Microsoft word - dig-early_letters.doc

Letters from Early Australia – Linguistic
Variation and Change
I Introduction1
This study of Australian English in general and the language of letters from nineteenth century immigrants in particular developed from two different interests. The first was a personal interest in Australia and its history, linguistic and social, the second an explorer’s fascination with a field he knows to be largely unknown territory. The letters and diaries that are preserved from 19th century Australia show the gripping lives of convicts, the lonesome toil of farmers and the daily experiences of city dwellers. Apart from the personal and historical interests that are connected with these testimonies, the language of the letters is also worthwhile studying. This is intended in the present analysis. The study of Australian English (AE) is still a rather neglected topic of linguistic research especially if it is compared with the work done on the American (AmE) and British (EngE) varieties of English. This situation has improved since the days of the pioneering works of Baker (1966), Mitchell and Delbridge but outside Australia there is virtually only a handful of studies to be found. Notable exceptions are the works of Dabke (1976), Görlach (1991) and Leitner (1984, 1989, 1990). The first works on Australian English focused on the origin of a particular Australian pronunciation, the mixing of dialects and the vocabulary of settlers and convicts. But in the 1970s the interest in the historical study of Australian English declined rapidly and gave way to studies of present-day usage with special attention devoted to the language contact situation of Aborigines and immigrants, sociolinguistic variables 1 This paper draws partly on the author’s MA thesis titled Early Australian Letters – A Linguistic Analysis which was completed in 1996 at the University of Regensburg. For help and comments on earlier drafts of this paper I wish to thank Philip A. Luelsdorff, University of Regensburg, and Brian Taylor, University of Sydney. and differences and similarities between the usage of English in Australia and other parts of the world. To the best of my knowledge, there is no historical study on Australian English that went beyond the topics of phonology, the mixing of dialects and lexis. Therefore the presentation of a history of Australian English, the investigation of letter-stylistics, determiners, verbal concord, topicalization, modality and orality vs literacy in a corpus (140,000 words) of nineteenth century letters and diaries and an attempt to link the features found there with present-day usage is a valuable contribution both to the historical study of AE as well as to the study of current AE. II From English in Australia to Australian English
Theories on the Origins of Australian English The fact that Australian English has a distinct pronunciation and lexis and that it is a variety of English in its own right is undisputed. But exactly how, when and where Australian English was ‘born’ is still a matter of intense debate. Three different approaches were developed to answer this question. The first states that AE developed independently all over Australia into a uniform dialect from the same set of ‘ingredients’ (cf. Bernard 1969), the second that it developed in England and was only later transplanted (cf. Hammarström 1980), and the third that it was shaped in Sydney and then spread from there all over the continent (cf. Horvath 1985). The discussion of the origin of the three acknowledged sociolects of AE, namely Broad, General and Cultivated Australian, received only scant attention in the studies dealing with the possible origins of an Australian variety of English (but cf. Horvath 1985, Gunn 1992). These categories were established by Mitchell and Delbridge in their revolutionary works on Australian pronunciation and have ever since been accepted terminology.2 Horvath (1985) claims that there were two different varieties from the start, which she calls high and low prestige varieties. The use of the respective sociolects corresponded to the social status of the speakers. From these two initial sociolects General Australian would finally emerge as a new accommodated variety. Gunn (1972) does not agree with this 2 Cf. Mitchell (1946), Mitchell and Delbridge (1965a, b). Letters from Early Australia – Linguistic Variation and Change explanation. His view is that Broad was spoken by the overwhelming majority of people in Australia and that Cultivated "developed later out of attempts to speak Standard English."3 A Critique of the Theories Presented The approaches presented fail to answer some important questions that are relevant to the purported mixing of dialects. First, they do not define the term dialect and whether it includes phonology, lexis or grammar or any combination of these. This is relevant since most studies do not go further than a simple comparison of the phonological systems of Broad Australian and some EngE dialect. Second, most of the theories presented share the implicit assumption that there is only a single unified Australian English variety. This assumption is derived from the alleged uniformity of AE in the fields of lexis and pronunciation. Cum grano salis, this is true. But there is also the undeniable fact that AE has three different sociolects, which are known to be distinct at least in the field of pronunciation. In