CHAPTER 9

MAIMING DICTIONARIES

Jane Simpson

One of the questions commonly asked about Aboriginal languages is: how many words do they have? In this chapter we explore the answers to this apparently simple question. In doing so, we have to consider how many words there can be in any language. In turn, we examine the ways in which the words of a language can be documented and the purposes for which this task are undertaken. In short — making dictionaries.

HOW MANY WORDS ARE THERE IN A LANGUAGE?

People often try to compare languages in terms of the numbers of words the languages have. They may even do so by comparing the sizes of the largest dictionaries available for the two languages. The OED (Oxford English Dictionary: A New English Dictionary on Historical Principles, second edition) is, in twenty fat volumes, much bigger than the sixty-three page vocabulary of the Adelaide language in Teichelmann and Schürmann (1840).

This is clearly not a fair comparison, for many reasons. But the main reason is the basis for such a comparison. How do we measure the number of words in a language? First, what is a word? For instance, should the compound ‘firehose’ be treated as a single word different from ‘fire’ and ‘hose’? Languages differ widely as to what is considered a word. Second, are we talking about all words ever used by any speakers of that language? Or about all words used currently? Or about all the words used by an individual speaker and, presumably, stored somehow in that speaker’s mind? Or about all the words ever recorded of the language? These questions show how hard it is to compare languages with respect to the number of words in them.

ASSEMBLAGES OF WORDS

Let us look at the terms which the English language has developed for assemblages of words. Many of these, unfortunately, overlap in meaning. Thus, the total of words in a language is sometimes called the lexicon or vocabulary of the language (‘lexicon’ from the Greek lexis speech, word, phrase; ‘vocabulary’ from the Latin voca:bulum, a name). Now, if the language is taken to consist of all the words ever uttered by speakers of that language, it is not possible to have a complete record of the vocabulary of the language. Many words are never recorded because they have such a short life span, and hence never make it into dictionaries. But it is possible to gather a lexicon of most of the words ever recorded as being part of a particular language. Dictionaries like the OED which are organised along historical principles have this as a goal. A language which has a long history of writing will have dictionaries containing more words than languages which have a short history of writing. But such dictionaries will contain many obsolete or archaic words, not used by present-day speakers.

Another way of looking at the lexicon of a language is to narrow it down to the words known by present-day speakers. Dictionaries described as ‘Contemporary English’ are of this type. Different speakers know different words, resulting from their different specialisations. So the number of words known by any of the community of present-day English speakers is much greater than the number of words known by any one speaker of English. We have in effect a linguistic division of labour; we divide among ourselves the labour of knowing the meanings of different words, or knowing the meanings to different degrees of precision. For example, I know that a grebe is a kind of water-bird and I assume that ornithologists know more precisely what a grebe is, and can say ‘This bird is a grebe; that bird is not a grebe’. Many of the words in a dictionary of contemporary English I do not know, but I assume that among the community of present-day English speakers are some people who use these words because they know what they mean.

When we talk of someone’s ability to use words, we sometimes say she/he has a wide vocabulary. The words known by a speaker are sometimes also called the speaker’s mental lexicon. They range over a continuum. At one end are those that we know actively (we understand and use them). At the other end are those that we know passively (we understand them, but would never use them). Allowing for individual variation, and for the fact that what counts as a word differs from language to language, the number of words used by the average speaker (her or his active vocabulary) probably does not differ significantly across languages.

The word ‘vocabulary’, however, can mean something different: language-teaching textbooks often contain texts accompanied by lists of the words in the text with explanations. Such a list is often called a vocabulary or a glossary (glossary from the Latin glo:ssa ‘word that needs explanation’). A glossary consists of a list of words and glosses (explanations of the meanings of the words).

Notice that with these meanings of ‘glossary’ and ‘vocabulary’, we are introducing another type of word assemblage — written collections of words. Written language provides a way of preserving older stages of a language. Spoken language changes as each generation learns it, while written language preserves older words and older pronunciations.

When texts of written language are highly valued in a culture, such as the Chinese Classics (written from the eighth century BC onwards), younger people may be expected to learn to understand the older texts, even though the language is unfamiliar to them. And so commentaries on the texts may be made, explaining unfamiliar ideas, references to people and places, and including glossaries of unfamiliar words. This glossary may lead on to the preparation of a full-scale reference book containing an explanatory list of words, that is, a wordbook, dictionary (from the Latin dictio ‘choice and use of words’), or thesaurus (from the Latin the:saurus, ‘treasure’). Thus, by AD 100 there was a dictionary of Chinese which helped the current generation read and pronounce the language used in the Classics. This was an early monolingual dictionary, a dictionary in which the words of one language are explained in the language itself.

Dictionary is the general cover-term for such a reference book. However, sometimes dictionaries are contrasted with word-books, word-lists and thesauri in different ways. Dictionaries contrast with ordinary word-lists and word-books by having complex entries (that is, the explanatory material associated with a word). Making dictionaries has become a specialised trade, known as lexicography, while dictionary-makers are called lexicographers.

A specialised kind of monolingual dictionary, a special purpose dictionary, arises when a society with a written language develops areas of specialisation, such as trades, which have their own specialised vocabulary. Learning the trade becomes in part learning the terms of the trade, and so dictionaries of these terms are useful in helping this process. In 1527 one of the earliest English special purpose dictionaries, a glossary of law terms, was published.

