Dictionary of Southern African Place Names

By Peter E Raper, Lucie A Moller and Theodorus L du Plessis

The Dictionary of Southern African Place Names - now in its 4th edition - helps you sort your Komkhulu from your Kommetjie with the most comprehensive glossary of Southern African towns, villages, railway stations, mountains, rivers and beaches. The 9 000 short entries incorporate data from sources dating as far back as 1486, encapsulating the linguistic and cultural heritage of all the peoples of the subcontinent, past and present. In this highly readable book the expert authors take you on a fascinating journey of the highways and byways of Southern Africa. Whether you are a motorist, an adventurer or merely an armchair traveller, this book has a multitude of facts and details that will fascinate you. This is much more than a reference book - it gives an insight into what shapes a place and its people through our heroes, events, beliefs, values, fears and aspirations.

• Language: English
• Publisher: Jonathan Ball
• Release date: Dec 8, 2014
• ISBN: 9781868425501

Book preview

PREFACE

The publication of this Dictionary of Southern African Place Names was encouraged by the enthusiasm with which the first three editions were received, and necessitated by the increase in the number of provinces from four to nine, and the implementation of the South African Geographical Names Council Act, which has resulted in numerous new names being standardised.

The Dictionary has been expanded, with the addition of many new entries. In this regard our appreciation is expressed to Dr BA Meiring, Mr Trueman Khubekha and Mrs Thulile Mthembu, who made the data of the South African Geographical Names Council available for inclusion in the Dictionary. Permission of the editors of the journals Acta Academica, Language Matters, Names, Nomina Africana and Onoma to use material published in these journals is acknowledged with gratitude.

Thanks are also due to Dr Peter Alcock, Mr TE Andrews, Mrs E Aphane, Revd JJE Barrow, Mr Hendrik Bohnen, Mrs Pauline Bohnen, Mr Hans Bornman, Prof TJR Botha, Dr KFR Budack, Dr M Burden, Dr Alfred W Burton, Mr D Clarke, Dr CE Claus, Mr J-R Bernhardt, Mr H Dys, Mrs CF de Jager, Mr JF Dick, Dr Franco Frescura, Mr AJ Greeff, Dr Joh Groenewald, Mr MJ Hall, Mr CJ Henry, Mr FJ Horak, Mr NSH Hughes, Dr JJ Kloppers, Mr D Kopke, Prof Adrian Koopman, Ms Chrismi Kotze, Mr D Krause, Ms Ruth Labuschagne, Mrs E le Roux, Mr JC Loock, Mr R Martin, Dr DL Milne, Mr P le S Milstein, Dr LA Möller, Mr PD Morant, Prof PT Mtuze, Mrs PN Mtintsilana, Mrs JC Mulder, Mr L Nel, Mrs A Nicolaides, Professor GS Nienaber, Professor PJ Nienaber, Mr J Orford, Brig-Gen Dr JH Picard, Dr G Quinting, Dr Richard R Randall, Dr IR Raper, Mr WV Read, Dr CJ Skead, Dr D Sleigh, Miss AH Smith, Dr PD Toens, Mrs DF van Breda, Mrs P van der Westhuizen, Mr GJ van Eck, Mrs M van Niekerk, Mr Theuns van Rensburg, Prof EB van Wyk, Mr WP Venter, Mrs Jo Walker, Mr WC Watson, Dr A Wessels, Dr TCW Wessels, Mrs M Wright and all others who supplied information and documentation. Improvements and amendments suggested by readers and colleagues have also been incorporated with gratitude. The authors express their sincere thanks to the editor, Frances Perryer, and proofreader, Valda Strauss.

Cognisance has been taken of resolutions of the United Nations, and each entry complies with the minimum requirements recommended. As this is a dictionary of Southern African place names, names of major features in neighbouring countries have also been included, specifically in Botswana, Lesotho, Namibia, and Swaziland, as well as names of features on Marion Island. However, it has not been the intention to cover all southern African countries. There has been no attempt at comprehensiveness, and in cases where essential information could not be provided, the name has been omitted.

The authorship of the Dictionary has been extended to include Dr Lucie A Möller and Professor L Theodorus du Plessis, and is a project of the Unit for Language Facilitation in association with the Department of Linguistics and Language Practice of the University of the Free State, and will be a continuing project of that University. I have pleasure in expressing my sincere gratitude to Professor du Plessis for this important initiative, which will ensure the continuation of one of the most comprehensive collections of the common heritage of the peoples of southern Africa.

Peter E Raper

Pretoria, October 2014

ABBREVIATIONS

Bots Botswana

CSIR Council for Scientific and Industrial Research

EC Eastern Cape Province

ESKOM Electricity Supply Commission

FS Free State Province

GP Gauteng Province

HSRC Human Sciences Research Council

ICA International Cartographic Association

IGU International Geographical Union

km kilometre(s)

KZN KwaZulu-Natal Province

Les Lesotho

Lim Limpopo Province

m metre(s)

Mar Marion Island

Moz Mozambique

MP Mpumalanga Province

Nam Namibia

NASA National Aeronautics and Space Administration (USA)

NC Northern Cape Province

nd not dated

NPNC National Place Names Committee

NW North-West Province

RGN Raad vir Geesteswetenskaplike Navorsing

SAGNC South African Geographical Names Council

SATS South African Transport Services

Swa Swaziland

UNGEGN United Nations Group of Experts on Geographical Names

USBGN United States Board on Geographic Names

WC Western Cape Province

INTRODUCTION

Background

The place names of a people constitute a very important part of the language of that people. They sometimes preserve words and elements no longer current in that language and, as in the case of the Bushmen (San) and Khoikhoi in South Africa, constitute a major part of the surviving vocabulary of that people. Although place names generally originate in a specific language, they are used by speakers of other languages as well, and become part of the vocabulary of the receiving or adopting language, albeit sometimes in a form adapted to the phonological and orthographic systems of that language. The place names of a country thus constitute a commonly shared vocabulary of all the people of that country and are a national asset. Clearly this linguistic heritage is priceless and should be cherished and preserved.

Place names also constitute an essential part of the cultural heritage of a nation. They reflect aspects of history, of agricultural, domestic, economic, humanitarian, industrial, military, political, scientific, sporting and other activities; they honour the heroes and leaders of the people, and commemorate important events; they reflect the beliefs, values, fears and aspirations of the people; and they embody words for and references to the flora, fauna, climatology, geology, hydrography, topography and every other aspect of the country in which the people live.

The oldest names in southern Africa are those of the San and Khoikhoi, and those of the indigenous Bantu peoples. From the second half of the 15th century, Portuguese, Dutch, English, French, German and other names were given. The evolution of Afrikaans further enriched the geographical nomenclature. Settlement, colonisation and immigration, particularly as a result of the exploitation of metal and mineral resources, and of urban and industrial development, added names from Latin, Greek, Hebrew, Italian, Indian and other languages.

