracket	Delete
Closing square bracket	Delete
Opening curly bracket	Blank
Closing curly bracket	Blank
Less-than sign	Blank
Greater-than sign	Blank
Semicolon	Blank
Colon:	Blank
Period, decimal point	Blank
Question mark	Blank
Inverted question mark	Blank
Inverted exclamation mark	Blank
comma	Comma or blank	The first comma in $a is retained; all other converted to blank

Back to Top

Other Special Characters
Character	Character	Comments
Music flat sign	Retain
Number sign	Retain
Slash	Blank
Reverse slash	Blank
Commercial at sign	Retain
Ampersand	Retain
Asterisk	Blank
Vertical bar (fill)	Blank
Percent	Blank
Equals sign	Blank
Plus sign	Retain
Plus or minus	Blank
Superscript plus, minus	Blank
Patent mark	Blank
Sound recording copyright	Blank
Copyright sign	Blank
Dollar sign	Retain
British pound	Retain
Degree sign	Blank
Spacing circumflex	Blank
Spacing underscore	Blank
Spacing grave	Blank
Spacing tilde	Blank
Euro sign	Retain	Do not use in NACO records
Music sharp sign	Retain
Alif	Delete
Ayn	Delete
Hard sign	Delete
Soft sign	Delete
Middle dot	Blank

Back to Top

Content Designation
Character	Character	Comments
Indicators	Delete
Subfield delimiters	Retain
Subfield codes	Retain
Tag		See comparison rules
Joiner, Non-joiner, Unicode Formatting Characters	Delete

Back to Top

Changes in this Revision
Character	Old Conversion	New Conversion
Commercial at sign	Blank	Retain
Vertical bar	Delete	Blank
Dollar sign	Blank	Retain
British pound sign	Blank	Retain
Middle dot	Delete	Blank
Greek alpha	A	Uppercase Greek alpha
Greek beta	B	Uppercase Greek beta
Greek gamma	G	Uppercase Greek gamma
Subfield codes	Delete	Retain

Back to Top

Last update: Apr. 9, 2020