Another kind of dictionary, a bilingual dictionary, is needed when writers of one language are learning to write another language. For example, some of the earliest Babylonian texts (seventh century BC) are word-lists, showing Sumerian pronunciations and Akkadian equivalents, thus enabling translation from the spoken language (Akkadian) into the language used for writing (Sumerian), and perhaps acting as a spelling guide. Until very recently, all dictionaries of Aboriginal languages have been bilingual dictionaries, the language of explanation being English in most cases (exceptions include the Diyari-German dictionary of J.G. Reuther, and the Arrernte-German dictionary of Carl Strehlow). But with Aborigines learning to read and write their own languages, some have started to prepare monolingual dictionaries.

WORD STORAGE IN AN ORAL CULTURE

Traditional Aboriginal societies had no large-scale writing systems, and so had oral cultures. Dictionaries are products of written cultures. Nevertheless, some of the functions of dictionaries have to be carried out in oral cultures too — learning words from other languages, explaining words from older stages of a language, explaining specialised terms. The first function, learning other languages, can be done orally, with little formal instruction. Traditional Aboriginal societies are, and were, multilingual societies. People learn other languages through having to talk to speakers of other languages, because of marriage, because of visiting for long periods in other people’s country, and because of a widespread convention of politeness that requires one to speak the language of the country one is visiting.

But the second two functions require greater efforts of memory and of social cooperation. Archaic language may be preserved in songs, and this is certainly true of the language of important song cycles of Aboriginal Australia. Young people learning the songs may be instructed by their teachers as to what the songs mean. In doing so they learn the older words preserved in the songs. Specialised terms of a trade will be learned by an apprentice from a master, as part of learning the trade. In Aboriginal Australia, this may include terms used in ceremonies, or even an entire respect language of the kind described by Alpher earlier (in chapter 7).

For words from older stages of the language or specialised terms of trade to be passed on, teachers must preserve information in their memories, while learners must train their memories, in order to learn. Socially accepted conventions of teacher-pupil relations develop so as to pass on information, while, in the society as a whole, there may be a division of labour as to who remembers what, and also as to how to keep a check on this.

The division of memory labour and of keeping check on memories has been achieved in many Australian Aboriginal societies by efficient social structures. In many areas, responsibility for knowing the songs, myths and ceremonies for different tracts of land is divided among people according to their family relationships. Sisters and brothers have responsibility for remembering information about both their mother’s country and their father’s country. A man must pass on to his children the knowledge of his father’s country, and his sister passes on to her brother’s children the knowledge of her father’s country. A brother and sister’s responsibility for their mother’s country requires them to check that the knowledge about that country is being correctly passed on to those whose father’s country it is (their mother’s brother’s children, for example). Likewise, some of their other cousins (father’s sister’s children) will check that the sister and brother are correctly passing on information about the country belonging to their common grandfather (father of both the cousins’ mother and the brother and sister’s father).

Concern with ensuring correct transmission of information has apparently been a part of Aboriginal societies for a long time. It is perhaps not surprising, then, that many traditional Aboriginal people have been interested in writing as a means of recording information, as a way of ensuring transmission of information, not simply to Europeans, but to their own grandchildren. Likewise, they have often enthusiastically taken part in dictionary-making, or made their own dictionaries, as a way of recording words and the associated concepts, artefacts, animals, plants, which they feel would otherwise be lost.

THE FIRST WORD-LISTS OF ABORIGINAL LANGUAGES

Word-lists of Aboriginal languages go back to Cook’s voyage to Australia, during which Joseph Banks recorded some words of Guugu Yimidhirr, spoken near Cooktown. Banks’s purpose was scientific, to record a sample of the language spoken by the people they met, just as he recorded the plants and animals they encountered. As Troy describes (in chapter 3), the people from the First Fleet tried to use Banks’s list for communication, and were surprised to find that the words recorded by Banks were not always understood by the local people.

Most of the best early vocabularies of Aboriginal languages come from missionaries. One of the earliest was that of Lancelot Threlkeld, who was sent in 1825 by the London Missionary Society to start work at Lake Macquarie, 110 kilometres north of Sydney. The Society encouraged their missionaries to learn the native languages in order to translate the scriptures into those languages and preach the gospel more effectively. As a result of learning Awaba, the local language, Threlkeld published in 1834 An Australian Grammar (Threlkeld 1834), which included the first extensive word-list of an Australian Aboriginal language to be published.

Threlkeld’s reason for publishing was partly scientific, and partly educational. He wanted to show the colonists that the Aborigines were human. As he writes:

it was maintained by many in the colony that the Blacks had no language at all but were only a race of the monkey tribe! This was a convenient assumption, for if it could be proved that the Aborigines of New South Wales were only a species of wild beast, there could be no guilt attached to those who shot them off or poisoned them as cumberers of the earth. (p 46)

Threlkeld decided not to use English spelling principles to write the Aboriginal language, because of the ambiguity and redundancy of the English spelling system. Instead, he adopted the spelling system that had been used by missionaries in the South Pacific. As a result he had to give an explicit pronunciation key, which makes his renderings of Awaba words less liable to mispronunciation.

Words in the word-list are listed (or ordered) in groups. These groups consist of lists of words on a particular topic. Topics include names of persons, names of places, parts of the body. As well, some groups consist of words with the same grammatical category (part of speech), such as common nouns. Within these groups, words are listed more or less according to alphabetic order. The entries are not complex, but do contain occasional ethnographic comments (see below for an example). Hence I have called it a word-list, rather than a dictionary.