Cultural and biological interaction between these different peoples resulted in the widespread transfer of cultural traits from one society to the other (Elphick 1985: 23). This interaction involved language contact and language shift, affecting also the toponymic landscape of the region. Some existing names were replaced and new names were given; some names were translated (wholly or in part); hybrid forms, with part of the name in one language and another part in a different language, came into being. Moreover, some individual places came to bear more than one name, bestowed by different language groups. It also happened that many names were spelt in different ways by different people and at different times. Many existing names were retained, but adapted to the language systems of the newcomers. As Pettman (1931: 10) points out:

Subsequent occupants of the territory have frequently taken over the place names from those they dispossessed, but in so doing – their meaning being unknown to them – they have accommodated them in sound and form to words that were known to them; others have been translated into the language of the newcomers; others, like worn coins, have had their distinctive features so nearly effaced by phonetic abrasion that only the most patient and careful examination will serve to ascertain what those distinctive features were, while some appear to be beyond recovery.

International standardisation

Due to the problematic nature of place names (also called geographical names) the United Nations found it necessary to establish a standing expert group to advise it in this regard. This group, the United Nations Group of Experts on Geographical Names (UNGEGN), is one of only seven standing expert bodies advising the UN, which indicates that geographical names are among the seven most problematic areas facing mankind.

Efforts towards the international standardisation of geographical names date from as early as 1820, when attempts were made to develop a script or alphabet which could be used world-wide to write place names. Since then, various national and international organisations have given attention to the problem of international standardisation, inter alia the International Committee of Onomastic Sciences, the International Civil Aviation Association, the Pan American Institute of Geography and History, and similar organisations. In 1953 the Economic and Social Council of the United Nations requested the Secretary-General to draft a framework of a programme aimed at maximum uniformity of writing geographical names, and to set up a small group of consultants to deal with the matter.

The United Nations Group of Experts on Geographical Names (UNGEGN) was established in pursuance of Economic and Social Council (ECOSOC) resolutions 715 A (XXVII) of 23 April 1959 and 1314 (LXIV) of 31 May 1968, to further the standardisation of geographical names at both the national and international levels. The UNGEGN comprises experts in the fields of cartography and linguistics designated by governments of states at the invitation of the Secretary-General of the United Nations. Experts participating in the sessions of the UNGEGN serve in their personal capacities as individuals of recognised competence and experience in their respective fields.

The UNGEGN has the task of providing for continuous coordination and liaison between countries to further the standardisation of geographical names. In order to accomplish this, sessions of the UNGEGN are held, normally every two years. In addition, United Nations Conferences on the Standardisation of Geographical Names are held every five years. At the sessions and conferences matters appertaining to the standardisation of geographical names are discussed, and problem areas entrusted to working groups. Resolutions are passed and recommendations adopted at the various UN conferences, and these serve as guidelines for national names authorities, state departments, agencies, cartographers, writers, journalists, lecturers, teachers, and other persons and organisations playing a role in the standardisation and dissemination of geographical names.

By international standardisation is meant ‘that activity aiming at the maximum possible uniformity in the form of every geographical name on the earth and of topographical names on other bodies of the solar system by means of national standardisation and/or international agreement, including the achievement of equivalences between writing systems’.

One of the fundamental guidelines for the international standardisation of geographical names is that it be based on national standardisation. With this in mind, the various UN conferences have stressed the need for the establishment of national names authorities. Guidelines have been formulated for the structure of such bodies, for the field collection and office treatment of names, and for the compilation of gazetteers. Countries everywhere in the world are urged to use the local standardised forms of names, i.e. the names of geographical entities as standardised by the various national names authorities.

In cases where so-called ‘exonyms’ (‘conventional names’ or ‘traditional names’) are used for international purposes, such as Vienna for Wien, Florence for Firenze, Cologne for Köln, and the like, the officially standardised form should also be indicated. The reduction of exonyms which have fallen out of use and become obsolete has consistently been recommended.

Other resolutions of the UN Conferences on the Standardisation of Geographical Names pertain to the romanisation of names from non-Roman languages such as Arabic, Chinese, Greek, Hebrew, Japanese, Korean, Thai, etc; the application of automated data processing techniques to the compilation of gazetteers; international co-operation in the standardisation of geographical names; maritime and undersea feature names; names of features beyond a single sovereignty; extraterrestrial feature names; geographical names from unwritten languages; and education and training in the treatment of geographical names.

National standardisation

The National Place Names Committee (NPNC)

As early as 1936 it was realised that a measure of standardisation of place names in South Africa was necessary, and a committee was appointed by the then Minister of the Interior to investigate the situation. At the recommendation of this committee, a Place Names Committee, later to become the National Place Names Committee (NPNC), was appointed in 1939. The principles and guidelines of the NPNC are given here, since they served as the basis for those of the South African Geographical Names Council (SAGNC) which superseded it.

Principles and guidelines of the National Place Names Committee

‘General suggestions

1.   The following names should be avoided:

(a)  Names already existing in the Republic of South Africa;

(b)  names of well-known places in other countries, and names of countries;

(c)  names having the same or almost the same spellings or having the same pronunciations as or very similar pronunciations to existing names: this would eliminate unnecessary confusion, e.g. in the postal, telegraph or telephone services;

(d)  names that are blasphemous, indecent, obscene, vulgar or unaesthetic;

(e)  names that are discriminatory or derogatory in regard to race, colour, religion, sex, political affiliation or other social factors;

(f)  names that are too long, clumsily compounded or impractical (more especially in view of the limited space available for such names on post office or municipal date-stamps, on railway tickets, on station name boards, etc.);

(g)  names that consist of no more than the name of a person and are therefore likely to cause confusion between place and person;

(h)  names that are too easily translated and may therefore lead to dual forms;

(i)  names that may be interpreted as advertising a specific firm or product.

2.   Existing names that are satisfactory and acceptable should not be changed without good reason.

3.   Where there is a clear connection between a proposed name and the specific place, e.g., where a motorbus halt is given the name of the farm on which it is situated, or where a post office is given the name of the suburb in which it is situated, the spelling of the new name should be the same as that of the place from which the name has been derived.

4.   New names should not be used on name-boards, in advertisements, etc, unless they have been approved by the Council.’

In pursuance of United Nations resolutions recommending the dissemination of standardised geographical names in gazetteers and other publications, the work of the NPNC led to the publication of Official Place Names in the Union and South-West Africa (1951), Official Place Names in the Republic of South Africa and in South-West Africa (1978), Official place names in the Republic of South Africa approved 1978-1988 (1991), as well as a Manual for the Giving of Place Names (1979), and a Concise Gazetteer of South Africa (1994). The spellings of names approved by the NPNC remain valid unless altered, revised, amended or replaced by the South African Geographical Names Council which superseded the NPNC in 1999.

The South African Geographical Names Council (SAGNC)

The SAGNC was established as the national geographical names authority of South Africa by the South African Geographical Names Council Act (Act No. 118 of 1998). The purpose of the Act is: ‘To establish a permanent advisory body known as the South African Geographical Names Council to advise the Minister responsible for Arts and Culture on the transformation and standardisation of geographical names in South Africa for official purposes; to determine its objects, functions and methods of work; and to provide for matters connected therewith.’

