To browse other articles on a range of HSL topics, see the A-Z index.
Authority – as applied to library and information science – refers two things: to sources of information (reference books and monographs) that are deemed to be trustworthy and authored (also authorized) by experts in the field. For many library and information professionals, authority is assessed by asking questions such as who are the "authors" of a given work and what are their credentials?. Part of that assessment includes their prominence in their chosen fields. Are they considered experts and what have they accomplished? What kinds of scholarship are they engaged in and what have they produced during their careers? In education, librarians learn to locate (and trust) "authoritative" sources and rank them against those that are not. If recent information is important, check the authors' own sources in bibliographies (as well as the reputation of the authors' of those sources) for recency. Often librarians define and come to understand the concept of authority as an intuitive thing and often they know authority when they see it, and look for it over other types of information.
Health librarians are given the responsibility to acquire authoritative resources and to provide access to the best sources of information for users. Further, they have ample opportunity to exercise these skills of discernment, and share their knowledge with their users. Due to the proliferation of websites and digital content, determining authority has never been more important. In theory, any piece of writing, any text, data or information can be assessed as to its authority. Moreover, any information may be amenable to authority controls.
In the world of authority control, it is a bit of an alphabet soup of acronyms. ORCID (Open Researcher and Contributor ID) which is a system to uniquely identify scientific and other academic authors, ISNI (International Standard Name Identifier) which identifies the public identities of contributors to media content such as books, television programs, and newspaper articles, and VIAF (Virtual International Authority File) a system that combines multiple name authority files into a single authority service, hosted by OCLC, all have their place when discussing identifiers. Identity issues and questions about identity for authors, researcher, and other content creators, persist around these identifiers especially when we talk about traditional authority control vs. self-asserted authority. Librarians should understand the implications and differences between ORCID, ISNI and VIAF, the proper use of each, and some of the benefits that come with using authority files and making that information available on the Web.
What happens when sources of information indicate strong authority but still need to be evaluated: to have their authority questioned? Do we know for certain if a website or reference source is authoritative? Even though librarians should not be expected to verify everything, they must engage in a process to determine authority whenever possible. Further, when might it be appropriate to question authority? Recently, I read a news item reporting on a clinical trial. The study seemed to be valid, according to experts, yet it was published by a non-indexed journal viewed as "second or third tier". This is a good example of questioning the authority of something. Critical thinking and appraisal skills are vitally important literacies for the health librarian.
When teaching how to evaluate information, librarians often rely on the notion of authority (and may invoke the idea of evidence-based information, but we leave that for another day). One of the challenges of providing user services in librarianship is to stay open to information that does not conform to our personal views, and to our own views on authority. The message is that we need to be open to new authors and new ideas but not to trust everything we find online. Secondly, the deeper point is that no matter how entrenched our beliefs get, we must be vigilant about being led astray. Will we use checklists to evaluate sources? Should we encourage trust after an evaluation has been done? Since peer-reviewed sources will have their own errors and biases, and occasionally are subject to fraud, we must be vigilant. In our desire to emphasize authority, we must nonetheless be careful not to put too much trust in it. Even when all components lead to the conclusion a work is valid, it may contradict secondary sources. You may end up consulting an expert in that case. One of the ways we can partner with subject experts is to build large and diverse library collections and to encourage students to use more than one or two sources -- and to check and or corroborate sources against each other. This is very simple advice yet I seldom see it recommended or used. It’s a tricky balancing act, but in our need for users to “consult authoritative sources” it may still be necessary to question what you read.
Authority & the web
Authority control is the process of establishing preferred forms of bibliographic elements which, when used consistently, uniquely and unambiguously help to decipher between variant names, titles, subjects and other bibliographic elements. Authority control in library catalogues and databases helps librarians enables to disambiguate items with similar or identical terms. Although theoretically any piece of information on a given book is amenable to authority control, catalogers typically focus on authors and titles. Subject headings fulfill a function similar to authority records, although they are usually considered separately. In databases where authority control is lacking, such as what is offered by Internet searching and web search engines, there is a concomitant drop in search relevancy and completeness. This reduces the experience to browsing. In order to accomplish the goal of helping users search more effectively, authority control must include: dealing with spelling and tagging problems, adding cross-references and keeping up with changes in terminologies and natural language variants. Commonly-known terms in English (indeed, any language) are often different from authorized terms such as those in the LCSH (Library of Congress Subject Headings). The ability of online systems to return relevant results based on the authority records of terms may be critical to search results. This notion is not applied with any consistency on the web.