Head word	English Gloss	Comment (ethnographic and occasionally etymological)
Ko-ro-wa-tul-lun,	The Cuttle fish,	literally, wave tongue.
Be-ra-buk-kán,	Sperm whale,	which is not eaten, only the black whale.

INTEREST IN THE HISTORY OF LANGUAGES

When Threlkeld was preparing his vocabulary, scientific interest in classifying languages had reached new heights in Europe. This was because several scholars had realised that languages were related to each other, and some could be thought of as forming families stemming from a common ancestor, or proto-language. Historical linguistics (the study of the history of languages) was getting underway. In 1822 Jakob Grimm put forward the hypothesis that German, Sanskrit, Greek, Latin and German all had their beginning as dialects of a single proto-language, Indo-European. The hypothesis received widespread publicity among scholars. Soon attention turned to Australia. Did the Australian Aboriginal languages all stem from one proto-language? Threlkeld noted that, although people living 100 miles or so apart were not able to understand each other at first glance, yet they were able to understand each other within a short space of time. He argued that this meant the languages could not be radically different, and tentatively suggested that Australian languages might all belong to one language family.

Perhaps the most influential spreading of this hypothesis among people working on Australian languages was by Captain George Grey, later Governor of South Australia. A few years after Threlkeld produced his An Australian Grammar; Grey went exploring around the southern part of Western Australia. On his travels he met Aborigines in different parts and tried to find out what he could about their languages. In doing so, he realised that much of what had been recorded on Aboriginal languages was quite unreliable. With the help of a friend he compiled his material into a book, published in 1840 and followed by a second edition. Like Threlkeld, Grey was struck by the correspondences among Australian Aboriginal languages. He noted that the word for water was kawi in Western Australia and kapi in Adelaide. He argued that this could not be coincidence — there must have been a language which was ancestor to the languages of southwestern Australia and the Adelaide language, and which contained a similar word for ‘water’.

Grey realised that reliable dictionaries were needed to check these correspondences. Being appointed Governor of South Australia in 1842 gave him the opportunity to do something about this, namely to act as patron and encourager of other people to prepare accurate word-lists and dictionaries. Previous work on the Adelaide language had been undertaken by people with little knowledge of the language, starting with the collection of a word-list on a French scientific expedition (Gaimard 1830-1835). After the establishment of the Adelaide colony in 1836, communication with Aborigines assumed more importance. Most colonists appear to have wanted them to learn English. However, there were moves to find out more about the language of the inhabitants of the Adelaide Plains. The second Protector of Aborigines in South Australia, William Wyatt, wrote a small account of the language, manners and customs of the Aborigines, mainly in order to instruct the population (reprinted as Wyatt 1879). Word-lists and phrases thought to be useful for conversing with Aborigines were published in the local newspaper in 1839 (Williams 1839). These word-lists of the Adelaide language all have simple structures, follow English spelling, and are often hard to interpret.

A DICTIONARY OF THE ADELAIDE LANGUAGE

A much deeper knowledge of the languages in South Australia was gained by several German missionaries, including Clamor Schürmann and Christian Teichelmann. Their arrival had been organised by a prominent English nonconformist Baptist, George Fife Angas, who was concerned about the welfare of the Aborigines, and who wanted the missionaries to help:

1. to preserve the native language;
2. to give the Aborigines a writing system and translations of the New Testament;
3. to teach the Aborigines to read.

Teichelmann and Schürmann were encouraged by George Grey and the then Governor, George Gawler, to publish their material, which they did in 1840 (see Plate 4). In the prefact they gave a number of reasons for publishing it: to help other Europeans learn the language, so as to be able to talk with Aborigines, and:

Plate 4: The frontispiece from Teichelmann and Schürmann’s study of the South Australian language (Teichelmann and Schürmann 1840).

to enliven the hopes of those who wish the christianization and civilization of their colored fellow-men, showing them that a race of human beings possessing a language so regular in its formation and construction as that of the South Australian natives, cannot be incapable of either; and to refute premature and unjust detractions concerning the mental capabilities of the Aborigines of Australia.

A third reason is scientific: ‘to render a small contribution or inducement to a general study of the manners, customs, and origin of these people’.

The book contains about 1,900 words, ordered alphabetically, and 160 or so illustrative sentences, both in the vocabulary and in the remainder of the book. The words are written using a version of Threlkeld’s orthography. The vocabulary contains words in almost every domain — material culture, ritual practices, kinship, bereavement, naming practices, plant names, animal names, meteorology, et cetera. The entries are more complex than Threlkeld’s, and are structured as follows (the italics indicate parts of the entry that are italicised to set them off from the rest of the entry). The entries in Teichelmann and Schürmann follow this pattern:

Head word (capitalised), part of speech. gloss/definition. example, translation of example. Comment; includes ethnographic, etymological, new term [that is, the word is a term for a new item introduced by the European invaders], some derived forms and cross-reference.