New place names, and proposals for changes or revisions of names, are submitted to the SAGNC and to the Provincial Geographical Names Committees. On their recommendation names approved by the Minister of Arts and Culture are published in the Government Gazette.

Principles and guidelines of the SAGNC

According to the Handbook on Geographical Names of the South African Geographical Names Council, standardisation is based on:

1.   The current orthographic (spelling) rules of the languages from which the names are derived;

2.   The wishes of the local population, provided they are not in conflict with the principles of the SAGNC;

3.   The historical use of the name;

4.   Redress, where a name is changed on the basis of historical considerations;

5.   United Nations resolutions on the standardisation of geographical names;

6.   Any other relevant factors which the SAGNC may identify (Handbook 2002: 4).

Geographical names that should generally be avoided are the following:

1.   Names that have already been approved for other places in South Africa;

2.   Names of places in other countries, and names of countries;

3.   Names of which the spelling or pronunciation is so close to that of an existing name that confusion might result;

4.   Names that are blasphemous, indecent, offensive, vulgar, unaesthetic or embarrassing;

5.   Names that are discriminatory or derogatory as regards race, colour, creed, gender, political affiliation or other social factors;

6.   Names that are too long or clumsily compounded;

7.   Names consisting of a personal name only, without an additional generic element (such as ‘Park’);

8.   Names that may be regarded as an advertisement for a particular commercial product, service or firm.

Naming places after persons

1.   Names of living persons should generally be avoided;

2.   Geographical entities named after persons should be in accordance with the stature of the persons concerned (not, for instance, naming an obscure feature after a person of national importance);

3.   Written permission should, where possible, be obtained from the individual or the individual’s family or heirs before that person’s name is used (Handbook 2002: 5).

Conventions regarding South African place names

Given the above guidelines of the SAGNC it is important to take note of conventions that have developed over time in South Africa and that still largely hold in the absence of amended rules and prescriptions. These conventions are language-specific, as the overview below will indicate.

The conventions regarding the writing of South African place names that have developed over time and that are largely reflected in this dictionary are discussed for the different language traditions below.

Conventions regarding Bushman (San) place names

Special attention is given to Bushman (San) place names, since recent research has revealed a hitherto unsuspected substratum of Bushman influence on the place names of the subcontinent. ‘Until about 2 000 years ago these people and their ancestors were the only inhabitants of the southern portion of Africa, and virtually every corner of the subcontinent housed a record of intermittent but substantial occupation extending back thousands or millions of years’ (Parkington 2007: 77). These hunter-gatherers roamed the country in search of the abundant game, occupying every part of the land, and presumably naming every feature. The first place names in the region were thus presumably derived from Bushman languages.

Werner (1925: 117) states that the Bushmen were ‘the aborigines of South Africa’, and that they inhabited the entire sub-continent of Africa ‘till they were encroached on, first by the Hottentots, then by the Bantu, and finally by the Europeans’. In due course many tribes have disappeared, and others are in the process of becoming extinct. In 1925 there were still several tribes in the north-western part of the Transvaal and in the Kalahari Desert (Werner 1925: 117). Bleek (1929: 1) recorded that the Colonial Bushmen or /Xam were then almost extinct.

The term ‘Khoisan’

The term Khoisan is sometimes used to refer to the non-Bantu click languages of southern Africa, derived from the first element Khoi of the term Khoikhoi, and San, the alternative name for the Bushmen.

The subject of the relationship between these languages is still a matter of debate and outside the scope of the present publication. When specific languages are referred to, the language in question is used, eg /Xam, Auen, Kung, Hadza and so forth. When referring to the Bushman languages in general, the term Bushman is used, although the term San also occurs when the Bushman people are referred to.

The terms ‘Bushman’ and ‘San’

In June 1971 the Royal Society of Southern Africa and the South African Institute for Medical Research jointly sponsored an international symposium on ‘The Peoples of Southern Africa’. The participants, drawn from the disciplines of social, cultural and physical anthropology, linguistics, archaeology and genetics, recommended inter alia that the term Bushman be used for the languages spoken by the hunter-gatherers known as San (Jenkins and Tobias 1977: 51). The Editorial column of the South African Archaeological Bulletin (vol XXIV, no 129, June 1979, 1) stated that ‘Such a call for standardization is greatly welcome and we hope that authors will find the terms both useful and usable.’ Rudner (1979: 3) prefers to use ‘the historical terms’ Hottentot and Bushmen ‘rather than the Khoikhoi terms Khoikhoi and San.’

It is generally accepted that the English term Bushman and the Afrikaans term Boesman are translations of the Dutch name Bos(ch)jesman, derived from the Dutch word bos(ch) ‘bush, forest’, of which the diminutive form is boschje or bosje (Nienaber 1989: 207). There are various explanations for the name. Sparrman (1785 I: 197) states that these people were so called ‘from dwelling in woody or mountainous places’; Barrow (1801 I: 56) wrote that they were called Bushmen ‘because, lurking in the cover of the shrubbery, they are said to shoot their poisoned arrows’, while Campbell (1815: 314) believed they received the name ‘from their country … being almost destitute of trees, but much of it being covered with bushes; [and] from their method of assault, as they never attack man or beast openly, but from behind bushes’.

The ethnonym San is said to be derived from a Khoikhoi verb sā, ‘auflesen, aufraffen (vom Boden)’, ‘to gather, pick up (from the ground’) (Kroenlein 1889: 293), plus the third-person common plural ending -n, ‘people’. Vedder (1938: 78) interprets the name as ‘the gatherers’, referring to their collecting the food that grows wild in the veld, their main article of diet. However, Budack (1969: 211) is of the opinion that the explanation of the name San as meaning ‘gatherers’ is not viable, because of tonal distinctions between the verb sā and the noun sāb. Nienaber (1989: 834 35) agrees with Budack’s linguistic argument, and also rejects other explanations of the name, for example that it means ‘pariahs, servants, subjugated ones’; ‘aborigines or settlers proper’; ‘the original inhabitants’; ‘those who sit and enjoy their leisure’, and the like (Nienaber 1989: 831). He concludes: ‘Die eindresultaat is dat geen enkele van hierdie taalassosiatiewe verklarings wetenskaplik oortuigend verantwoord is nie. Dit is die huidige stand van sake’ (‘The end result is that not a single one of these language-associative explanations has been convincingly validated scientifically. That is the present state of affairs’) (Nienaber 1989: 835).

