Note: accents do not appear within this article.
Toponyms present little difficulty in common parlance; no one thinks of the city of Chicago when hearing the words "New York," "Nueva York," "Manhattan," or "the Big Apple." When large sets of them need to be organized the task is not so easy, however. Over time certain procedures have become traditional in the English-speaking world for dealing with place names. These traditions are the basis for the Anglo-American Cataloguing Rules (AACR2r), whose three primary principles are summarized below.
One viewpoint
AACR2r adopts a single viewpoint, that of English speakers. "Florence" is preferred spelling to "Firenze" and "Louvain" to "Leuven." Sources that aspire to an international viewpoint, such as RILM Abstracts, cannot rely on toponyms adopted by AACR2r users, such as the Library of Congress.
The name is primary
Because AACR2r applies to the organization of names, words
take precedence over historical and geographical matters. Complicated
geographical issues are avoided wherever possible, such as the
complexities of toponyms like "Hong Kong" and "New York."
Everything in the present
The current name for a location is preferred for that location throughout history. "Babylonia" is placed under "Iraq" and "Dahomey" under "Benin."
Relating toponyms to one another
Though AACR2r is as good as any other, it is certainly not perfect; articles by Wellisch and Rodriguez discuss some of the problems. Grouping relevant information is one such issue; political jurisdictions that change boundaries over time seldom have all relevant information in one place. A good example of this is the area that is now eastern South Dakota. At times this area was part of the District of Louisiana (1803-05), the Territory of Louisiana (1805-12), the Missouri Territory (1812- 20), the Michigan Territory (1834-36), the Wisconsin Territory (1836-38), the Iowa Territory (1838-49), the Minnesota Territory (1849-58), and the Dakota Territory (1861-89). Studies of all periods confined to modern eastern South Dakota are catalogued by AACR2r rules under the current toponym "South Dakota." Therefore, a person researching any of the jurisdictions of which eastern South Dakota was a part must look under the name of this state (and several others).
Incidentally, policies of the Canadian National Library circumvent this kind of problem in selected cases by being logically inconsistent; materials on several Canadian historic jurisdictions, such as "Acadia," are placed under that toponym rather than under the modern provinces of New Brunswick and Nova Scotia.
Research resources
Webster's and Westermann remain good places to begin research on a toponym. The geographic information they provide can't always be found on the internet. There are a number of free, public webpages that can help with place-name problems. The largest is supported by the United States Board on Geographic Names (BGN), whose pair of databases cover the world. Users need to be mindful that the BGN is "the interagency board established by public law to standardize geographic name spellings for use in U.S. Government publications." Obviously, political and cultural ramifications flow from this mandate. Virtually all of the several other countries that have public webpages devoted to their toponyms are English-speaking, perhaps reflecting the difficulties the internet has with accented letters. Obtaining comprehensive topographical information on the internet is far from a reality.
Toponyms: Some Practical Problems
After the war
The international RILM office ran into some thorny issues concerning Vietnamese toponyms appearing in music literature written after the change of governments in 1975. Toponyms referring to regions within Vietnam appeared that could reasonably be names for non-political historic areas, official province names for the governments before 1975, or official province names for the new government. RILM's indexing policy, that only current toponyms be used in the index, required distinguishing the three groups and changing old province names to new ones. The political atmosphere made conflation of old and new likely since a number of pro-democracy authors were publishing materials outside the country.
This kind of problem, far from unique, would hardly merit discussion had a map or gazetteer of the newly organized Vietnam been readily available. Unfortunately, such maps in the United States seemed to be completely unavailable in the 1980's; my research at the map division of the New York Public Library and contact with major map distributors turned up nothing. An additional concern was that some of the toponyms as submitted on RILM forms seemed (to my eye) to not conform to the double-accent system of the Vietnamese language. Clearly a detailed, official map or gazetteer of post-1975 Vietnam was needed.
This problem was eventually resolved through contacting the office of the Vietnamese delegation to the United Nations. Officials there were very helpful, promptly supplying a detailed map with current, official province names. Comparing old and new maps of Vietnam allowed for the old province names to be identified and ambiguities to be sorted out.
The Whole and the Parts
Delineating metropolitan areas and their components usually pose difficulties; such a case occurs in my index for Cultivating Music in America: Women Patrons and Activists since 1860, edited by Ralph Locke and Cyrilla Barr. A chapter in that book extensively discusses musical life in Brooklyn during the 1880's, a period before Brooklyn became a part of New York City. After proposing solutions to relevant parties, I decided to place information about Brooklyn under "B" rather than under a subheading for "New York." My reasons for contravening practices of the Library of Congress were practical. I felt that interested readers could find the heading with little difficulty and the table of contents showed "Brooklyn" in a chapter title. Had there been numerous references to musical life in twentieth- century Brooklyn, when it was part of the city of New York, I likely would have decided differently.
Who goes to an unfindable place?
Recently my work for ABC-Clio, the producers of databases such as Historical Abstracts and America: History and Life, led me to a pair of articles in the journal Etudes/Inuit/Studies concerning petroglyphs of the Dorset people. In the articles the site is described as "Qajartalik dans le Nunavik." The texts described an island and the conservation actions by a Quebec government agency. What and where is "Qajartalik?"
First, the term "Nunavik." It is listed in neither Webster's,Westermann, nor the GNS database. In the NRCan toponyms database of the Canadian government it appears as a "geographical area" of northernmost Quebec, with an area of some 400,000 square miles and a population of less than 10,000. NRCan considers "Nunavik" to be completely synonymous with "Kativik," but provides no explanatory information. In other sources, the first-rate series Paleo-Quebec prefers "Kativik" to "Nunavik" (1994), while Picard (1994) orders the terms the other way. Before my research began I had assumed that anglophone-francophone issues would crop up (many geographical features of the area, originally of English origin, have been renamed, such as the "Payne River" being renamed the "Arnaud River"). Disagreements over pairs of toponyms apparently both of Inuit origin was something I hadn't expected.
The names of the larger government jurisdictions subsuming this area also vary; the 1991 Canadian Federal census considers the area as part of the "Territoire Nordique" region, while Quebec province uses the term "Nord-du-Quebec." Nunavik's representative to the Canadian Parliament is from the "Abitibi" district, while the provincial assembly member is from the "Ungava" district.
At any rate, the island of "Qajartalik" must be off the coast of northern Quebec, right? This reasonable assumption brings to light another issue, for all islands in the region belong by definition to the Northwest Territories, not to Quebec. About the only way that "Qajartalik" can be an island within Quebec's jurisdiction within Nunavik is that it is very, very small.
Now, to "Qajartalik" itself. The term is not in any international geographic source known to me. It is not in the NRCan database. It is not found in two excellent, extremely detailed publications, Repertoire Geographique du Quebec (1969) and Repertoire toponymique du Quebec (1979). Two close cognates appear in these sources, Point "Qajariartalik" and "Qajurtuvik" Cove. Perhaps the petroglyphs are on a very small island close to one of these areas and that language issues account for the variant spellings. Perhaps not. An archeological webpage confirmed the spelling of "Qajartalik" but provided nothing new concerning its location. My close examination of small-scale topographic maps, most of which showed no human settlement, likewise produced nothing more. Using the best government geographic resources of a province that takes territory issues very seriously, I was unable to find this unique cultural and historical site in my own country; the same, unfortunately, cannot be said for those who defaced the petroglyphs in the last several decades!