order to avoid this problem, most studies base their investigations on Broad Australian and do not address the question how the two other sociolects of AE do fit into the proposed framework. Finally, on the basis of phonological and lexical investigations, almost all studies categorically discount any Irish influence. This seems doubtful in view of the large numbers of Irish immigrants and convicts. The Origins of Australian English First, it is necessary to establish whether there was such a thing as a language contact situation in late eighteenth century Australia. Here, the following can be stated: There were a substantial number of people involved over a considerable time. The contact they had was frequent and permanent. There were, however, from the beginning two very distinct social strata that allowed substantial contact only within but not across them. Hence, both strata have to be considered separately. The upper echelons of early Australia shared a very uniform Standard Southern EngE dialect. Therefore any accommodation was unnecessary. On the other hand, the convicts and emancipists, the assisted immigrants and the common soldiers formed another closed social stratum. The dialect they spoke was a relatively uniform urban lower class dialect with a Southern EngE basis. English as it was spoken in early nineteenth century Australia was thus relatively uniform among each of the two social groups. Moreover, both had a Southern EngE basis and so did not differ much from each other. The use of lexis was undoubtedly extremely unified from very early on. This is due to two reasons. First, the lexica of the individual systems of the speakers involved was not very different within the two social groups. Second, there were functional restrictions on the use of many words since the new environment, the convict system and the new work methods made the use of a specially designed vocabulary obligatory. This relative uniformity changed from the 1820s onwards when larger numbers of Irish convicts and immigrants poured into Australia and when settlers from all over the world, especially in the gold rush period in the 1850s, came in increasing numbers to Australia. The linguistic situation did not change for the upper social stratum in Australia. Their linguistic systems were constantly reinforced in their use of Standard EngE by gentlemen newcomers. It was only the slow weakening of social barriers and the gradual development of a national identity in Australia that eventually put some pressure on the language use of this group. But it is definite that their use of language remained relatively unchanged in the course of the nineteenth century. This is reflected in formal writing from that period. The situation was different for the lower social strata speaking an early version of Broad AE. The new arrivals, especially the Irish, had individual systems that differed from those of the established settlers. Moreover, the number of people coming to Australia in the 1820s and onwards was very high in comparison with the resident population and contact between the two groups was frequent and permanent. Since life in Australia was extremely different from life in Europe the factor of colonial experience to a great extent determined the social standing of a person. Hence, there was great social pressure on the ‘new chums’ to blend in with the ‘old hands’ as quickly as possible which favoured uni-directional accommodation processes among the newcomers. This is most obvious in the adoption of colonial pronunciation features, for attitudinal reasons, and of colonial vocabulary, for attitudinal and functional reasons. By way of summary it can be stated that Australia did not evidence a mixing of dialects in the sense that a number of radically different dialects accommodated to each other. There were from the beginning two distinct social groups that experienced different linguistic developments. Both Letters from Early Australia – Linguistic Variation and Change groups showed an ‘Australian’ lexis early. The later social developments enabled the two distinct language uses to mix and to form what is now called General Australian. Finally, the term Australian English can be defined as follows: Australian English, if used to describe the linguistic situation in 19th century Australia, denotes a variety of English spoken by people in Australia who share to a great extent features of lexis and pronunciation but can evidence differences in their morphological and syntactic systems. III The Corpus – Letters and Diaries
The self-edited corpus used for the present investigation comprises altogether 143,565 words in 359 letters, diaries and various excerpts. Each letter was given a corpus designation (CD) which consists of a string of letters indicating the source, e.g. catch for letters written by a person named Margaret Catchpole, and a serial number (1-x). The data for the corpus come from three different sources. First, there is a number of unedited letters from the Mitchell Library in Sydney, New South Wales. These can be further divided into three groups, namely letters from the Campbell family, letters by and about Margaret Catchpole and letters to John Piper. The second source is a collection of all extant letters of the Reibey family. They are edited in Nance Irvine's book Dear Cousin: The Reibey Letters. The last and by far the largest part of the corpus comes from Patrick O'Farrell's