At one stage Bushman was apparently regarded as a derogatory name and the term San was preferred, but recently Bushman has become acceptable again, and San is said to be regarded as a derogatory Khoikhoi name for the Bushmen (Raper 2010). A perusal of the Internet indicates the different points of view. ‘According to Andries Steenkamp, chairman of the South African San Council, the word Bushmen is preferred, since it underlines his people’s status as first nation. By contrast, Alex Thoma, the adviser to the South African San Council, told me that a representative gathering of San in 1993 decided to use that term’ ([e-mail]). ‘The names San and Basarwa are sometimes used, but the people themselves dislike these names and prefer the name Bushmen’ (Kids.Net.Au-Encyclopedia>Kalahari Desert). ‘Although some consider the name Bushmen derogatory as well, it is less so than the term San’ (www.diversitysouthafrica.co.za/khoi.html). ‘The different tribes and clans started using the name Bushmen. But political correctness proclaimed this an insult and declared they should be referred to as San. Where Bushmen merely means they are people of the bush (or close to nature), San was a Khoi word meaning vagrants, people who have nothing. And because of this misguided correctness we now use the more derogatory name for our first inhabitants’ (www.southafrica.com/forums/open-board/5901-whats-name.html).

The debate concerning the terms Bushman and San is continuing, but most sources seem to prefer the term Bushman.

Bushman tribes and languages

There are similarities in the roots of recorded words from different Bushman languages (Traill 1978: 145). On the basis of these similarities, the Bushman tribes have been classified or categorised into three groups, namely the Southern, Northern and Central Group (Bleek 1929: i; 1956: iii-iv). Because there is no capitalisation of the click symbols in Bushman, Bleek and other scholars write the names of many Bushman languages that begin with a click, with a lower case second letter, e.g. /kam-ka !ke. To avoid confusion between these names and other words, the names have been written in this dictionary with an upper case letter after the click, e.g. /Xam. The classificatory designations S1, N2, etc, are placed between brackets in this book to avoid misinterpretation should these designations occur so close to other words as to be confused with them.

They are the following, given with the area where the ethnic groups live or lived (Bleek 1956: iii-iv):

Southern Group

S1:  /Kam-ka !ke , /Xam Old Cape Colony, south of Orange River

S2:  //Ŋ !ke Gordonia and Griqualand West

S2a: ≠Khomani Northern Gordonia

S2b: //Kxau Near Kimberley

S2c: //Ku //e Near Theunissen

S2d: Seroa Near Bethanie, southern Free State

S2e: !Gã !ne Eastern Cape north of Kei River

S3:  Batwa Lake Chrissie, Mpumalanga

S4:  /Auni Between Nossob and Auob (Auhoup)

S4a: Khatia, Xatia East of Nossob River, southern Kalahari

S4b: Ki /hazi West of Auob, southern Kalahari

S5:  Masarwa (Kakia) Kakia, southern Kalahari

S6:  /Nu //en Upper Nossop and Auhoup

S6a: /Nusan South of Auhoup

Northern Group

N1:  //K"au-//en Auen Oas-Ngami road to 19º E

N1a: Nogau Omuramba-Epikuro and Naukluft

N2:  !Kuŋ, Kung Ngami to Grootfontein to Okavango

N2a: Hei //kum, Heikum Near Etosha Pan

N2b: a dialect Near Ukuambi

N2c: a dialect Near Ukualuthu

N3:  !O !kuŋ Eastern half of central Angola

Central Group

C1:  Hie, Hiet∫ware, Masarwa Near Tati

C1a: Sehura South-east of Ngami

C1b: Mohissa East of Ngami

C2:  Naron, //Aikwe Near Sandfontein, west of Oas

C2a: Tsaukwe West of Ngami, north-east of Naron

C2b: Hukwe Caprivi region, north of the Tsaukwe

C3:  Hadza, Hadzapi At Lake Eyassi, north of Mkalama

Aspects of Bushman languages

The Bushman languages are characterised by clicks or suction consonants. Bleek (1929: i; 1956: 640) and Traill (1978: 137-138) describe the clicks and their mode of production as follows:

(a)  /, the dental or alveolar fricative click, pronounced by placing the tip of the tongue against the upper front teeth or alveolar ridge and withdrawing it rapidly;

(b)  //, the retroflex fricative click, made by spreading the tip of the tongue against the front palate and withdrawing it gently backwards, with a sucking sound;

(c)  !, the cerebral or palato-alveolar click, pronounced with the tongue-tip placed firmly on the point of division between palate and alveolar ridge, the back of the tongue placed against the velum and the sides of the tongue against the side upper gums, and releasing the tongue-tip sharply downwards, the resulting click resembling the sound of a cork being drawn from a bottle;

(d)  ≠, the alveolar click, formerly called the palatal click, pronounced with the upper part of the tongue behind the tip pressed firmly against the gum-ridge behind the central upper teeth, the back of the tongue raised to touch the velum, and the sides of the tongue raised to complete the space of rarefaction between velum and alveolar ridge, and bringing the front of the tongue sharply down, the resulting click resembling the sound made by a child when tasting something sweet;

(e)  the labial click, or lip click, usually represented in writing as a circle with a dot in the middle, is here represented as • •; and

(f)  the retroflex click, variously represented in writing as !! or /// .

Each of the clicks is pronounced with distinctive releases, accompaniments or effluxes, namely aspirated, ejected, fricative, glottal, nasal, preglottal, prevoiced, voiced, and others. In the Bushman languages, the voiced efflux is indicated in writing with g, as in /gã, ≠gã, //gã, !gã; the nasal efflux with n, as in /na, ≠na, //na, !na; the fricative efflux with x, as in /xã, ≠xã, //xã, !xã; the ejected efflux with k, as in /k?a, ≠k?ã, //k?ã, !k?ã; the aspirated efflux with h, as in /ha, ≠ha, //ha, !ha; the glottal efflux with a glottal stop, as in /?ã, ≠?ã, //?ã, !?ã and so forth (Traill 1978: 138).

In addition to the clicks, the Bushman languages have many other complexities of pronunciation involving both consonants and vowels. ‘The vowels of these languages are notable for their complex colourings, including plain, nasalised, breathy and pressed vowel colourings. These combine with each other to produce up to seven vowel colours for each of the five vowels’ (Traill 1978: 139).

A further point to note is that the Bushman languages are tone languages. Different words may have identical vowels and consonants but be distinguished by their tone alone (Traill 1978: 139). A high tone is indicated by the symbol ¯, and a low tone by the symbol _.

Conventions regarding Khoikhoi place names

The Khoikhoi languages are characterised by suction consonants or ‘clicks’. These clicks were, or could be, semantically determinative. In other words, the meaning of a word would be different if spoken or written with different click consonants, just as the meanings of English ask, bask, cask, task differ from each other because the initial consonant (if any) is different. For example, /gui means ‘one’; ≠gui means ‘many’; //gui means ‘lie down’; !gui means ‘to roar’; gui, with no suction consonant, means ‘milkbush’, Euphorbia species; also ‘to swarm’, ‘to teem’.

In 1856 the written form of four clicks was standardised, namely the dental click, /; the palatal click, ≠; the lateral click, // and the cerebral click, !. Prior to that date, various letters, combinations of letters, diacritic signs, and the like, were used to indicate the click sounds, e.g c, k, t, ’k, ’t, tk, t’¹, t’², tq, etc.

Nowadays place names of Khoikhoi origin in South Africa are written without click signs. However, to determine or verify the lexical meaning of a Khoikhoi place name it is necessary to establish which clicks, if any, were used when the name was given.