Kundo, s. chest; breast. kundo punggondi, to hurt one’s
feelings
Kundobakkurta, s. ornamental dots on the chest
Kundomanka, s. ornamental stripes on the chest
Kundomuka, s. the breast of the male
kundopungorendi, vn. to long; linger; languish; to be
uneasy; anxious

Unlike the previous word-lists, parts of speech are recorded for each word: s. ‘substantive’ = noun, vn. ‘intransitive verb’. Notice also that, while the first four words are clearly related, there is no attempt to bring out this relatedness in the structure of the dictionary entry. The fifth word, kundopungorendi, is probably related, since kundo seems to have two senses — one, the actual chest, and the other, the seat of emotions. But again, this relationship is not made explicit.

Exactly which words are included in such word-lists is usually dictated by the interest and circumstances of the dictionary-maker, and in the case of Aboriginal languages, of the people teaching the dictionary-maker. Dictionaries produced by people who are not native speakers of a language at the start of their work tend to focus on concrete things (words for plants, animals, food, artefacts, topography). Only with a much deeper knowledge of the language do words for complex emotions, sensations, thought and judgement appear. However, early word-lists and dictionaries often include words for concepts which were of contemporary concern. Detailed entries for terms for land ownership sometimes appear, suggesting that some colonists had scruples about their dispossession of Aboriginal people. Here is part of an entry from Teichelmann and Schürmann’s dictionary.

Pangkarra, s. a district or tract of country belonging to an individual, which he inherits from his father. Ngarraitya paru aityo pangkarrila, there is abundance of game in my country. As each pankarra [sic] has its peculiar name, many of the owners take that as their proper name, with the addition of the term burka; for instance, MuLleakiburka (Tam O’Shanter), Mullawirraburka (King John), Kalyoburka, Karkulyaburka, Tindoburka &c. [...]

In fact, most dictionaries of Aboriginal languages contain only a small fraction of the number of words in the languages. This is partly because, until recently, they have all been prepared by people who were not native speakers of the languages, and partly because the dictionary-makers have not had the time, resources and the source material available to some makers of English dictionaries.

In Teichelmann and Schürmann’s dictionary there is no English to Adelaide-language word-list. In fact, most of the early word-lists and dictionaries tend to be lists of vernacular words with explanations in English, much as if the dictionary-makers recorded the words they heard, and then compiled the results into lists. In terms of communication, it suggests that people wanted to know the meanings of words in Aboriginal languages that they heard, rather than wanting to know how to express meanings in that Aboriginal language. But the lack of such reverse lists must have reduced the usefulness of the vocabulary for colonists who wanted to use the book to help them learn the language.

Under Grey’s auspices, three more dictionaries of South Australian languages were published (Meyer 1843; Schürmann 1844; Moorhouse 1846). In 1857 Grey, who by then had moved to South Africa to be Governor of Cape Town, wrote to Threlkeld and Teichelmann asking for copies of their more recent material. They both sent him material, and noted that the languages were dying. Threlkeld wrote ‘I think a memorial of a language passing out of existence ought to be preserved for posterity ...’

Teichelmann sent a manuscript dictionary of the Adelaide language. It is a very impressive piece of work which contains about 2,500 words (including both head words and sub-entries). The words are more fully glossed than in the 1840 dictionary, and (except for concrete terms) usually include at least one illustrative sentence. It has some unusual words, which indicate his familiarity with the language and speakers. But perhaps the major innovation in Teichelmann’s manuscript dictionary is the structure of the entry. A good example of this is the word kundo. In the 1840 dictionary no attempt was made to show the fact that kundo has several senses, or that words could be derived from it. In the 1857 manuscript, however, three senses are given, and derived forms are listed under the senses:

Kundo, the chest, whereas ngammi the female breast;

2. anything projecting similar as the chest
3. as the seat of several passions, as
kundo wilta, ‘brave, bold; fearless’;
kundo punggondi, to dislike, hate; [...]
kundo punggorendi, to be concerned about; to be sorry;
kundo punggorendaii ngaityo yungakko, ‘I am concerned
about, or long for my elder brother.’ [...]

The hierarchical structure of this entry, indicated in the manuscript by indenting, contrasts with the flat structure of entries in the published dictionary.

The head word is capitalised. Senses are numbered. Some derived forms are still given under the head word entry, but many are given separate, although indented, sub-entries, with lower-case initial. Sub-entries can in turn have several senses, and even sub-entries of their own. Comparisons are made with lexemes that are clearly related. This more complicated structure makes for entries that capture the complexities of meaning better.

The dictionary has never been published. After the burst of dictionary publishing encouraged by Grey in the 1840s, interest in publishing dictionaries declined — such a burst was not seen again until the 1960s. Other large dictionaries compiled in the nineteenth century were either never published, or not published until long after their compilation. Such was the fate of James Günther’s grammar and dictionary of Wiradjuri (Wiradhuri), compiled in 1840 on a mission station in the Wellington Valley, but not published until 1892. Late in the century, other German missionaries produced some large dictionaries, perhaps the most impressive being J.G. Reuther’s unpublished manuscript A Diari Dictionary, originally written in German, and used at the mission at Lakes Kopperamanna and Killalpaninna in northern South Australia. Again, because it was unpublished, and because it was not translated into English until late this century, it had little influence on non-Aboriginal Australians. Some Diyari people learned to read and write their own language. But, like Awaba, Wiradjuri and the Adelaide language, Diyari has not survived, and these dictionaries remain important records of the languages.