book Letters from Irish Australia 1825-1929, who collected letters from Irish immigrants to Australia. These are to be found in the Public Record Office of Northern Ireland and the State Paper Office in Dublin. Prominent in the latter collection are letters by James Twig and the Maxwell family. Figure 1 shows the total number of words to be found in the corpus for a particular period. This reveals a slight preponderance of linguistic evidence from the late nineteenth century. Since most of the letters in the corpus were written not by seasoned Australians but by recent arrivals from the British Isles they can be expected to reflect more the individual and dialectal standards of the immigrants than the contemporaneous Australian pattern. Figure 1: Number of words per decade The next four tables show the origins of the letter writers, the states where they lived in Australia, the Australian residences of the recipients and the overall residences of the recipients. Figure 2 reveals that the overwhelming majority of the letter writers in the corpus came from Ireland. This is, however, counterbalanced by the fact that the letters from Irish immigrants are generally much shorter. To be more specific, the number of words in the Irish letters is 89,622 compared to a number of 53,943 words for the other letters. Figure 3 shows that most of the letters written in Australia come from the state of Victoria. The states of New South Wales and Western Australia are also fairly represented. Letters from the other states are either not in the corpus, e.g. South Australia, or only in small numbers, e.g. Van Diemen’s Land. This fact makes an investigation of regional differences for the states of New South Wales, Victoria and Western Australia possible though this was not attempted in the present study. Letters from Early Australia – Linguistic Variation and Change Figure2: Origins of the letter writers Figure 3: Australian residences (writers) A similar picture can be gathered from Figure 4 showing the Australian residences of the recipients of the letters that were sent either to or within Australia. The only difference lies in the number of letters sent either to or within NSW. This reflects the fact that the letters written in Western Australia and Victoria were mostly written home and not sent to another part of Australia. Figure 5 indicates that most of the letters of the corpus were sent to Ireland. Figure 4: Australian residences (recipients) Figure 5: Residences of all recipients 4 The abbreviations read: NSW = New South Wales; QuL = Queensland; VDL = Van Diemen’s Land; Vic = Victoria; WA = Western Australia. IV Stylistic and Linguistic Analyses
Formal letters are distinguished from personal letters in that they were written to serve a certain purpose. Consequently, all formal letters focus on a single topic which explains the relatively high frequencies of words like send (28), country (13), state (12), pay (9), power (9), money (8), present (7), situation (7) and poor (6). The formality of the language is also evident from the frequencies of formal forms of address like Mr (35), Sir (28) and Mrs (10) which indicates that the relationship between the writer and the recipient of the letter is formal and not friendly. The formal letters are all structured according to the following pattern: They all start with a formal address that ranges from Sir (5) and Dear Sir (5) to Hon’ble Sir and Sirs (1) to Gentlemen (1). This then is followed immediately by the statement why the letter has been written: Dear Sir, The painful task has devolved upon me of apprizing you of the death of your gallant and lamented Brother, Alexander [.]. [Camp 1] If a letter is a reply to another letter, the date of the first letter is given and a short summary of it is presented. Every effort is made to ensure that the relationship between the two letters is clear. The paragraphs of the first letter are then answered in an ordered way: In your second Paragraph you state, that altho you had permission for the building a vessel [.]. In your fourth paragraph you state that you have been obliged to have recourse to the Salted Pork intended for the settlements [.]. Your letter of the 7th is fully answered by my pre-ceding paragraphs [.]. [Piper 2] Often a petition is presented that requires the recipient to undertake a certain duty or to send back some desired information: i therfore Hon’ble Sirs look up to you for assistance hoping you will enable me by an order from your Hon’ble Board to receive my pay as second officer [.]. [Reib 1] The letters close with non-committal formulas like "Your most obedient, humble servant" [Piper 2] or "Thanks for disinterested kindness" [Reib 16]. Letters from Early Australia – Linguistic Variation and Change The degree of formality of the address correlates with the personal rela- tion of the letter writer and its recipient. The explicit naming of an ad-dressee is significant since most formal letters refrain from this, especially when the actual reader of the letter is unknown. This can be the case when the letter is directed to an office or a bureaucratic institution. The address with the first name, e.g. "My Dear Patrick", is undoubtedly the most personal of all and reveals an intimate relationship between the writer and the recipient of the letter. An even greater intimacy can be as-sumed when the first name is modified with an adjective, e.g. "My Dear old Lottie". The