The Getty
Research Home Search Tools & Databases Learn about the Getty Vocabularies Editorial Guidelines Art & Architecture Thesaurus Online
Art & Architecture Thesaurus Online
Art & Architecture Thesaurus (AAT): Editorial Guidelines



Purpose of these Guidelines
Purpose of the AAT

    Introduction and Overview (PDF, 8.9 MB, 223pp)

    1.1.1 Scope and Structure
    1.1.2 What is a Thesaurus?
    1.1.3 What is a "Concept" in the AAT? (warrant)

    1.2.1 Review process
    1.2.2 Does contributors' data follow editorial rules?

    1.3.1 Web browsers
    1.3.2 Licensed files
    1.3.3 Sample Displays of AAT Data

    1.4.1 Database
    1.4.2 Merged Records
    1.4.3 Operating VCS


    2.1.1 Following the rules
    2.1.2 Required fields and minimal records
    2.1.3 Format and values
    2.1.4 Capitalization and abbreviation
    2.1.5 Language of the Record
    2.1.6 Production goals
    2.1.7 Leaving unfinished records overnight
    2.1.8 Quality control
    2.1.9 Avoid plagiarism
    2.1.10 Uncertainty and ambiguity in display fields
    2.1.11 Uncertainty and ambiguity in indexing fields
    2.1.12 Uncertain identification of a concept

    2.2.1 Rules for merging
    2.2.2 Procedures for merging

    2.3.1 Rules for moving
    2.3.2 Procedures for moving

    2.4.1 Sample AAT record
    2.4.2 Sample AAT record in VCS

    2.5.1 About the fields
    2.5.2 List of VCS Fields


    3.1.1 Parents (required)
    3.1.2 Sort Order
    3.1.3 Historical Flag: Current or Historical parents and other flags
    3.1.4 Dates for relationship to parents
    3.1.5 Parent String
    3.1.6 Facet or Hierarchy Code

    3.1.7 Relationship Type for Hierarchy

    3.2.1 Subject ID
(required default)
    3.2.2 Parent Key
    3.2.3 Merged Status
    3.2.4 Published Status
    3.2.5 Review Status
    3.2.6 Record Type
    3.2.7 Candidate Status
    3.2.8 Label
    3.2.9 Contributors for the Subject Record
    3.2.10 Sources for the Subject Record

    3.3.1 Term ID
    3.3.2 Term
    3.3.3 Preferred Flag
    3.3.4 Qualifier
    3.3.5 Sequence Number
    3.3.6 Historical Flag
    3.3.7 Term Type
    3.3.8 Part of Speech
    3.3.9 Vernacular Flag
    3.3.10 Language for Terms(required-default)
    3.3.11 Preferred Flag for Language
    3.3.12 Language Status
    3.3.13 Contributor for Term
    3.3.14 Preferred Flag for Contributor
    3.3.15 Sources for Terms
    3.3.16 Page Number for Term Source
    3.3.17 Preferred Flag for Source
    3.3.18 Dates for Terms
    3.3.19 Display Term Flag
    3.3.20 LC Flag (formerly AACR flag)

    3.3.21 Other Flags

    3.3.22 Assigned To

    3.4.1 Scope Note
    3.4.2 Sources for the Scope Note
    3.4.3 Contributor for the Scope Note

    3.5.1 Related Concepts
    3.5.2 Relationship Type
    3.5.3 Historical Flag
    3.5.4 Dates for Related Concepts

[Note: There are no sections 3.6 and 3.7 for the AAT Guidelines]

    3.8.1 Comment Flag
    3.8.2 Problem Flag
    3.8.3 Assigned To
    3.8.4 Special Project
    3.8.5 Facet Code
    3.8.6 Legacy ID
    3.8.7 Class Notation
    3.8.8 Image
    3.8.9 Index Note
    3.8.10 Not Found Note
    3.8.11 Status Note
    3.8.12 Editor Note
    3.8.13 Revision History

    4.1.1 How to Use Diacritical Codes
    4.1.2 Diacritical Codes: Quick Reference
    4.1.3 Diacritical Codes: Full List

    4.2.1 How to Record Dates
    4.2.2 How to Use the Date Authority
    4.2.3 Date Authority

    4.3.1 How to Record Sources
    4.3.2 Rules for Sources
    4.3.3 Merging Sources

    4.4.1 How to Record Contributors
    4.4.2 Contributing Large Translations

    4.5.1 How to Record Languages
    4.5.2 List of Languages


Guidelines for Multilingual Equivalency Work



    compiled and edited by
    Patricia Harpring, managing editor, and
    the Getty Vocabulary Program:
    Antonio Beecroft, editor
    Robin Johnson, editor
    Jonathan Ward, editor

    Revised: 22 November 2022


This document contains information about editorial practice for the Art & Architecture Thesaurus (AAT) ®, one of the vocabularies produced by the Getty Vocabulary Program. The other vocabularies are the Getty Thesaurus of Geographic Names (TGN)®, Union List of Artist Names (ULAN)®, Cultural Objects Name Authority (CONA)®, and Iconography Authority (IA)™.