COMPILATIONS OF WORD-LISTS FROM DIFFERENT LANGUAGES

In the late nineteenth century, attention turned to collecting small word-lists from many different languages, and comparing them. In 1876 R. Brough Smyth published The Aborigines of Victoria: With Notes Relating to the Habits of the Natives of Other Parts of Australia and Tasmania. This contained a variety of word-lists, some comparative (tables comparing the same concept — for example, ‘crow’ — in several Aboriginal languages), some organised by semantic domain. Over thirty of them were the result of sending lists of English words to people in different parts of Victoria and asking them to give the corresponding words in the local Aboriginal languages. It also contains lists of placenames and their meanings, something that has intrigued Australians for a long time. In 1886 E.M. Curr published The Australian Race: Its Origin, Languages, Customs, Place of Landing in Australia, and the Routes by which It Spread Itself over That Continent, which contained 300 word-lists of approximately 125 words, again the results of sending out English word-lists to people all over the country and asking for translations. The lists in Brough Smyth and Curr vary greatly in quality and reliability, but very often are the most important record we now have of the languages concerned.

While Brough Smyth and Curr were careful to record where the word-list came from, whether by named group (for example, the Karnathun group) or place (for example, Colac), the early part of the twentieth century saw the start of a pernicious practice which has done a great deal to hinder the recognition of Aboriginal languages. It probably stemmed from the fact that urban Australians were becoming less aware of Aborigines, and of the differences between Aboriginal societies. This was the spread of the ‘Aboriginal word books’.

In 1930 Justine Kenyon published An AboriginaL Word Book which has been reprinted many times since. It consists of lists of meanings for ‘Aboriginal words’ and ‘Aboriginal placenames’. The words come from many different languages, but there is no proper recognition of this. Several other books containing word-lists followed: H.M. Cooper’s AustraLian Aboriginal Words and Their Meanings, R. Praite and J.C. Tblley’s Place Names of South Australia, and perhaps the most popular of all, A.H. and A.W. Reed’s Aboriginal Words of Australia. The problem with these books is that, whether their authors intended this or not, they give the impression either that there is only one Aboriginal language, or that the differences between Aboriginal languages are quite unimportant. Furthermore, because they give no language sources for a particular meaning, it is very hard to check whether they are right or not. Finally, in instances where the meanings can be checked, they are often misleading or even wrong.

THE DEMANDS FOR MODERN DICTIONARIES

In the last fifty years, work on dictionaries of Aboriginal languages has grown enormously. Some excellent large dictionaries have appeared, mostly because the reading public for dictionaries has changed. In the nineteenth century the main audience for dictionaries of Aboriginal languages was non-Aboriginal people, those who wanted to learn the local Aboriginal languages, and those who were generally interested in Aboriginal languages. But in the last thirty years or so, bilingual education programs have become more widespread in schools on Aboriginal communities. Aboriginal people are learning to read and write in their own languages. So there is a need for learner’s dictionaries of Aboriginal languages, both monolingual and bilingual. Goddard’s dictionary of Pitjantjatjara/Yankunytjatjara (Goddard 1987) is an excellent example. There is also a continuing need for reference dictionaries of Aboriginal languages.

Because the audience for dictionaries of Aboriginal languages has changed, the need for different kinds of dictionaries is emerging. Adults require different dictionaries from children. People who have learned to read and write in English use dictionaries in different ways from people who have never learned to read and write. People whose first language is an Aboriginal language might use a monolingual dictionary in preference to a bilingual dictionary. On the other hand, people whose first language is English could find the English explanations of a bilingual dictionary quite helpful.

What do they use dictionaries for? Native speakers may use them for checking the spelling of words, or for learning the basic principles of spelling their own language, or for finding out the meaning of an unfamiliar word. For people who are learning to read and write, a dictionary is a useful tool which frees them from complete dependence on their teacher for checking spelling. Furthermore, listing normal English words with synonyms helps people literate in English to increase their English vocabulary.

The language to be included in a dictionary clearly depends on its intended audience. A learner’s dictionary would not include difficult words that are rarely used. A reference dictionary might include these words, but might exclude some slang words which people find offensive.

A more intractable problem is often what dialect is to be used in a dictionary. We face this problem with English dictionaries, since over time the English language has diverged widely in different places. Similarly, certain Aboriginal languages have dialects, as Sharpe describes for Bundjalung (chapter 5).

One way of representing these differences is to have different dictionaries for each dialect. In effect, this is what has happened with the group of dialects known as the Western Desert Language. Yankunytjatjara, Ngaanyatjarra, Pintupi can all be considered dialects of this language, but they each have their own dictionary. For school use, it is important to keep dialects distinct.

However, it is also useful to see what is common among dialects. And so another way of handling dialect differences is to have a single dictionary in which each head word is marked as to which dialect it comes from. By having a dictionary of many dialects of a language put together, one can see the circulation of words, and understand shades of meaning and extension of vocabulary. Such a dictionary would be David Zorc’s Yolngu Matha Dictionary. Perhaps the largest attempt to cover dialect variation is the on-going Arrernte dictionary project, which is providing dictionaries for a number of Arandic languages and dialects.

Once the lexicographer has decided what language to record in the dictionary, the next task is to decide what should be included in the dictionary entry for each word, and how it should be structured. We have seen how Teichelmann and Schürmann structured entries in their dictionaries. What information should be included is an interesting question. For instance, Teichelmann and Schürmann include information about the part of speech, thus that kundo is a noun. This is useful for a learner of a foreign language, who needs to know what part of speech a word is so that she/he can use it correctly in a sentence. But it is not generally useful for a native speaker of the language, who probably already knows how to use the word, or can work out how to use it from looking at its meaning.