adjective dear is used ambiguously. On the one hand, it can indicate that the addressee is cordially regarded, e.g. in "My Dear Cousin", or it is used as a faded phrase of esteem that can be used in every context, e.g. in "My Dear Sir". The closing formulas of the letters can also show how the writer and the reader relate to each other. In personal letters the conclusion is very short and does not form a complete sentence. It can even be wholly omitted and the letter ends then very abruptly with the signature: My dear aunt I hav sent you a Lock of my Darkest of my hair. Margaret Catchpole [Catch 2] Other criteria are the sending of greetings and blessings. Typical adjec- tives for the concluding formulas in personal letters are loving and affec-tionate. In contrast to this are the qualities of the closing sentences in formal letters. They tend to be longish and always form a complete sentence: With every feeling of sympathy for your loss, I am, dear Sir, your faithful and obedient servant, Peter Knox, Assistant Surgeon, 1st Btn Pioneers. [Camp1] It is significant that the writer of the above letter states his official position here. By this he justifies why he has written the letter. There are no greetings and blessings to be found and typical adjectives are humble, faithful and obedient. 2 An investigation of various morphosyntactic features in the Corpus Determiners are a category that is worthwhile studying in the letters mainly because there is such a high number of determiners omitted in places where Standard English would place them. This phenomenon requires some explanation. In the following, a few examples will be presented that express some general quality of a person and where an indefinite article would be placed in Standard English. This seems to be categorical for some speakers when the noun phrase is preceded by an ‘as of quality’: I see by yours that you have taken a situation as [a] landsteward. [Iri 108b] Similar to the above, articles can be omitted where abstract general or specific qualities are seen as more important than the fact that they correspond to real entities. The land is too hilly and I am too far from [the] market [.] [Iri 108b] The paying it is not so heavy [.] too far from [the] railway or good market I mean. [Iri 135] The next sentences show that the placing of an article or another determiner is not obligatory where the specificity or the generality of the noun phrase is obvious: [.] the Consequence is that [the] Govt. at last took notice of it. [Reib 15; Government never takes an article when the colonial government, i.e. the obvious one, is talked about] [.] as also my cousins [the] McCormacks [.]. [Iri 12b] It is also possible for a determiner to be omitted when there are two noun phrases the first of which has already been modified by some determiner and where this modification is regarded as applying to the second noun phrase as well: The moment she sees anyone or [a] dog she runs into holes [.]. [Iri 47c] [.] but the rise in price won’t make up for [the] falling off in yield. [Iri 202] Letters from Early Australia – Linguistic Variation and Change Indefinite articles can also be omitted when it is obvious that there is After i arrived i took [a] hackney coach [.]. [Diary] I had taken [a] Coach and drove off for his House [.]. [Diary] Immediate purposes that are apparently strong enough to force a break- up of the ‘normal’ sentence patterns are the emphatic fronting of items or the adherence to a theme-rheme structure, which means that phrases that contain an already known information are fronted. Examples of this are extremely rare in Standard English but relatively frequent in the corpus (as they are in English dialects): Mr P. Broadfoot I have not seen by reason of my being here when he [.]. [emphasis; Reib 9] Your Communication with Sir Thomas [.] I am exceedingly obliged for. [emphasis; Reib 15] Hunting the kangaroo with dogs I never cared for. [theme-rheme; Iri 118] The fronting of whole phrases for reasons of emphasis is a well-known though hardly ever used pattern in Standard EngE. The fronting of phrases to express a theme-rheme structure in the sentence is less well known and even less frequent in this variety. Both have in common that whole phrases are fronted without a change in the structure of the fronted phrase. In the corpus there are examples of fronting that occur within phrases and thus the items fronted receive special marking. Again, the function of this fronting can be either emphasis or the adherence to a theme-rheme structure. It was a struggle to keep money enough together to pay for them [.]. [emphasis; Iri 202] I have to again recommend you to try the leghorn fowls. [theme-rheme; Iri 205] [.] as I intend someday going squatting [.]