Authority in the catalogue
Authority in the catalogue is a concept that aims to ensure authority control (or uniformity) within a cataloguing system. In this context, authority encompasses several processes that ensure the accuracy of data points or access points and consistency and authoritative forms for personal and corporate names, uniform titles, authors, subjects and other bibliographic information. Authority control is typically carried out by technical experts or cataloguers in a library's technical department, and will result in production of authority records (or authorized files) that create a solid foundation for research accuracy and consistency. The painstaking work of authorizing names and titles may be done at a high level such as at the national library level, see WorldCat.
Talmacs defines authority work as the “intellectual decision making that determines variant forms” of headings and titles (1998, 129). Users of the library catalogue have to expect, and reasonably so, that when they look under names of a novelist, writer or researcher, that all variations of the name will be subsumed under a single authoritative form. Don't look under "Clemens, Samuel Langhorne" but see Twain, Mark. Put simply, the reason librarians emphasize authority is due to the need to create accurate listings of the library's collections, and machine-readable records represent an important part of that equation. In doing this work, librarians perform a critical function for authoritative forms of titles translated from one language to another, and when the first language of a title is based on a non-Roman script such as Arabic. From the Arabic, there are at least six or seven spellings for Gadhafi, for example. One authoritative form must be chosen, and typically it will be determined by consulting three or more reference sources to see the name that is most-commonly used.
Burger (1985) identifies three primary parts of a catalogue record for items in a library:
These three parts of a catalogue record are implicit in the functionality of most catalogue systems. Talmacs describes the steps necessary to achieve access control, and identifies a direct correlation between the products of authority work and aspects of catalogue records. Authority records consist of the authoritative form of an access point (heading); alternative forms of the access point (see and see also references); and the history of terms and their scope notes. According to Talmacs, an authority file "is an organized assemblage of authority records" (1998, 129). The processes performed under the umbrella of authority work are traditional librarianship practices, and date back to the work of Antonio Panizzi. The creation of access points, references, and notes in the bibliographic record has been a common practice in cataloguing systems for many decades. Authority files are ultimately created and linked to bibliographic files which form an authority system. This authoritative representation of the library's holdings is why the catalogue has primacy over other catalogues or databases. By maintaining and reevaluating authority files as items are added to the catalogue, the authority system – and access points – will be continually improved and enhanced.
In the absence of authority, there is a corresponding drop in finding relevant works and precision in searching. This is not only frustrating but a considerable waste of time for busy scholars and researchers who rely on getting the information in hand as soon as possible.
Charles Cutter and his rules
See also Charles A. Cutter
Authority control is directed towards access points and ensuring the consistency creation of headings and titles. In Cutter's Rules for a Printed Dictionary Catalog published in 1876, Cutter states that controlling the variant forms of names was one of the ways to meet the objectives (he called them ‘objects’) of the catalogue (Tillett, 2004, 24). The three objectives are commonly referred to the finding objective, collocation and the identifying objective. Authority control, as it is currently practiced around the world, tends to focus on enhancing these three cataloguing functions which harken back to Cutter’s objectives. Finding, collocating, and identifying intellectual objects is achieved efficiently by maintaining a consistency in the verbal forms used to represent these access points in a catalogue. Futher, there should be a process in place that reveals the relationships among headings for names, works and subjects (Taylor 2006, 283). Lubetzky in his Principles of cataloging from 1969 says that uniform headings and titles are especially successful in accomplishing collocation as they "[enable] an inquiry under any variant form to retrieve the works of an author under any of the names he/she used – even a citation from a bibliography" (cited in Tillett, 2004, 25). The complicated system of references outlined in AACR2R in conjunction with rules for headings and unified titles theoretically create the ideal collocation functions in the catalogue.
Strangely, perhaps due to the costs of authority control, some indexes such as PubMed and the Web of Science suffer from poor name authority, making it difficult to find all papers by a given author. The searcher, whether a librarian or scholar, will therefore need to employ techniques to ensure that no citations are missed during the information retrieval process.