Teichelmann and Schürmann also include ethnographic comments, such as the fact that people inherit a tract of land from their fathers. This kind of encyclopaedic information is very useful for learners, whether these are native speakers or foreign language learners. Including the cultural context in which a word is used also often helps to give a better understanding of the meaning of the word.

Kinds of information that are sometimes put into dictionary entries for Aboriginal languages include local or folk classifications for classes of words. Thus, ordinary English speakers classify some plants as grasses, others as trees, shrubs, vegetables, fruit, grains, and so on, while others they find hard to classify (for example, Venus flytraps). Not surprisingly, Aboriginal people have different folk classifications of plants, and these may be included in the dictionary entries for plants (for example, are they seed-bearing? are they edible? what kinds of implements can one make out of them?). Another area that may be included is the mythological significance of something, for instance, if fire is an important dreaming for speakers of a language, they may want this to be included in the dictionary entry for that word.

The heart of a dictionary entry is the explanation it gives of the meaning of a word. The dictionary-maker has to decide how the reader will best grasp the meaning of an unfamiliar word — whether the meaning of the word should be spelled out as a definition, or synonyms should be used, or pictures should be given, or illustrative sentences should be used. It seems that different kinds of words require different types of explanations — it might be more helpful to show a picture of a wombat than to describe it in words, but it would be hard to provide an unambiguous picture of an activity like ‘winning’.

Dictionary entries can be ordered in several ways. For languages like English and Aboriginal languages which use the Roman alphabet, a common way is according to the first symbol (letter) representing the first sound of the word, or what we customarily call alphabetical order. Of course, there’s no need to arrange the letters in that order. The Russian alphabet, for example, has the letter representing the sound pronounced [v] as the third letter of its alphabet. But English-speaking societies, by and large, have chosen to use the ABCDEF alphabetical ordering, and so have the lexicographers working on Aboriginal languages (with some exceptions, usually relating to sounds written with two letters, such as ng).

There are other ways of arranging material. For instance botanists and gardeners have arranged a special way of classifying plants, and in botany books, the plants are often listed in the order of that classification, rather than in alphabetical order.

A thesaurus is a whole dictionary arranged by topics. Two important decisions that have to be taken before making a thesaurus are, first, what will be the topics, and second, what order will the topics be arranged in? For instance, an English thesaurus might have a major topic devoted to architecture, while a thesaurus of Nunggubuyu (a north coast people) might have a major topic on dugong. The order and structuring of topics also reflects the importance of particular concepts in a culture. A society whose main source of meat is dugong, rather than beef, might give a different priority to dugongs and cattle in a thesaurus of its language. Several good thesauri of Aboriginal languages have appeared, including Heath’s thesaurus of Nunggubuyu (1982) and Douglas’s thesaurus of the Western Desert Language (1977) (the latter contains excellent illustrations).

CONCLUSION

Much work is being done creating dictionaries of Aboriginal languages, and there are some major dictionaries in preparation, including a Warlpiri dictionary and a comparative Arrernte dictionary. The production of dictionaries has been greatly assisted by computerisation. The availability of texts on computer means that concordances (lists of words in a text with references to where they occur) can be created automatically, providing valuable material for compiling dictionary entries. Putting dictionary entries on to a computer simplifies many of the lexicographer’s tasks. It is much easier to be consistent in structuring entries and checking cross-references. Furthermore, from one set of entries different dictionaries can be created semi-automatically. These might be alphabetically ordered, ordered by topic, abridged, or, in the case of bilingual dictionaries, reversed. Finally, computerised dictionaries can be useful for creating educational computer games for children, and for providing people using computers for word processing with computer spelling checker programs. On the other end of the continuum, there is growing interest in small vocabularies and picture dictionaries as useful material for Aboriginal Studies courses.

There is a great need to do more basic recording work on the Aboriginal languages which have no dictionaries. While no one ever saved a language just by making a dictionary of it, there is no doubt that, if Aboriginal languages are to achieve their proper place in the education of children, those children and their teachers must have access to good comprehensive dictionaries.

FOR DISCUSSION

1. Consider the validity of comparing the OED and Teichelmann and Schürmann’s vocabulary, bearing in mind the following points.
   1. The OED is a monolingual dictionary, while the other is a bilingual dictionary.
   2. The people who prepared the OED were native speakers of English, whereas the two people who wrote down the vocabulary of the Adelaide language were native speakers of German.
   3. In the nineteenth century, when these dictionaries were compiled, English had been a written language for over five centuries, whereas the Adelaide language had not been written down before the European invasion in 1836.
   4. It took more than forty-four years to complete and publish the OED. By contrast, the two authors of the Adelaide language vocabulary arrived in Adelaide in 1838 to start learning the language, and eighteen months later published their vocabulary.
   5. In 1838 there were probably not more than a few hundred speakers of the Adelaide language.
2. In this chapter, partial explanations of words like ‘vocabulary’, ‘glossary’ are given together with an explanation of what they meant in the language from which English borrowed the term. The source of a word is called its etymology. Many dictionaries contain etymologies for words (indeed, etymological dictionaries are devoted to providing etymologies). Why do dictionaries include etymologies? Do they help you understand the meaning of words? Can you know the meaning of a word without knowing its etymology? Look at the list of words below, (taken from a learner’s dictionary (Goddard 1987, 62), and provide etymologies for them.