. [emphasis; Iri 209] 3 Modal verbs in the Corpus and present-day AE, EngE and AmE This particular investigation relies on the following assumptions: (1) The language of the Corpus can be profitably compared with current AE usage since the former is a precursor of the latter. (2) Present-day AE can be profitably compared with present-day EngE and AmE since they represent different varieties of today's English. (3) Since no 19th data of EngE and AmE modal usage were readily available and since this study focuses on past and present AE, data of present day EngE and AmE modal usage were adduced for comparison with data from the Corpus. The possibility that the use of modals in the latter two varieties may have changed in the last 150 years remains, however, disregarded. This drawback is counterbalanced by the fact that both varieties are much older than AE and thus were less likely to change extensively in their accepted standards. Present-day AE usage has been extensively investigated in the works of Collins (1978, 1988, 1991a,b). He also provides the necessary comparative material for present-day EngE and AmE. Collins (1988) presents the use of can, could, may and might in a corpus of spoken and written Australian English that comprises altogether 225,000 words. This he compares with findings in the linguistic literature on the use of these modals in EngE and AmE. The semantic distinctions he makes are those between root modality (Ability (A), Permission (PE), Root Possibility (RP)) and epistemic modality (Epistemic Possibility (EP)). Table 1: Raw frequencies of can, could, may and might Table 2 shows the different uses of can as a modal. Significant6 differences for PE can be found between AE and the Corpus as well as 5 The data for AE, EngE and AmE are always Collins’s. The corpus data are my own. 6 The terms ‘significant difference’ and ‘difference’ will be applied to those differences that have been statistically tested and whose level of confidence is at least below .05. Letters from Early Australia – Linguistic Variation and Change between AE and EngE and AmE usage. Significant differences are also to be found between AE and AmE and the Corpus and EngE and AmE for RP meaning. Finally, usage differs significantly for A meaning between the Corpus and AE, EngE and AmE. Table 2: Can: meanings in the Corpus and in current AE, EngE and AmE Can and may Collins states to have very little semantic overlap due to contextual constrictions. He postulates that the greatest difference between present-day AE and present-day EngE and AmE lies in the relative infrequency of the PE meaning for may, which corresponds to the preferred PE use for the modal can in AE. Statistically significant differences can be found between AE and AmE for PE meaning, between AmE and EngE and the Corpus for RP meaning and between AmE and AE and EngE for EP meaning (cf. Table 3). Table 3: May: meanings in the Corpus and in current AE, EngE and AmE Collins observes that the modal could can be used both as a marker of past time and as a marker of hypotheticality. Many differences can be found for past RP and for A meaning. For the first, AE differs significantly from the corpus and from AmE. With the latter, the Corpus differs from EngE and AmE usage. Hypthetical RP shows AmE to differ from all other varieties. In hypothetical A, the corpus differs from Ae and EngE. Here AmE differs from EngE. Hypothetical EP shows the Corpus to be significantly different from all other varities (cf. Table 4). 7 The figures given are percentages to allow better comparability. The abbreviation Table 4: Could: meanings in the Corpus and in current AE, EngE and AmE The modal might is, like may, primarily used to express EP, i.e. Past EP, Hypothetical EP and Present EP. Collins suggests that might is superseding may as the prime modal used to express epistemic meaning and that, accordingly, its uses as root meaning become less frequent. This change he sees well advanced in AE. Past RP evidences significant differences between AmE and all other varieties. The same applies to Past EP and Hypothetical EP for the Corpus. Usage differs most for Hypothetical RP, EP and PEP. The first shows differences between the Corpus and AE and EngE, between AE and AmE and between EngE and AmE. The second evidences differences between the Corpus and all other varieties as well as differences between AE and EngE and AmE. Finally with PEP, the Corpus differs from all other varieties and AE differs from EngE and AmE (cf. Table 5). Table 5: Might: meanings in the Corpus and in current AE, EngE and AmE Collins draws the following conclusions about the state of modal verbs in present-day AE: (1) The expression of Epistemic Possibility is undergoing a sweeping change in AE with might becoming the main modal used. May is restricted in this use to formal contexts. Could has developed into a third alternative. (2) May can be used as a hypothetical epistemic modal and as a past epistemic modal in AE. Neither use is recorded in the studies of EngE or AmE that Collins cites. 8 The abbreviation ‘PEP’ reads Present Epistemic Possibility. Letters from Early Australia – Linguistic Variation and Change Two trends in the development from language-use in the corpus to present-day AE can be discerned. First, there is an increased use of can and could for PE which corresponds to a decreased use of may and might in this function. Second, the use of epistemic could can be considered an early ‘Australianism’. For the investigation of the modals of necessity and obligation, Collins (1991a) used the same corpus of AE as described above. Each modal verb can express an Epistemic Meaning (EM) or a Root Meaning (RM). The first includes notions of Certainty and Epistemic Necessity and the latter notions of Obligation and Compulsion. Collins states that must, have (got) to and should are the only modals that can reasonably be studied, since the others are to infrequent to allow a fruitful comparison (cf. Table 6). Table 6: Raw frequencies of must, should, ought, need, have to and have got to The frequency of RM meaning for must differs