Khoikhoi nouns, and place names, are further characterised by gender endings, masculine -b (or -p), and feminine -s. These markers may, on occasion, be interchangeable, depending on the context. Thus ‘mud’ occurs both as ≠goab and ≠goas. In the present publication the gender markers may be omitted when the root word is being discussed as a component in a place name.

The Khoikhoi languages have become virtually extinct within the borders of the Republic. The possibilities of these languages producing any new place names are therefore very slight. There is, however, a large number of farms and other places with Khoikhoi names of which some may be submitted for approval as official names. The following principles used to apply in such cases.

(a)  Khoikhoi place names are normally written as one word.

(b)  Diacritical signs to indicate pitch, nasalisation, etc, are not rendered in writing.

(c)  Clicks are not indicated.

(d)  Established forms of spelling, such as Henkries, Kango, Knysna, are left unchanged.

(e)  Spelling should be changed as little as possible except that at the end of a name -p may be standardised as -b; -bep/-beb/-bes, -beep/-beeb/-bees, -biep/-bieb/-bies may be standardised as -bib/-bis; -sep/-seb/-ses/-sieb/-sies, etc, can be normalised as -sib/-sis; -rep/-res as rib/ris, etc.

(f)  The sound (x) is represented by ch, except in cases where it has already become established, e.g. Gamka, Khorixas. Compounds of which one element is Khoikhoi and the other Afrikaans, or English, are dealt with according to the principles applying to Afrikaans or English place names. We therefore write Gouritsrivier, Cango Caves, Kei Mouth, Naab se Berg. Khoikhoi place names that have been taken over via a Bantu language and have been Europeanised, are dealt with according to the principles applying to place names from Bantu languages, i.e. as one word if the second part is Afrikaans, and as two words if the second part is English, e.g. Gamtoosrivier, Kaya Fort.

Conventions regarding names from Bantu languages¹

The Bantu languages (sometimes inaccurately referred to as ‘African languages’) of South Africa and its neighbouring countries belong to four groups, namely:

(a)  The Nguni group, including Ndebele, Swazi, Xhosa and Zulu;

(b)  The Sotho group, including Northern Sotho, Southern Sotho and Tswana;

(c)  Tsonga and Shangaan;

(d)  Venda.

In most names from Bantu languages the generic term precedes the specific, e.g. Mlambongwenya, NokayaMatlala, Thabambovu.

Prefixes or class markers

Names in the Bantu languages are nouns, as in other languages, and to understand these Bantu language place names, it is necessary to take cognisance of the structure of nouns and the noun class system (Koopman 2002: 267). Each noun consists of a stem and a prefix or class marker. The stem conveys the meaning of the word, while the prefix indicates number (singular or plural), concord or subject marker (which set of concordial agreement should be employed for agreement with other parts of speech in a sentence), and so forth. As indicated above, the stem of the noun carries its ‘true’ or ‘real’ meaning, i.e. its lexical meaning (Koopman 2002: 267). By removing the prefix from the noun the stem is revealed, and it is frequently this stem that can be compared to Bushman words.

Zulu prefixes

The class prefixes or markers in Zulu are as follows (cf. Doke and Vilakazi 2005: xvii, Koopman 2002: 268):

According to the Meinhof numbering

Note:

(i)  Class 10 serves as plural class for classes 9 and 11.

(ii) Classes 12 and 13 are not found in Zulu.

(iii) Class 16 does not form nouns in Zulu, but only locative adverbs.

(iv) Class 17 is mainly of adverbial significance with shortened prefix ku-.

According to Zulu grammatical treatment

(Doke and Vilakazi 2005: xviii):

Sotho prefixes

In Sotho there are six different classes of nouns, with their prefixes indicated in bold type

Locative affixes

Nguni (Zulu, Xhosa and Swazi) place names, like other nouns, also have a locative inflection, typically with the prefix e- and one of the suffixes -ini, -eni, -wini or -weni, e.g. eThekwini, Empangeni. Some names take the prefix o- instead of -e, e.g. oFeni, and some have no suffix, e.g. eGoli; but for the most part locatives fit the pattern of eLangeni and eMthunzini. For the most part, the locative form is determined by the manner in which the noun or name is used in a sentence. Thus in a locative context the locative form must be used, e.g. eMgungundlovu, whereas in non-locative contexts the form uMgungundlovu is used (Koopman 2002: 114-15).

Names from the Nguni languages are often characterised by locative prefixes ku and kwa, e.g. KuHlange, KwaMbonambi, and by locative suffixes eni and ini, in conjunction with which the initial vowel is replaced by e or o, e.g. eMakhazeni, oThobothini. Although most names are locatives, many of them also have non-locative forms when they are used non-adverbially. Umnambithi is the Zulu name for Ladysmith, but in a sentence like ‘I am going to Ladysmith’ or ‘I live in Ladysmith’ it becomes Emnambithi. The syntactic variability of these prefixed vowels has led to uncertainty as to the correct capitalisation of some names and to the occurrence of forms such as uMnambithi, eMnambithi, Mnambithi and Umnambithi. The entries in this dictionary follow the conventions of international usage, eMnambithi under the letter E, uMnambithi under the letter U, etc.

The orthographic rules for Xhosa dictate that the place name be written without the prefix or subject marker. Names previously standardised with the prefix or class marker are now written without it, e.g. Dutywa (previously Idutywa), Mthatha (previously Umthatha), Ngcobo (previously Engcobo).

Clicks

The Nguni clicks are described as follows:

‘The radical form of the dental click is represented by c, the aspirated by ch, the voiced by gc, and the nasal by nc. The velar nasal consonant before the voiced form is represented by ngc, and the rare instances of the velar nasal before the unvoiced form by nkc. The radical form, c, under homorganic nasal influence, becomes ngc’ (Doke and Vilakazi 2005: 97).

‘The radical form of the palato-alveolar click is represented by q, the aspirated by qh, the voiced by gq and the nasal by nq. The velar nasal consonant before the voiced form is represented by ngq, and the rare instances of the velar nasal before the unvoiced form by nkq. The radical form, q, under homorganic nasal influence, becomes ngq’ (Doke and Vilakazi 2005: 684).

‘The radical form of the lateral click is represented by x, the aspirated by xh, the voiced by gx, and the nasal by nx. The velar nasal consonant before the voiced form is represented by ngx, and the rare instances of the velar nasal before the unvoiced form by nkx. The radical form x, under homorganic nasal influence, becomes ngx’ (Doke and Vilakazi 2005: 858).

Conventions regarding Dutch place names

Historically a fairly large number of geographical names are derived from Dutch. The Place Names Committee maintained the general principle of giving an Afrikaans form to certain names that were submitted in the Dutch form, e.g. Blouberg, Noupoort, Seekoeivlei and Sondagsrivier. Dutch names were retained where the Dutch spelling had become the accepted form and was being used as such, e.g. De Doorns, Franschhoek, Volksrust, Zeerust. This rule normally applies to important places. The Dutch spelling can also be retained if a post office or siding, etc, is named after the farm on which it is situated and the farm name has a registered Dutch spelling and the Place Names Committee was requested to retain the old spelling. Each case was decided on merit.