NOTE: The guidelines in this document are subject to frequent modification and addition.


Purpose of these guidelines
This document contains rules and guidelines intended for use by contributors and editors of the Getty Vocabularies. Developers and other implementers of the vocabulary data may consult these guidelines to extrapolate information and guidance for implementation of the Vocabularies.


Purpose of the AAT
The AAT, TGN, ULAN, CONA, and IA are structured vocabularies that can be used to improve access to information about art, architecture, and material culture. The Getty Vocabularies are not simple 'value vocabularies'; they are unique knowledge bases in themselves. Through rich metadata and links, the Getty Vocabularies provide powerful conduits for knowledge creation, complex research, and discovery for digital art history and related disciplines.

  • Cataloging: For some users, the Getty Vocabularies are utilized as data value standards at the point of documentation or cataloging, to promote consistency in assignment of a term or to provide options among multiple terms referencing the same concept. The Vocabularies provide preferred names/terms and synonyms for people, places, and things. They also provide structure and classification schemes that can aid in documentation.

  • Linking: For other users, the Getty Vocabularies are used in linking, in order to reference the unique identifier of the Vocabulary record, or to otherwise reconcile their data.

  • Retrieval: For other users, the Getty Vocabularies aid in retrieval and discovery, utilizing synonymous terms, broader/narrower contexts, and other rich contextual data in search assistants, in database retrieval systems, and more broadly in a linked environment. The Vocabularies are rich knowledge bases that contain dozens of fields of rich contextual data about each concept, and semantic networks that highlight links and paths between concepts.

  • Research tools: For other users, the Getty Vocabularies are used as look-up resources, valuable because of the rich information and contextual knowledge that they contain.

  • In order to meet the needs of these various user communities, the Getty Vocabularies are made available in several ways.

  • Data formats: Releases include Linked Open Data (LOD) (JSON, RDF, N3/Turtle, N-Triples for GVP and Linked.Art), XML, Relational Tables, Web Services APIs. These files are used by developers or incorporated in various tools by vendors or others. These releases may be transformed to comply with other formats, such as the MARC format used for ULAN and TGN in the Virtual International Authority File (VIAF). Some of these releases contain simplified versions of the data, while others contain the full, rich data sets, providing versions to meet the requirements of various developer communities. The AAT, TGN, and ULAN are available as LOD, relational tables, and XML. AAT, TGN, ULAN, CONA, and IA are available through APIs. The data file releases are refreshed periodically throughout the year. The Getty Vocabularies are published under the Open Data Commons Attribution License (ODC-By) 1.0.
  • Online search: The five Getty Vocabularies' online search pages are consistently the top sites visited at the Getty Research Institute Web site each month. Using these search tools, catalogers copy-and-paste Vocabulary terms and IDs as part of their daily workflow. Researchers use the search to locate rich information about the Vocabulary concepts. In the results displays, for each concept the data fields are presented in a logical full-record display for end users, as well as in hierarchical views. The online search data is refreshed monthly.

    Various releases or utilizations of the Getty Vocabulary data may contain more or less of the full, available data for each Vocabulary record, depending upon the purpose of the release. An implementation that intends to ask complex queries using the Vocabulary data would require the full available data. In another example, if a developer only needs to link to the unique identifier for the concept, perhaps a streamlined data set would be more appropriate; e.g., the ULAN data (and soon TGN data also) that is included in the VIAF (Virtual International Authority File) is a subset of the full data available, which is streamlined and parsed to fit the particular requirements of MARC.

Catherine wheel or rose window? AAT is a structured vocabulary, including terms, descriptions, and other metadata for generic concepts related to art, architecture, conservation, archaeology, and other cultural heritage. Included are work types, styles, materials, techniques, and others.

For further discussion of the history and scope of the AAT, see About the AAT.