strikingly between the Australian data (Corpus data and Collins's data) and the non-Australian data (current EngE and AmE) (cf. Table 7). It seems likely that the emphasis placed on an egalitarian approach in Australia is responsible for the low frequencies of meanings like ‘Obligation’ or ‘Compulsion’. There is also evidence for an epistemic use of mustn’t corresponding to can’t in Standard EngE. This feature is also evident in Irish English.9 EM meaning, again, shows EngE and AmE usage to be significantly different from usage in AE and the Corpus. Table 7: must: meanings in the Corpus and in current AE, EngE and AmE The use of the modal should is apparently very different in the varieties looked at. Significant is the difference between the Corpus and all other varieties for RM meaning. EngE also differs here from AE and AmE. Almost the same picture can be derived from the usage of EM. The corpus differs from all other varieties. AE differs from EngE and AmE and EngE and AmE differ from each other. Finally, the use of the quasi-subjunctive is significantly low in the Corpus and current AE compared with EngE and AmE usage (cf. Table 8). Table 8: should: meanings in AE, EngE and AmE Collins observes that have got to, unlike have to, is modal-like in its formal properties. He further claims that it is semantically indistinguishable from must. Collins, contrary to the study of EngE he quotes, finds that root have got to can be performative and that it can have a habitual sense. It is realized sometimes with be rather than have and in one example with and rather than to. Since there are no figures available for EngE and AmE, the comparison must be restricted to the Corpus and Collins's data. Here we find that have (got) to was significantly not as restricted to RM meaning in the Corpus as it is in present-day AE. The EM meaning of AE have to then seems a remnant of an earlier, more frequent use. The rise of the modal-like construction have got to from a very low frequency in the language of the letters to a very high frequency in today’s AE is an intriguing observation. This ‘intrusion’ represents a remarkable success story, especially since there were already other modals like must and should that also represented RM. It can be assumed that the rise of have got to is related to the rise of have to because of their phenotypical similarities (cf. Table 9). Letters from Early Australia – Linguistic Variation and Change Table 9: have (got) to: meanings in the Corpus and in Collins (1991a) All the modals of obligation discussed exhibit differentiation into RM and EM. Must is the only modal that strongly favours EM, i.e. it is the primary modal for the expression of Epistemic Necessity. Have (got) to, on the other hand, is the primary modal for the expression of Root Obligation. Should, which also strongly favours RM, is differentiated from have (got) to in terms of subjectivity, i.e. it is used in sentences where the speaker expresses advice, whereas have (got) to expresses an obligation binding on the speaker.10 Overall, a look at the modals of Necessity and Obligation shows two interesting trends. First there is the rise of the construction have got to in present-day Australian usage and second there is the unexplained fact that should changed its role from a modal that in most instances expressed an epistemic meaning to a modal that primarily denotes root meaning. Evaluating differences in modal usage in the Corpus, present-day AE, EngE and AmE Table 10 lists and measures differences in the use of modals in the abovemention varieties. The first column lists the modals in their respective semantics. The modal construction have (got) to was omitted since no data were available for EngE and AmE. The following columns compare the varieties discussed with each other, e.g. comparing modal usage in the Corpus with those in present-day AE. The last column shows the total number of differences for a particular modal meaning. The numbers in the cells represent different levels of confidence for the differences in the usage of a particular modal meaning. If it was <.05 the 'difference factor' 5 was chosen. <.02 gave 10, <.01 15 and <.001 20. This was meant to represent the fact that a higher level of confidence corresponds to a greater difference in usage. Table 10: Differences in the use of modals in the Corpus, current AE, EngE and AmE Corpus/ Corpus/ Corpus/
Most interesting are the last row and the last column of Table 10. The last row reveals the total difference between the use of all modals in all varieties investigated, or, in other words, it reveals how much the use of modals in two varieties is related. It is apparent that the language of the Corpus is most closely related to present-day AE, less related to present-day EngE and least to present-day AmE. This result seems like a foregone conclusion but it is much more than that. It undeniably proves that the language in the corpus is indeed a reliable sample of early AE and that the development of a distinct AE variety, at least as regards modality, was influenced much more by antipodean factors than by external influence from either 'Merry Old England' or from the USA. Letters from Early Australia – Linguistic Variation and Change Nevertheless, the number of differences between the Corpus and current AE is very high, especially in comparison with the number of differences between present-day AE and EngE or present-day EngE and AmE. This suggests that the letter writers used systems that had to change to a great extent in order to achieve today’s AE standard. In the last column of Table 10 the total number of differences for a particular modal meaning is listed. This yields an insight into what modals were most likely to change and in what meanings. Information of this kind is extremely helpful in deciding as to which modals can be profitably studied in future investigations. 