Conventions regarding Afrikaans place names

(1)  Simplexes

Names consisting of a simplex should not cause any problems.

(2)  Compound names

Compound names are normally written as one word. The following will serve as examples. Boesmankop, Bultfontein, Klawerkuil, Koedoedraai. Each of these names consists of two parts which in a non-onomastic context are two nouns. Sometimes the two parts are also linked with an -s, e.g. in Boesmanskraal, Broedersput, Klawershoek.

Note:

(i)  Sometimes there is uncertainty as to which one of the following pairs is admissible: laagte or leegte, rand or rant, olien or oliewen, wilge or wilger, etc. It is suggested that local custom should decide the issue. However, this freedom of choice no longer exists in the case of -stat and -stad. The form -stat is now retained only in Dingaanstat, in all other cases the form -stad is used.

(ii) Names of the following kind are also written as one word: Brugo (from Bruwer and Hugo), Krudoring (from Kruger and doring). Palfon (from Palmietfontein); also letter names such as Eljeesee (LJC) and syllable names such as Delmyn (from delwery and mynbou), etc.

(iii) Diepkloof, Nuweland, Swartrand, Warmbad. These, and most names compounded of an adjective plus a noun, are written as one word.

(a)  Hyphens

The hyphen is used in combinations with contrasting or distinguishing additions, such as agter, voor; groot, klein; nuwe, ou; bo, onder; noord, suid; oos, wes. They occur before or after a place name, e.g. Agter-Sneeuberg, Groot-Brakrivier, Nuwe-Mosbank, Noord-Rand, Somerset-Oos, Riebeek-Wes. In an ordinary word which is not a distinguishing place name, such additions are normally affixed to the ordinary word without a hyphen. In this way one finds Bo-Tautesberg, but Boplaas; Groot-Drakenstein, but Grootvloer; Wes-Transvaal, but Weskoppies. Note: Where certain vowels occur together in such combinations, a hyphen is used to facilitate legibility, e.g. Bo-erf (instead of Boerf), Perde-eiland (instead of Perdeeiland), etc. Hyphens are also used in place names consisting of two words linked by en, e.g. Haak-en-Steek, Rus-en-Vrede, Hoog-en-Droog.

(b)  Compound names that are written separately

Cases do occur in which parts of a compound place name are written separately.

(i)  We write Jan Smutslughawe. All compounds consisting of a Christian name and a surname are treated in this manner. The Christian name or names are written separately, the surname is dealt with according to the rule, and in this case it is joined to the next component. Initials are dealt with in the same manner as Christian names, in other words they are also written separately from the surname, e.g. JG Strijdomtonnel.

(ii) We write De Clerqville, Du Toitskloof, Le Rouxsrivier, Van Blerkskraal, Van der Merwesrus; in other words, when a place name consists of a surname of the type beginning with De, Du, Van, Van den, Van der, the prefixed parts in the place name are also written separately. Note: There used to be a tendency to write names of this kind all as one word, e.g. Dewetsdorp, Vanderbijlpark, Vanwyksvlei. Where this form of writing has become traditional it must be retained.

(iii) Ou is written separately when it precedes a personal name or nickname in place names such as Ou Thomas se Loop, or where the Ou no longer has any distinguishing or contrast value, e.g. Ou Handelspos. Note: Compare this to what was said under ‘hyphen’ above.

(iv) Place names consisting of combinations with the possessive se between the words are written separately, e.g. Beck se Plaas, Booi se Kraal, Lof se Dam, and many others.

(v)  Where the definite article Die forms the first word in a place name, it is written separately, and there is a tendency for the words following it to be written separately as well, e.g. Die Hollandse Saal, Die Onderste Aar, Die Ou Elands, Die Ou Vaal.

(vi) The components of place names consisting of a numeral plus a noun used in the plural form are written separately, e.g. Drie Susters, Veertien Strome, Twee Riviere, each element beginning with a capital letter. If the numeral is followed by a noun in the singular form, the elements are written as one word: Driefontein, Tweespruit, Vyfhoek.

(vii) Place names comprising phrases such as Agter die Berg, Hoek van die Berg, Koppie Alleen, Op die Tradou, are written separately. In these cases only the main words begin with capitals.

(viii) Place names consisting of two verbs such as Aanhou Hoop, Help Soek, Kom Kyk are written as separate words.’

To summarise:

1.   Afrikaans place names are generally written as one word, e.g. Drakensberg, Heksriviervallei, Krugersdorp, Witkleigat.

2.   Names incorporating a numeral are written as one word if the generic term is in the singular (Driefontein, Tweestroom), but as two words if the generic term is in the plural (Drie Fonteine, Vier Strome).

3.   A hyphen is used

(a)  in names containing an accumulation of vowels which could affect legibility and intelligibility, e.g. Goeie-uitsig, Perde-eiland;

(b)  in names consisting of elements connected by en, e.g. Berg-en-Dal, Hoog-en-Droog, Rus-en-Vrede;

(c)  in names that include adjectives or adverbs of contrast or distinction, e.g. Agter-Sneeuberg, Klein-Karoo, Noord-Kaapland. (Names in which the adjective or adverb is not contrastive or distinctive are written as one word, e.g. Boplaas, Kleinfontein, Weskoppies).

4.   Names written as separate words are

(a)  those incorporating a surname with an ‘unfused particle’, such as De Bruynshoogte, Van der Merwesrus;

(b)  those comprising a first name or initials plus surname, e.g. Abel Erasmuspas, JG Strijdomplein;

(c)  those incorporating the possessive morpheme se, e.g. Achab se Kloof, Lof se Dam;

(d)  those consisting of two verbs, e.g. Help Soek, Kom Kyk;

(e)  those incorporating the definite article die, e.g. Die Moot, Die Ou Sloot.

Conventions regarding English place names

English names already existing overseas that are given to places in South Africa in most cases retain their original spelling and form. English names that were formed and given in this country, however, may differ in spelling and form from the ‘imported’ names.

(1)  Simplexes

There should be no problems in regard to the spelling of simplex names.

(2)  Compound names

It is not easy to provide guidelines for the spelling of compound English place names. In practice a certain degree of irregularity is found. For example, names ending in crest, end, gate, hill, view, may be written either as one word or as two. Thus are encountered Wavecrest but Leisure Crest; Teaksend but Flats End; Westgate but North Gate; Foxhill but Calf Hill; Redhouse but Blue House; Aloeridge but Gravel Ridge; Bayview but Mountain View.

The Place Names Committee studied the written form of English place names in South Africa. An analysis of certain types of word combinations indicated that in some cases a pattern, or at least a tendency, can be discerned in the writing of English place names.