The Getty Vocabularies are copyrighted: Copyright © J. Paul Getty Trust, released under Open Data Commons Attribution License (ODC-By) 1.0. The data is available in a Web search interface and in various formats, including Linked Open Data, XML, Relational Tables, and APIs.


AAT is a compiled resource; it is not comprehensive. It grows over time to become gradually more comprehensive, to reflect changes in the development and usage of language and terminology, to accommodate new research in art history and archaeology, and to become ever more multilingual and multicultural.

The AAT grows through contributions. Information in the AAT was compiled by the Getty Vocabulary Program (and earlier groups in Williamstown, Massachusetts) in collaboration with many institutions. Institutions interested in becoming contributors to the AAT should write to, explaining the scope of their collections and likely contributions. For translation projects, see Contributing Large Translations.


For further information, please contact

the Getty Vocabulary Program

Getty Vocabulary Program
1200 Getty Center Drive, Suite 1100
Los Angeles, CA 90049







General Information about the AAT

For a briefer general introduction, see AAT Introduction and Overview (PDF, 8.9 MB, 223pp).

For translating the AAT, please see Guidelines for Multilingual Equivalency Work.





Scope and Structure



Scope of AAT
The AAT includes generic terms, and associated dates, relationships, and other information about concepts related to or required to catalog, discover, and retrieve information about art, architecture, and other visual cultural heritage, including related disciplines dealing with visual works, such as archaeology and conservation, where the works are of the type collected by art museums and repositories for visual cultural heritage, or that are architecture. It is our goal to be ever more inclusive of various cultures and their visual works. Also, in recognition of diverse collections found in art museums, the AAT contains terminology to describe objects and associated activities that are ceremonial or utilitarian in nature, but are not necessarily labeled as art according to traditional Western aesthetics.

The AAT is used for work types, roles, materials, styles, cultures, techniques, subject, and other fields in a work record, so long as terms fit into established AAT facets.

What is excluded in AAT? In general, terminology from domains other than visual arts is excluded from AAT, except to the extent that such terminology is necessary to catalog or discover information about visual works, including new media art, conceptual art, and performance art (as distinct from the performing arts).

Excluded are proper names and iconographical subjects for the domain of art and architecture, but which are included in the other Getty Vocabularies: examples are names of people (ULAN), geographic names (TGN), titles of works of art (CONA), or iconographical topics (IA). Also excluded from AAT are unbound compound terms, which combine words from distinct hierarchies; for example, Baroque cathedral is an unbound compound term combining terms for a style and for a work type. Excluded are phrases or words that do not have literary warrant, organizational warrant, or user warrant. Overall, records that lack the minimal information for an AAT record are excluded; see guidelines for contributions for more information regarding minimum contribution requirements.



Structure of the Data
The focus of each AAT record is called a concept. In the database, each concept's record (also called a subject in these guidelines) is identified by a unique numeric ID. Linked to each concept record are terms, related concepts, a parent (that is, a position in the hierarchy), sources for the data, and notes.

  • The AAT is a hierarchical database; its trees branch from a root called Top of the AAT hierarchies (Subject_ID: 300000000). There may be multiple broader contexts, making the AAT polyhierarchical. In addition to the hierarchical relationships, the AAT has equivalence and associative relationships. AAT is a thesaurus, in compliance with ISO and NISO standards.

  • Unpublished facets in AAT are used for candidate records. Unpublished facets and hierarchies are designated by the "name" temp.parent (e.g., temp.parent/candidate records)



Facets and heirarchies
Facets constitute the major subdivisions of the AAT hierarchical structure. A facet contains a homogeneous class of concepts, the members of which share characteristics that distinguish them from members of other classes. For example, marble refers to a substance used in the creation of art and architecture, and it is found in the Materials facet. Impressionist denotes a visually distinctive style of art, and it is found in the Styles and Periods facet.

  • The conceptual framework of facets and hierarchies in the AAT is designed to allow a general classification scheme for art and architecture. The framework is not subject-specific; for example, there is no defined portion of the AAT that is specific only for Renaissance painting. Terms to describe Renaissance paintings will be found in many locations in the AAT hierarchies.




What is a Thesaurus?

  • The AAT is a thesaurus. A thesaurus is a semantic network of unique concepts, including relationships between synonyms, broader and narrower (parent/child) contexts, and other related concepts. Thesauri allow three types of relationships: equivalence (synonym), hierarchical (whole/part or genus/species), and associative. Thesauri may be monolingual or multilingual.

  • The structure of the AAT supports multilinguality insofar as terms and scope notes may be written and flagged in multiple languages.





