The Australian National Placenames survey.

The aim of the Australian National Placenames Survey (ANPS) is to prepare a national register of geographical names. The National Register will be a supplement to the Australian National Gazetteer, and will be a database constructed on recognised principles for the research and documentation of history, geography, linguistics, cultural knowledge, surveying and mapping. It will record all known Australian names, documenting their pronunciation, generic class, status (gazetted, obsolete, nongazetted, sensitive, disputed etc.), origin, meaning, history, cultural significance (of both name and site), and map reference and location. The database will enable the production of a series of place name dictionaries, both local and national, over the next few decades.

The project will have important beneficial effects not only on the cultural aspect of geographical names but also on the technical aspects of government function (such as efficiency in communications, transport and defence). The project involves support from government agencies, academic institutions and individuals. A national structure is being put into place to implement the project, ensuring effective representation of interested parties at national, state and local levels.

THE SCOPE OF THE ANPS

The task of the Australian National Placenames Survey is to investigate the history, meaning, and motivation for use of each name ever current for a geographic feature or inhabited locality in Australia, and to make public the results of these investigations. The cultural aspects of placename study have never formed an area of systematic research in Australia, and the Survey aims to remedy this deficiency.

The roots of the project extend back to the earliest days of the Australian Academy of the Humanities, under the patronage of its Foundation President, Sir Keith Hancock. An earlier pilot phase took place at the University of New England between 1971 and 1974, when Dr John Atchison acted as research fellow. However, the political and economic climate toward the end of that period meant that the project was then rested, and the intention lay fallow for the subsequent quarter century.

Ironically, it was in 1974 that K.S. Inglis commented (in the introduction to The Australian Colonists) that with the imminent publication of the final volume of the initial sequence of the Australian Dictionary of Biography, two further definitive reference works were needed in order for Australians to achieve a full understanding of their cultural heritage: a dictionary of Australian English, and a survey of placenames. With the publication of the first edition of the Macquarie Dictionary in 1981 and the Australian National Dictionary in 1988, only the last remains to be fulfilled.

In the meantime the technical aspects of toponymy received a great boost with the formation in 1984 of the Committee for Geographical Names in Australia (since 1998, with the joining of New Zealand, the Committee for Geographical Names in Australasia). The related cultural aspects are addressed by the ANPS, and close collaboration with state/territory nomenclature bodies through this forum is crucial: in November 1998 Macquarie University, as the host institution of the Survey, was admitted to full membership of the Committee. The CGNA has recently been working to develop a National Placenames Data Model, which will form one element of the Australian Spatial Data Directory being created by the Australia New Zealand Land Information Council.

In 1996 the Australian Academy of the Humanities was successful in obtaining funding under the Australian Research Council's Learned Academies Program to relaunch the scholarly study of cultural aspects of Australian toponymy. The initial pilot phase, the National Placenames Project (1998-99), involved the employment of a full-time research fellow, Flavia Hodges, with extensive experience in the fields of onomastics, lexicography, and publishing management, to develop the methodologies and management structures needed to underpin the Survey as a whole.

An expansion of funding for 2000 allowed the launch of the Australian National Placename Survey proper and the employment also of research associate Susan Poetsch, with experience in teaching and linguistics, and IT officer Robert Iverach, with expertise in database development and geographic information systems. Robert Iverach left the Survey at the beginning of 2001 to complete his PhD in Geographic Information Systems at Curtin University of Technology, Perth, but has continued to advise on database development.

Since the beginning of 2002 the work of the ANPS has been carried out under the umbrella of the Asia-Pacific Institute for Toponymy - it is the principal activity of the Institute's historical and cultural toponymy section, and funding for the core directorate is assured until 2006 by a grant from the Vice-Chancellor's Millennium Innovations Fund of Macquarie University.

The ANPS collaborates closely with the state and territory nomenclature authorities who are responsible for the technical aspects of toponymy and placenames standardisation, to which we seek to add cultural information about the history, origin and meaning of placenames. This practical research work will be carried out by a network of volunteer Research Friends, coordinated by state/territory ANPS committees representing a wide range of interests and expertise, such as history, geography, Australian languages and archaeology. The results will be entered in a uniform national database available for browsing and searching on the Internet.

The ANPS has arranged a series of day conferences on placenames of Indigenous origin, from which selected papers have been published in a monograph (Hercus et al., 2002). We also undertake theoretical and practical training in cultural toponymic research through university seminars, continuing education courses and self access study programs, and supervise candidates undertaking Masters and PhD research in the area.

PROGRESS

Progress on the Survey is encouraging, though slow. The main hurdle is completion of the project database, which will have to cope with as many as six million placenames and their related items of information. Design of the data model is complete, but implementation (most probably as an Oracle DBMS) is going to be time-consuming and costly. In the meantime, data is entered via MS Excel files. Public access to the database through our Web interface is intended, but will be some considerable time in the future.

Data is currently being received from dedicated researchers around the country. Informants are either Research Friends in the community who typically belong to Historical Societies or Progress Associations, or academics, typically research students working on Masters theses or PhDs.

Funding issues are, unsurprisingly, at the forefront of our minds. Major surveys which will not be completed within decades are not favoured applicants to Australia's research grants committees. The availability of academic supervision for research students is also an issue: there are no departments of toponymy in Australia, and no academics whose full-time research interest is placenames.