(a)  Names that are written separately

The following types of place names are normally written as separate words:

(i)  most names of which the first part is one of the following adjectives: Bonny, Golden, Lower, Old, etc; New, Rocky, Sweet, etc. Examples are Bonny Rest, Golden Grove, Lower Adamson, Old Place, New Centre, Rocky Hill, Sweet Hope;

(ii) names in which the generic term is still strongly felt as a common noun, e.g. Amatole Basin, Algoa Bay, Brighton Beach, Nagle Dam, Albert Falls, Cape Flats, Bretby Mine, Table Mountain, Sand River, Berg River Valley;

(iii) most names of which the second part is a plural noun, e.g. Birch Acres, Beecham Woods, Broken Slopes;

(iv) names with Crown, Fort, Loch, Mount and Port as the first part, e.g. Crown Reefs, Fort Beaufort, Loch Maree, Mount Frere and Port Alfred;

(v)  names of which the second part indicates the situation, e.g. Beaufort West, Boksburg North, Modder East, Randfontein South;

(vi) names consisting of a numeral plus a noun, e.g. Four Pines, Three Sisters, Twenty Four Rivers;

(vii) names consisting of Glen plus a personal name, e.g. Glen Karen, Glen Lynden;

(viii) phrases used as place names, e.g. Ascot on Vaal, Henley on Klip, Ebb and Flow, The Hole in the Wall;

(ix) names of which the first part is written with an apostrophe, e.g. Davey’s Halt, Gordon’s Bay, Eagle’s Crag, Lion’s Head;

(x)  names consisting of the abbreviation St plus a proper name, e.g. St Andrew’s, St Mark’s.

(b)  Names written as one word

The following kinds of place names are normally written as one word:

(i)  names ending in bourne, bury, combe, dene, hurst, lea, leigh, mere, wick, etc. For example Ashbourne, Woodbury, Ashcombe, Forestdene, Meadhurst, Birchleigh, Buttermere, Thornwick, etc;

(ii) names beginning with broad, cross, gay or middle, e.g. Broadlands, Crossmoor, Gayridge, Middlebrook;

(iii) names beginning with bal, brae, clan, craig, dal, holm, pen, sel, strath, etc. For example Balcraig, Braeview, Clanville, Craigend, Dalview, Holmleigh, Pendale, Selcourt, Strathcona, etc;

(iv) names composed of syllables or parts of names or words, such as Atcem (from Atlas Cement Company), Corobrick (Coronation Brick), Navex (Navarre exploration), Soweto (South Western Townships).

Conventions regarding dual names

In South Africa, where English and Afrikaans are two of the official languages, one may expect a large number of place names that are used in the one language also to be used in a translated form in the other language. In the course of time dual forms of names have established themselves for the same places.

Three kinds of translations can be differentiated:

(a)  names of which all the parts are translated, e.g. Bloedrivier – Blood River, Coffee Bay – Koffiebaai; Drieankerbaai – Three Anchor Bay, etc;

(b)  names of which both parts are ordinary words in the language concerned, but only the second part is translated, e.g. Bergrivier – Berg River, Melkbosrand – Melkbos Ridge, etc;

(c)  names of which the first part is a personal name and the second a generic term, e.g. Boshoffweg – Boshoff Road; Caledonplein – Caledon Square.

In terms of the country’s policy of multilingualism the members of each language group have the right to insist on the form they use in the natural context of their own language. For official purposes, however, precedence may be given to one form, that is, the ‘first of the two equals’. This precedence is based on the derivation and linguistic composition of the name, its age, the population group preponderating in the locality concerned, etc.

We write Brighton North, but Randfontein-Suid. In combinations of this kind the precedence form in the first type is normally English, and in the second, Afrikaans. In the first type the main word is derived from English, in the second from Afrikaans. The part of the name which indicates the name type or the point of the compass, or which constitutes a descriptive addition, will be indicated in the same language as the main word in the precedence form. Thus we write Bay Road (Bayweg), Bergrivier (Berg River), Oos-Rand (East Rand). The precedence form is given in italics. Combinations of which the first part is a word derived from an indigenous language are written as one word if the second part is Afrikaans, and as two words if the second part is English, e.g. Gamtoosrivier, Kaya Fort.

Origins of place names

As Nicolaisen (1976: 4) pointed out, place names have a power of survival that other words in a language do not have, and tend to outlive in some instances even the language in which they originated. Furthermore, their lexical or descriptive meaning becomes irrelevant when they are used as names, and they can be used by people speaking other languages. Over time, names may change, be adapted by later comers, reinterpreted, and gain new components, sometimes from the languages of later users of the names. Thus when names are examined for their original etymology and origin, they may be seen to be composed of elements from different languages. For example Tayside is derived from Indo-European *te or *ta ‘to flow’, and the word side is derived from Old English sīde, akin to Old Scandinavian sīda, Old High German sīta, Old Norse sītha, this noun derived from a prehistoric adjective (Gove 1961: 2111). The name Casseldale comprises elements from German and Old Norse, and so forth. Other place names in southern Africa display components from different languages, e.g. Tugela Ferry, Hout Bay, Randfontein South, Matopo Hills, Lambasi Bay.

Similarly, components of some place names in southern Africa may come from different Bushman languages. but it will be borne in mind that words in different Bushman languages display similarities, and in cases where a cognate is indicated, that word is thought to be similar to an older unrecorded word. The ancient Bushman languages were never recorded in writing, and we have no idea of the migratory or settlement patterns of these people, or of the influences their languages could have had upon each other and on their place names. Bushman words have been recorded only over the past few hundred years, and name components, probably much older, may be obsolete words similar to recorded words. In addition, it will be noticed that in many cases more than one Bushman word may be similar to the name component.

Structure of place names

Generally speaking, it may be stated that the structure of place names varies according to the language of origin of the names. Khoikhoi names, for instance, are characterised by masculine endings -b or -p and feminine endings -s, thus Goab, Neip, Goms. In addition, they may include locative morphemes such as -se, -si, -re, -ri, -te, -ti, -be, -bi, -bee, etc, which generally occur in the penultimate position, as in Nababeep, Komtes, Gobabis, Gobaseb.

Place names from the Bantu languages are characterised by locative prefixes such as Sotho ga (GaDikgale), Nguni kwa (KwaMashu), Venda ha (Hamasia), Tsonga eKa (eKa Mhinga), or by prefixes and suffixes, such as Nguni o-ini (Otobotini), e-ni (Edulini), o-eni (Obenjeni), etc.

What seems to be a universal feature of place names is that some are compounded, consisting of more than one element, while others are non-compounded, consisting of one element. An example is Springs as opposed to Table Mountain. Compounded names apparently preponderate. Compounded names generally consist of a term indicating the type of feature to which the name refers (mountain, river, etc), and a term that describes, qualifies or modifies it. The former is known as the generic term (or generic) and the latter the specific term (or specific).

An analysis of place names has indicated that the generic term usually refers to elevations and depressions, to the presence of water, or to settlement, demarcation, allocation or division of land. Thus occur names such as Attakwas Mountains, Enselsberg, Girinaris, Thaba Bosiu, Bonnievale, Gamkaskloof, Otjikoto, Tsubgaos, Albert Falls, Bronkhorstspruit, Nossob, Manzimnyama, Bellville, Boland, Nuhub, Roggeveld, Windhoek.