Thesauri may have the following three relationships:





Equivalence relationships
The relationships between synonymous terms or names that refer to the same concept, typically distinguishing preferred terms (descriptors) and non-preferred terms (variants, or ALTs and UFs).





Hierarchical relationships
Broader and narrower (parent/child) relationships between concepts. Hierarchical relationships are generally either whole/part or genus/species; in the AAT, most hierarchical relationships are genus/species (e.g., chalice is a type of drinking vessel). Relationships may be polyhierarchical, meaning that each child may be linked to multiple parents.





Associative relationships
The relationships between concepts that are closely related conceptually, but the relationship is not hierarchical because it is not whole/part or genus/species.





What is a "Concept" in the AAT?

  • Terms in the AAT represent abstract concepts, physical attributes such as shape, pattern, and color, style or period, activities, performers of activities, materials, objects, and visual and verbal communication forms discussed within the literature of art, architecture, and material culture. In the context of the AAT, a "concept" (or "subject record") represents terminology needed for cataloguing art, excluding proper names; thus it can be described as containing information about generic concepts (as opposed to proper nouns or names).

    The AAT may include terminology to describe the type of work (e.g., sculpture), its material (e.g., bronze), activities associated with the work (e.g., casting), its style (e.g., Art Nouveau), the role of the creator (e.g., sculptor), and other attributes or various abstract concepts (e.g., symmetry).

    AAT does not include proper names of persons, organizations, geographic places, named subjects, or named events.





Warrant for a concept
A major issue in creating AAT records is how to determine if a word or words truly represent a definable, unique "concept" in common and scholarly usage, or if it is simply a string of words (which would not be included in the AAT). A concept is single word or multiple words that are used consistently to refer to the identical generic concept, type of work, material, activity, style, role or other attribute.

  • In order to determine if the words are truly accepted by the community, that they consistently represent a definable concept, and that the preferred term (descriptor) is the term most often used to refer to this concept, the AAT requires three pieces of literary warrant for the preferred term. See 3.3 Terms for further discussion.





Discrete concepts
A "concept" in the context of the AAT is a discrete thing or idea. Records in the AAT generally represent discrete concepts, not subject headings. In contrast to a discrete concept, a subject heading typically concatenates multiple terms or concepts together in a string. For example, Pre-Columbian sculptures is a heading composed of terms representing two discrete concepts: Pre-Columbian (a style and period) and sculpture (a type of work). Pre-Columbian as a style and period term may be combined with many other terms and retain its meaning; sculpture also could be combined with many other style or period terms and still retain its meaning.

  • Note that a term for a concept is not necessarily composed of only one word; terms describing discrete concepts include the following: rose windows, flying buttresses, book of hours, High Renaissance, and lantern slides. We maintaining discrete concepts, as opposed to headings or compound terms, in the structure of the AAT in order to make it more versatile in cataloguing and more powerful in retrieval.





Range of coverage
The AAT includes the following types of terms: Associate Concepts (abstract concepts and phenomena, e.g., beauty, balance), Physical Attributes (characteristics and features, e.g., strapwork, borders), Styles and Periods (e.g., Xia, French), Agents (e.g., printmakers, landscape architects) and Living Organisms, Activities (e.g., archaeology, engineering), Materials (e.g., iron, clay), and Objects (e.g., paintings, amphorae). Brand Names are also included, especially as required to support indexing for the Conservation community. See 3.1 Hierarchical Relationships for a fuller discussion regarding what is included in each facet.





What is excluded from the AAT?
The AAT contains terminology for "concepts"; see What is a "Concept" in the AAT above. The AAT contains no proper names; therefore specific named buildings, individual people, named corporate bodies, and named historical events are out of scope for the AAT. All terms in the AAT must refer to a case of many (generic things), not a case of one (specific things). In contrast, a proper name refers to a unique thing: a case of one. The scope of the AAT is further narrowed to only concepts having to do with art, architecture, and related disciplines.





Contributions and Editorial Control









Review Process

  • Contributions must meet the following criteria: they must be submitted by an authorized contributor; must be within scope of the vocabulary; must include the minimum information; and must be submitted in one or the prescribed formats. By contributing data to the Vocabulary, the contributor agrees to its contributed data becoming a permanent part of the Vocabulary, which is made available in several formats, including as Linked Open Data. See How to Contribute.