   kalatji	glass, mirror
   lapaturi	toilet
   mutuka	car
   nipa-nipa	scissors
   tjiila	prison, gaol, the lockup
   tjapila	shovel
   tina	lunch
   tiinta	1. tent, canopy on vehicle. 2. piece of canvas, tarpaulin
   walypala	white man
   waya	1. wire 2. wire handle on billy 3. power cord 4. things made of wire, for example, a cooking-grill

   And now some harder ones:

   kaliki	canvas
   laimi	(noun) plaster
   makati	rifle
   miita	spouse
   tiki	(on) credit
   tjaintjimilani	cash a cheque

3. In the nine years between 1838 and 1846 six dictionaries of Aboriginal languages were published. In 1857 Threlkeld wrote to George Grey that he had hoped to publish his updated dictionary by subscription, but Australians showed a lack of interest in science: ‘there is but little Encouragement for such things in this Colony’. In 1971 Geoffrey O‘Grady (O’Grady 1971) could only think of four dictionaries of Aboriginal languages published in the twentieth century. Think of reasons why people lost interest in publishing dictionaries. Track down O‘Grady’s article and compare the reasons with his suggestions.
4. You are presented with a dictionary of ‘European’. On page one you see the following lists of words:

   abercoc	apricot
   alamo	poplar tree
   aller	to go
   antworten	to answer
   amba	it’s all up!
   arktos	bear
   alcatraz	pelican

   By dint of asking your Catalan, Spanish, French, German, Russian, Greek and Portuguese friends, you find out that these words come from their languages. What uses could you make of such a book?
5. Look at the following entry from the Warlpiri dictionary (provided by Mary Laughren).

   JAALJAAL(PA) [Noun or Preverb] (Yuendumu dialect)

   1. feeling, hunch, premonition

      Ngaju karna jaaljaal jarnmi nyiyakurra. Marda kapulu kurdu ngajunyangu pakarni. I have a feeling about something. Perhaps they are going to hit my son. Kari! Nyiyakurra karna jaabaal-jarrimi miyalu nyampuju? Ngajunyangukurra kajanyanukurra? Jungajuku ngajunyangukurralparna jaaljaal-jarrija manjurruju. Jungajuku wantija miyalu-purdanji. Oh, why do I have this feeling in my stomach? Is it because of my son? It was because of my son that I had this twitching feeling. He fell down on his stomach.

      • When used with the Allative case

   2. (Yuendumu and Lrxjamanu dialects) have an urge, desire (to do something), want, feel (like), have a yen for.

      Jaaljaal-jarrimi, ngulaji yangka kujaka yapa kiyikiyi-jarri manu jalajala-jarrimi nyiyarlanguku majuku marda ngurrjuku marda — ngurrju-maninjaku — yangka nyanungu yangka yapa — wati marda, karnta marda, kurdu marda. Jaaljaal-jarrimi is when a person feels like or gets the urge to do something, either something bad or somethiny good — to make something — just that person himself — either a man or a woman or a child. Jaaljaal-jarrimi karna janyungukupurda. I feel like some tobacco. [Alma Granites Nungarrayi, Y27.9.88]

      Wara! Janyunguku karna jaaljaal-jarrimi. Oh I really feel like some tobacco. [Jean Napananyka Brown Y 1988]

      • When used with the Dative case

   Work out the different kinds of information contained in the entry, how they are coded (for example, by typeface), and speculate as to how readers could use this information. [NOTE: ‘Allative‘ is the case of motion towards, expressed by the suffix -kurra in the examples. It is roughly equivalent to the English prepositions ‘into’ and ‘onto’. ‘Dative’ is the case of the recipient, or the thing desired, of the person benefiting from something expressed by the suffixes -ku and -kupurda in the examples. It is roughly equivalent to the English prepositions ‘to’ and ‘for’].

6. Provide a dictionary entry for the English word ‘urge’ in ‘have an urge to’. Compare your explanation with one given in a large dictionary, such as the Shorter Oxford English Dictionary or the Macquarie Dictionary.
7. Suppose the sounds of an Aboriginal language are written as follows: aiu p t ty k m n ng rn ny l rl ly r rr w y (all the double letters represent a single sound, just as the sh sound in ‘shop’ represents a single sound). How would you order these sounds? Do you foresee any problems with your ordering?
8. Discuss the advantages and disadvantages of organising a word list alphabetically or by topics. You might consider points such as:
   • is it a learner’s dictionary?
   • if it’s a learner’s dictionary, what’s the first language of the learners?
   • what uses is the dictionary intended to have? Learning about a particular topic? Searching for synonyms? Checking the spelling of a word? Looking up the meaning of an unfamiliar word?
9. ‘A picture is worth a thousand words’. Discuss the truth of this statement with respect to sets of words — for example, compare words for things, such as saucers, bowls and plates, with words for emotions, such as anger, love and hope.