Specific terms are usually of a more varied character. They may be descriptive of colour (Blouberg, Heigariep), size (Great Fish River, Klein-Karoo), or shape (Gouib, Spitskop, Table Mountain); they may refer to fauna or flora encountered there (Crocodile River, Gamka, Taung, Knysna, Umhlume, Wilge River); they may be descriptive of the type of soil or rock (Duineveld, Kalk Bay, Modder River); they may refer to climatological conditions (Cold Bokkeveld, Noagore, Terra dos Bramidos); they may refer to people in authority (Colesberg, Land van Waveren), or to owners of land (Adendorp, Odendaalsrus), or someone who had a link with the place (Daveyton, Trappes Valley).

It will be noted that in most cases, and in most languages, the specific term precedes the generic, as in Piketberg, Soetap, Westville. Examples of names in which the generic term precedes the specific are Lake Mentz, Mount Frere, Ntababovu, Port Beaufort, Thabatshweu.

Because the original lexical meaning is irrelevant in place names, a generic term is sometimes not regarded as a word with meaning, and another, different generic may be added to the name; or else a new generic may be added to an existing name to indicate an extended entity. Thus names occur such as Breede River Valley, Karkloofrivier, Fairmount Ridge.

Hybrid place names are those in which elements from more than one language occur, e.g. Hout Bay, Kei Mouth, Buffalo Nek. Sometimes a generic term from a specific language is added to an existing name already containing a generic. In such instances tautology may occur, e.g. Cheridouws Poort, Nossob River, Heidelberg Hills, Waterkloof Glen.

The converse may, of course, also happen, namely that the generic term is omitted altogether, e.g. The Dargle (River), Knights (Station), Warden (Town). In such cases the names are said to display a zero generic term.

As in other languages, place names in Bantu languages may be either simple (or uncompounded), or compounded or composite, consisting of a descriptive term and a feature type designator, otherwise called a specific and generic term respectively, e.g. Thabankulu, ‘big mountain’, from thaba, ‘mountain’, kulu, ‘big’. The generic term is generally the first component of the name, but names do occur with the descriptive term first, e.g. Nhlazatshe, ‘green stone’, from nhlaza, ‘green’, tshe, ‘stone’. However, it is frequently the case that the latter are adapted Bushman names that preserve the customary Bushman structure of specific plus generic.

Meanings of place names

Conceptual, descriptive or lexical meaning

A cursory glance at any reasonably comprehensive list of place names will probably indicate a number of things. First, it will be noted that each name consists of a word, or of more than one word. Second, some of the names will immediately be ‘understandable’, or apparently semantically transparent, while others will be semantically opaque. Third, each name will be the appellation of a particular place or geographic entity.

Several comments on these observations may be in order. Place names, or toponyms, as they are also called, are proper names. As is the case with common names, or common nouns, each consists of a spoken or written form and an internal content, sense or meaning. A name also refers to, or denotes, an extra-linguistic entity.

There may be a number of reasons why the ‘meaning’ of a name is not readily discernible. The name, or a part of the name, may be in an unknown language; the name may be so old that the word(s) from which it is derived is or are no longer in current use; or the name may have been so greatly adapted (some say ‘corrupted’) that it is no longer recognisable.

In addition to the fact that there are names that do not have a readily discernible meaning, there are also names whose meanings are not relevant to the entities, e.g. geographical names used for cultural features. These and similar considerations have prompted linguists, philosophers and others to ask the question whether names have meaning at all. Some scholars maintain that proper names, including place names, have no meaning, while others aver that names are the most meaningful of all words.

Names are an integral part of language, and a primary function of language is to communicate. It would therefore seem to be unreasonable to assume that names have no meaning at all. Yet some clearly do not appear to ‘mean’ very much, except to indicate what a particular place is called. The solution to the problem seems to lie in what is meant by ‘meaning’.

Normally the meaning of a common name may be determined by consulting a dictionary. Thus the Concise Oxford Dictionary (6th ed, 1980, p 712) gives the meaning of mountain as, inter alia, ‘large natural elevation of earth’s surface, large or high or steep hill, esp. one over 1 000 ft. high’. This definition will hold for every mountain, and a mountain can be defined in terms of these characteristics, which are applicable to all members of the class of mountain. No such definition is possible for a proper name. Rietfontein may appear to mean ‘reedy fountain’, but when it is the name of a suburb of Pretoria, or of a post office, that is not, or no longer, its meaning.

In the above cases what is meant by ‘meaning’ is dictionary meaning, also termed logico-semantic, lexical, descriptive or conceptual meaning. Most scholars today maintain that proper names do not have current lexical meaning. Yet Table Mountain immediately calls to mind the idea of a mountain in some way resembling a table, namely being flat on top; Blue Lagoon obviously refers to the colour of such a feature. If, on the other hand, current lexical value were a prerequisite or defining characteristic of a name, Cango, Namib, Nossob, Parys and others would not qualify as such.

The problem may be solved by distinguishing between synchronic and diachronic meaning. Synchronic meaning is that which is discernible in the name at a point in time (the present), diachronic meaning that which may be discerned by viewing the name through the course of time. Synchronically, names are generally regarded as being devoid of lexical or conceptual meaning. Diachronically, when a name is viewed as being derived from one or more descriptive common noun(s), the etymological meaning may be either discerned or traced. Thus an apparently transparent name is homonymous with its appellative counterpart. For place names such as these, which have counterparts that are common nouns or appellatives (Table Mountain as opposed to ‘table mountain’), the terms ‘appellative name’ or ‘semi-appellative name’ have been coined. At which stage a common noun like ‘(the) pine grove’ becomes a proper name like Pinegrove cannot always be determined. But when it does attain this status of a proper name, the dictionary meaning becomes irrelevant. Thus a name like Coldstream, originally applied to a brook in which there is cold water, is also applied to a post office and a town. Rietfontein, originally ‘a fountain at which reeds grow’, has also become a residential suburb, a farm, a hospital, and so forth.

In the course of time the lexical or dictionary meanings of the common nouns underlying the place names may become forgotten, but the name itself survives. Thus there are place names still existing derived from languages which have died out, and of which even the peoples who spoke these languages no longer exist. In such cases research is necessary to trace the etymology.

The determination of that etymology has been described as only the first step in onomastic investigation. Yet it is often an indispensable step. Too often have name scholars fallen into the trap of assuming that a name means what it seems to mean. Take a name like Koppies, for example. Patently derived from Afrikaans, this name could mean ‘hillocks’, ‘little heads’ or ‘cups’. Since Koppies is a geographical name, one would presume the meaning to be ‘hillocks’. One would be wrong. Referring to a place in the Northern Cape, the name Koppies is derived from a Bushman or Khoikhoi word meaning ‘wild sage’ (Tarchonanthus camphoratus).

Similarly, an obviously English name such as The Coombs is not English at