PLACENAMES AND NSW MAPPING

Our field surveys in NSW begin with the data of the Geographic Names Register of NSW, and continue with the 1:25 000 topographic maps of NSW. The fieldwork, then, is a process of 3-way checking: attempting to match the GNR data with the topo map features, with on-ground observation.

Rural areas and older mapping

The area unit we have chosen for our field research in NSW is the parish, and our on-ground observations are carried out with the relevant 25k map already marked up with the parish boundaries. Our early work was largely rural, and was carried out using 1st and 2nd edition maps. The field surveys of Singleton and Cessnock, for instance, were based on mapping that had field revision dates of 1975 and 1985 respectively.

In those surveys, we were not surprised that there were discrepancies between the mapping and on-ground observations. Most of these, understandably, had to do with housing developments that simply did not exist at the date of the last field revision. A few, however, were oddities in the mapping:

* Loder Creek near Singleton changed its name in 1976 from Loders Creek. The prophetic Singleton map of 1975 gives the yet-to-be-assigned name; but the Bulga map of the same revision date does not. Indeed, the road sign at the Putty Road bridge even now calls it Loders Creek.

* Skinners Hill in Cessnock Parish is absent from the map, for no obvious reason. At this stage our assumption is that the GNR is correct, although we were unable to confirm the name from local evidence.

* Homesteads are frequently marked on the maps of these rural areas. One is encouraged to assume that they are of considerable significance, both in their historical value and their current imprint on the landscape. In practice though (at least in Singleton and Cessnock) they usually are no more than typical suburban homes on the outskirts of the towns; and we've had to work hard to convince ourselves that their presence in 25k mapping was a good reason to include them in our Survey.

The bulk of discrepancies, however, were with the GNR data itself. The entry of data into the GNR was not 100% accurate by any means, and we've been glad to return some added value by sending occasional lists of corrections to our colleagues at the Geographical Names Board at Bathurst. In some cases the conflict with reality arose, we suspect, from overenthusiastic local government responses to the Boundaries Program: proposed new developments were named without consultation with the local community and the names were submitted to the GNB, while the reality of rural life has not so far managed to catch up to the shire president's vision.

Suburban areas and 3rd edition maps

In 2002, with the help of some enthusiastic undergraduate students, we've checked several parishes in the Sydney suburban area. The district is now covered by the newly-released 3rd editions of the 25k scale maps. What is the rate of discrepancies now?

Again, as we might expect, the GNR itself is the source of some problems. Cochrane and Dunheved Railway Stations, we are assured by the Register, are stations on the Penrith and Blacktown lines respectively. Since in fact they are discontinued stations on the unused Ropes Creek Railway, it's not surprising that our student researchers could not locate them on the 3rd edition maps. (They do not appear on the Ropes Creek line either, but it would perhaps be unfair to criticise the cartographers for omitting discontinued stations.)

There are some problems, however, with the mapping. Most of these have to do with the location of localities (suburbs, neighbourhoods) rather than with other geographical features.

* The Prospect map includes a tag for North Rocks in an area known as Northmead, and a tag for Castle Hill within Baulkham Hills. Similarly, the Parramatta River map tags Mobbs Hill north of its true location, in an area known to all as Carlingford. What seems to be happening cartographically is that location tags are being used as direction tags, since they all occur near the borders of the map. In other words, "North Rocks isn't really here, but if you look on the adjacent map you'll find it ..."

* The GNR records a suburb named Holroyd in Prospect parish, identified as "a neighbourhood situated SE of Greystanes and NNW of Woodpark". Although Holroyd High School appears there (and various street directories identify the locality), no suburb name is in evidence on the map. The GNR data for latitude and longitude, in fact, would lead us to believe that the suburb is 50 or so kilometres away, somewhere near Springwood in the Blue Mountains; this is another common problem with GNR data entry.

The remaining few discrepancies relate to natural features, where the GNR data or field information conflict with the mapping (all examples from Prospect parish).

* Girraween Creek is locally labelled (on the Great Western Highway) Greystanes Creek, in conflict with both the GNR and the map.

* Similarly, the authority responsible for signage on the Highway prefers Pendle Hill Creek to the GNR's Pendle Creek.

* A hill named Mount Dorothy (its location confirmed by local street names Dorothy St, Mount St) is not mapped. A nearby commemorative plaque for Constitution Hill and a nearby Constitution Road suggest an alternative name and will encourage us to enquire further.

* Lalor Creek, signed on the Prospect Highway, appears to be a branch of Blacktown Creek, but is absent from both the GNR and the 25k map.

* Seven Hills Trig Station is typical of a fairly frequent discrepancy: the GNR lists it as being within the Prospect parish, but the map disagrees. The coordinate listing appears to be correct--the former name Seven Hills Reservoir Trig Station matches the existence of a reservoir at the given coordinates. The error is simply one of the parish attribution when the GNR data was being entered.

SUMMARY

The publication of 3rd edition mapping is a great benefit to our fieldwork. The map revision has considerably reduced the number of discrepancies, at least for most types of features. And we are not particularly disheartened by the instances that remain, since they indicate fruitful areas for more local investigation and in many cases may enable us to return a favour to our colleagues in Bathurst.

REFERENCES

Hercus, L., Hodges, F. and Simpson, J. eds. 2002. The land is a map. Canberra: Pandanus Press

Inglis, K.S. 1974. The Australian colonists: an exploration of social history 1788-1870. Carlton: Melbourne University Press.

David Blair, Director, Australian National Placenames Survey. [e-mail]