REFERENCES

Austin, P.
1991	‘Australian Lexicography’. In F.J. Hausmann, O. Reichmann, H.E. Wiegand (eds), Encyctopaedia of Lexicography, Walter de Gruyter, Berlin, 2638-41.
Capell, A.
1971	History of Research in Australian and Tasmania Languages. In T.A. Sebeok (ed), Current Trends in Linguistics 8, Mouton, The Hague, 661-720.
Cooper, H.M.
1952	Australian Aboriginal Words and Their Meanings, South Australian Museum, Adelaide.
Curr, E.M.
1886-87	The Austratian Race: Its Origin, Languages, Customs, Place of Landing in Australia, and the Routes by Which It Spread Itself over That Continent, 4 vols, J. Ferres, Government Printer, Melbourne.
Dixon, R.M.W., W.S. Ramson and M. Thomas
1990	Australian Aboriginal Words in English: Their Origin and Meaning, Oxford University Press, Melbourne.
Douglas, W.
1977	Illustrated Topical Dictionary of the Western Desert Language: Warburton Ranges Dialect, Western Australia, Australian Institute of Aboriginal Studies, Canberra.
Gaimard
1830- 1835	Vocabulaire de la Langue des Habitants de Golfe Saint-Vincent. In Voyage de Découverte de l’Astrolabe 1826-7-8-9 sous le Commandement de M.J. Dumont d‘Urville, vol 1 Philologie, J. Tastu, Paris, 6-8.
Goddard, C.
1987	A Basic PitjantjatjaralYankzcnytjatjara to English Dictionary, Institute for Aboriginal Development, Alice Springs, Northern Territory.
Grey, G.
1845	‘On the Languages of Australia’, Being an Extract of a Dispatch from Captain G. Grey, Governor of South Australia, to Lord Stanley. Communicated by his Lordship, Journal of the Royal Geographical Society 15, 365-367.
Günther, J.
1892	Grammar and Vocabulary of the Aboriginal Dialect Called the Wirradhuri. In J. Fraser (ed), An Australian Language, as Spoken by the Awabakal [L. E. Threlkeld], Appendix D, 56-120.
Heath, J.
1982	Nunggubuyu Dictionary, Australian Institute of Aboriginal Studies, Canberra.
Kendon, A.
1988	Sign Languages of Aboriginal Australia: Cultural, Semiotic and Communicative Perspectives, Cambridge University Press, Cambridge.
Kenyon,J.
1982	Aboriginal Word Book, Lothian, Melbourne.
Kilham, C. et al.
1986	Dictionary and Source Book of the Wik-Mungkan Language, Summer Institute of Linguistics, Darwin.
Meyer, H.A.E.
1843	Vocabulary of the Language Spoken by the Aborigines of the Southern and Eastern Portions of the Settled Districts of South Australia, J. Allen, Adelaide.
Moorhouse, M.
1846	A Vocabulary and Outline of the Grammatical Structure of the Murray River Language, A. Murray, Adelaide.
O’Grady, G.N.
1971	Lexicographical Research in Aboriginal Australia. In T.A. Sebeok (ed), Current Trends in Linguistics 8, Mouton, The Hague, 779-803.
Praite, R. and J.C. Tolley
1970	Place Names of South AustraLia, Rigby, Adelaide.
Reed, A.H. and A.W.
1965	Aboriginal Words of Australia, A.H. and A.W. Reed Pty Ltd, Sydney.
Reuther, J.G.
1981	The Diari, microfiche ed. Translated from the German by P.A. Scherer, Tanunda. Australian Institute of Aboriginal Studies, Canberra.
Sandefur, J. and J. Sandefur
1979	Beginnings of a Ngukicrr-Bamyili Creole Dictionary, Summer Institute of Linguistics, Darwin.
Schürmann, C.W.
1844	A Vocabulary of the Parnkalla Langvcage, Adelaide.
Simpson, J.A. and E.S.C. Weiner
1989	Oxford English Dictionary, 2nd ed, Oxford University Press/Clarendon Press, Oxford/New York, 20 vols.
Smyth, R.B.
1876	The Aborigines of Victoria: With Notes Relating to the Habits of the Natives of other Parts of Australia and Tasmania, 2 vols, John Currey, O’Neil, Melbourne.
Strehlow, C.F.T.
n.d.	Aranda-Loritja-English Dictionary, Translated from the German by P.A. Scherer. 223-page typescript held at Australian Institute of Aboriginal Studies.
Teichelmann, C.G. and C.W. Schürmann
1840	Outlines of a Grammar, Vocabulary, and Phraseology, of the Aboriginal Language of South Australia, Spoken by the Natives in and for some Distance aroicnd Adelaide, Published by the authors, at the native location, Adelaide.
Threlkeld, L.E.
1834	An Australian Grammar, Comprehending the Principles and Natural Rules of the Language, as Spoken by the Aborigines in the Vicinity of Hunter’s River, Lake Macquarie, &c. New South Wales, Stephens and Stokes, Sydney.
Williams, W.
1840- [1839]	A Vocabulary of the Languages of the Aborigines of the Adelaide District, and other Friendly Tribes, of the Province of South Australia,
	published privately by A. McDougall, Rundle St, Adelaide, and then republished in South Australian Colonist, Adelaide, 297—98.
Wyatt, W.
1879	Vocabulary of the Adelaide and Encounter Bay Tribes, with a Few Words of that of Rapid Bay. In J.D. Woods (ed), The Native Tribes of South Australia, (no publisher) Adelaide, 157-82.
Zorc, R.D.
1986	Yolngu-Matha Dictionary, School of Australian Linguistics, Darwin Institute of Technology, Batchelor, NT.