4 The letters between orality and literacy This chapter discusses the question to what extent the writers of the letters were aware of the standards of written language and what standards they used when they wrote their letters The encoding and decoding of an utterance can be assisted by extra- linguistic means in spoken language. This, in effect, increases the amount of acquired co-textual knowledge and hence enables the encoder to be less precise. Since this is not possible in written language, more effort is required to encode an utterance clearly and without ambiguity. Written communication also requires verbalization of all information to be conveyed. This means that written language has to place more linguistic cues, like pro-forms, existential markers, punctuation markers etc., and that it has to use signs and structures that are less ambiguous. The most important difference between spoken and written language is that the former is variable in its use of such linguistic cues depending on the amount of shared previous and situational knowledge. This is not the case in written language. Here the placing of the maximum amount of linguistic cues and the use of unambiguous signs is obligatory, because the amount of shared knowledge, irrespective of its actual amount, is assumed to be minimal. This is due to historical reasons.11 The majority of the letter writers in the corpus evidence a use of language that reveals that they were unaware of the traditional conventions of written language. They stem, for the most part, from the lower orders of 19th century British society and therefore had experienced little schooling. Some had only very recently learned to read and write and thus had not been exposed to the standards of written language to a greater extent. This explains why the writers of the letters and diaries in the corpus imported many familiar structures from their use of spoken language into the unfamiliar written mode of communication. There are three distinct qualities apparent in the letters that show the language used to be closer to the standards of spoken language than to the standards of written language. The first is the use of direct addresses and tag questions, the second the use of unembedded and context-dependent structures and the third the omission of linguistic cues that were considered to be redundant. The fact that for many of the letter writers the norms of spoken communication were considered to be very close to the norms of written communication is evidenced by many direct addresses and tag questions in the letters. The use of these show that writing was considered to be more like a process than a product, thus marking the breakdown of one of the major barriers distinguishing spoken and written language. Typical examples of the use of direct address are now presented: [.] you will send me sum of Lucesy and Charles, hear. [Catch 4] [.], but I write in love to you, Alick my own. [Iri 77] The use of tag questions often presupposes that an immediate reaction to a statement is expected. Since this is only possible in spoken communication, the use of tag questions in the letters again reveals that for many writers these two modes of communication were considered to follow a single standard and not different standards. Typical instances of this phenomenon are: [.] rather absurd is it not! [Reib 23] [.] but I love you very tenderly, you know this don’t you [.]. [Iri 77] He has been on the late shift this week [.] pretty late isn’t it? [Iri 163b] The standards of written language strictly require an author to use context-independent and embedded structures which allow the unambiguous Letters from Early Australia – Linguistic Variation and Change decoding of a statement. Many letters in the corpus however show the use of unembedded and context-dependent structures. This is not due to some lack in the writers’ linguistic competence but reflects the reasoning that these structures are not ambiguous when the amount of shared knowledge is taken into account. [.] I was bled and blistered and both me and Eliza so ill [.]. [Diary] [.] if you were in this colony we were never happier at home. [Iri 10] Another thing father did very wrong by sending David to Australia without giving him [.]. [Iri 189] It is a particular quality of the written mode of communication that every piece of information, when referred back to, has to be stated again or at least has to be represented by a pro-form. Likewise, it is obligatory to use aspect markers and auxiliaries to express particular meanings. These explicit linguistic cues make it easier for a reader to decode an utterance. The question where such cues are to be set and what they have to be like is rigidly prescribed in the standard form of written language. The omission of linguistic cues not considered to contribute to the comprehensibility of a statement is a very powerful principle in the language of the letters. Examples of omitted aspect markers/auxiliaries [.] you [may] know long ere this that Eliza Foster was married to a Mr Pitman. [Reib 15] I have been in places where we [have] never seen a newspaper from one Christmas [.]