    Submitted contributions are subjected to an established review procedure, which will determine if they meet the required standards, are in line with the goals and mission of the Getty Vocabulary Program and the Getty Research Institute, and are in a format compatible with processing the data. If accepted, contributions will be prioritized, processed, and published according to internal priorities and timetables.

  • As time permits, the Vocabulary Program reviews individual records from contributors before they are released in the AAT. All contributions are checked, but with less supervision required for trained, established contributors.

  • Vocabulary Program (VP) editors follow strict rules when adding new records to the AAT. VP editors edit the contributors' records to comply with VP policy and practice; however, given the large number of records in the AAT, editors do not have the time or resources to edit every record. An editorial goal is to have uniform and homogeneous records throughout the AAT, but employing flexible standards for contributors' data means that the AAT database as a whole is not entirely consistent or totally uniform.

  • The VP collects new issues that arise during the course of accepting contributions and editing the AAT. The resolutions of these issues are periodically transferred to an updated version of the manual.









Does contributors' data follow editorial rules?

  • The Vocabulary Program communicates with and trains potential contributors, to assure that 1) the incoming data will be within the scope of the AAT, and 2) the incoming data will be in appropriate format and generally consistent with the AAT standards.

  • Every effort is made to ensure that the AAT data is consistent. However, given that the AAT may be compiled from various contributors' automated records, it is necessary to allow "flexible standards" in order to accept contributions. Compliance with the critical standards regarding technical rules, structure, content, and editorial guidelines are required; however, certain other content and editorial guidelines are considered non-critical and are therefore not strictly enforced for some contributors.









Releasing the data









Web searching

  • On the Getty Web site: Users may search for individual terms and names, qualified by other fields, in the Getty Vocabularies online. Using these search tools, catalogers may copy-and-paste Vocabulary terms and IDs as part of their daily workflow. Researchers use the search to locate rich information about the Vocabulary concepts. In the results displays, for each concept the data fields are presented in a logical full-record display for end users, as well as in hierarchical views. The online search data is refreshed monthly.









Data files

  • Data files: Releases of AAT, TGN, and ULAN include Linked Open Data (LOD) (JSON, RDF, N3/Turtle, N-Triples for GVP and Linked.Art), XML, and Relational Tables. AAT, TGN, ULAN, CONA, and IA are available through APIs. These files are used by developers or incorporated in various tools by vendors or others. These releases may be transformed to comply with other formats, such as the MARC format used for ULAN and TGN in the Virtual International Authority File (VIAF®). Some of these releases contain simplified versions of the data, while others contain the full, rich data sets, providing versions to meet the requirements of various developer communities. The data file releases are refreshed periodically throughout the year.
  • Incorporated in vendors' systems: The Getty Vocabularies are used in various collection management systems or other information systems. Examples are in TMS (The Museum System) by Gallery Systems and in the Refine cloud app in Alma by Ex Libris.
  • Data Dictionaries: The data dictionaries and other documentation describe the release formats and content of the Getty vocabularies. They do not give step-by-step instructions on how to construct a database or interface based on the data files; analysis and programming resources will be required to implement the vocabularies. The Getty does not provide technical support. View the data dictionaries and data from the Download Center.









Sample Displays of AAT Data














Vocabulary Coordination System (VCS)

  • VCS is the editorial system that was custom-designed to house and edit the Getty Vocabularies. AAT, TGN, and ULAN are each stored in a separate iteration of VCS; CONA and IA share an interation of VCS. References to "the system" refer to VCS. References made to "fields" refer to data elements in VCS. References to a "record" or "subject record" refer to an intellectual record comprising all the data linked to a given Subject ID in the data structure.










  • VCS uses a relational database; the database models for each of the five Vocabularies are identical in the core fields, differing only where additional tables or controlled values are necessary. See the Data Dictionary for further information.









Merged Records

  • The AAT is compiled from terms and other information that has been collected by the Getty and other institutions. When multiple contributors have submitted information about the same concept, all the terms and information about the concept are merged into a single record ("merge" is a function of the VCS editorial system).









Operating VCS

  • The chapters in this manual contain definitions of the fields, suggested values, sources where the values may be found, and rules for entering the data where relevant. The fields are presented roughly in the order in which they are found in VCS.

  • While there is some mention of the functionality of VCS in this manual, detailed instructions for the system are not included here. Instructions regarding how to use VCS are provided during training.


Updated 2 February 2023
Document is subject to frequent revisions


Back to Top

The J. Paul Getty Trust
The J. Paul Getty Trust
© J. Paul Getty Trust | Privacy Policy | Terms of Use