. [Iri 58c] I told you in my last [letter], of them all comeing up on a visit to Sydney. [Reib 12] [.] there is another [possibility] you can place yourself on a piece of Government Land. [Iri 12b] i did not Lik that so i [went] to nurs one Mrs. Skinner. [Catch 2] [.] and lodgers would soon [eat] her out of house and home. [Iri 178b] The attendance [was] very good. [Iri 22] I would esteem it as a very great obligation done [to] me. [Piper 5] [.] my family here are all well [in] health. [Reib 10] [.] and would marrey me if [I] Lik But i am not for marring. [Catch 1] [.] & never can bear to look over a Letter after having written it in order to Correct [it], [.]. [Reib 9] [.]; [he] has got nothing to do but walk up and down [.]. [Iri 157a] V Summary and Conclusions
This study of Australian English in general and a letter corpus of nineteenth century Australian letters in particular was divided into three main parts. First, the historical development of the Australian variety of English was presented. Various theories regarding the development of new dialects and the origins of a typical Australian pronunciation and lexis were shown not be be powerful enough to explain the process of the formation of AE in a principled and unified way. The possible origins of Australian English were sketched taking into account linguistic and social factors. Following that, the letter corpus used in the present study was presented and relevant sociological and other data were discussed. The letter writers were shown to be mostly Irish male first generation immigrants. The evaluation of stylistic and linguistic features found in the corpus made up the third part. A detailed comparison of the modal systems of present-day AE, EngE and AmE and the language of the letters was given. It was shown that modal use in the corpus was most similar to AE, less similar to EngE and least similar to AmE. In the last part, questions of orality and literacy were looked at. It was contended that many letter writers did not differentiate between standards for spoken and written language. Questions of shared knowledge and comprehensibility were shown to be the relevant factors in the textualization of information. This was used to explain the many examples of context-dependent and unembedded structures and the many instances of absent linguistic cues. Letters from Early Australia – Linguistic Variation and Change Baker, Sidney J. 1966². The Australian Language: An Examination of the English Language and English Speech as Used in Australia, from Convict Days to the Present. Sydney: Angus and Robertson. Bernard, John R.L. 1969. "On the Uniformity of Australian English." Orbis 18: Collins, Peter. 1978. "Dare and Need in Australian English: A Study of Divided Usage." English Studies 59: 434-41. –––. 1988. "The Semantics of some Modals in Contemporary Australian English." Australian Journal of Linguistics 8: 233-58. –––. 1991a. "The Modals of Obligation and Necessity in Australian English." In: Karin Aijmer and Bengt Altenberg, eds. 1991. English Corpus Linguistics: Studies in Honour of Jan Svartvik. London: Longman, 145-65. –––. 1991b. "Will and shall in Australian English." In: Stig Johansson and Anna- Brita Stentström, eds. 1991. English Computer Corpora. Berlin: Mouton de Gruyter, 181-99. Dabke, Roswitha. 1976. Morphology of Australian English. München: Wilhelm Fritz, Clemens. 1996. Early Australian Letters: A Linguistic Analysis. Görlach, Manfred. 1991. "Australian English: Standards, Stigmata, Stereotypes and Statistics." In: Manfred Görlach, ed. 1991. Englishes: Studies in varieties of English 1984-1988. Amsterdam: Benjamins, 144-73. Gunn, John S. 1972. "Early Australian Pronunciation". Address given at the Australian Universities Literature and Language Association Congress in Dunedin. Halliday, Michael A.K. 1985. Spoken and Written Language. Burwood (Victoria): Hammarström, Göran. 1980. Australian English: its Origin and Status. Hamburg: Horvath, Barbara. 1985. Variation in Australian English: The Sociolects of Sydney. Cambridge: Cambridge University Press. Irvine, Nance, ed. 1992. Dear Cousin: The Reibey Letters. Sydney: Janet Press. Leitner, Gerhard. 1984. "Australian English or English in Australia: Linguistic Identity and Dependence in Australian Broadcast Media." English World Wide 5, 1: 55-85. –––. 1989. "‘You’re not supposed to say nothing to nobody’ – about Australian English Grammar." In: Dieter Riemenschneider, ed. 1989. Critical Approaches to the New English Literatures and Language Studies. Essen: Die Blaue Eule, 142-54. –––. 1990. "Ist das australische Englisch wirklich wie das Londoner Cockney?" In: Gerhard Stilz and Heinrich Lamping, eds. 1990. Australienstudien in Deutschland. Frankfurt (Main): Peter Lang. Mitchell, A.G. 1946. The Pronunciation of English in Australia. Sydney: Angus –––, and Arthur Delbridge. 1965a. The Pronunciation of English in Australia. –––, and –––. 1965b. The Speech of Australian Adolescents. Sydney: Angus and Newbrook, Mark. 1992. "Unrecognized Grammatical and Semantic Features Typical of Australian English." English World Wide 13: 1-32. O'Farrell, Patrick. 1989. Letters from Irish Australia 1825-1929. Sydney: New

Source: http://www.clemens-fritz.de/publications/DiG-Early_Letters.pdf