“Ik heef niets gezeg”: the past tense in an Anglo-Boer War concentration camp diary
-
Johanita Kirsten
Abstract
This paper presents a case study of the writing of the young woman Rensche van der Walt in her personal diary during the Anglo-Boer War, specifically 1901 to 1902. A corpus of about 24,000 words was compiled from the diary, detailing the personal experiences of Van der Walt in concentration camps during the war. This study focuses on her use of the past tense to determine the extent to which it aligns with modern Afrikaans norms rather than Dutch norms, and it is contextualized against a control corpus of letters written during the same war. The results show that Van der Walt uses the Dutch preterit to a limited extent, while the modern Afrikaans past tense (based on the Dutch perfect) is the preferred choice. This aligns with the usage in the letter corpus. However, the past participles used in the diary are more regularized, with much fewer inflectional markings than in the letters, indicating that Van der Walt is slightly ahead of her time regarding this development – it is not unexpected, as young women are known to lead the adoption of new variants during periods of language change. The findings also shed light on the extent of variation still present in the past tense use of semi-literate writers of early Afrikaans, a topic which has not received much attention in corpus-based studies yet.
1 Introduction
The Afrikaans language developed from an extra-territorial variety of Dutch used in southern Africa from the seventeenth century until the twentieth century. Changes from Dutch were influenced by a number of factors, especially contact with the indigenous Khoekhoen, European travelers and settlers, and slaves from elsewhere in Africa and the East. Even though it was originally believed that the major changes took place before the nineteenth century, later studies revealed a dialect continuum with significant variation to exist well into the twentieth century, with some developments still underway (Deumert 2004; see also Kirsten 2019).
The establishment of Afrikaans as a language separate from Dutch was initiated in the late nineteenth century (Deumert 2004: 287), but these attempts were arrested by the outbreak of the Anglo-Boer War at the turn of the twentieth century. As is often the case during times of war when loved ones are separated, many letters were written during this time, and these were often preserved by families and/or donated to museums and archives in the course of the twentieth century. The same is true for diaries, such as the one by a concentration camp nurse Sarie Roos (reported on in Pretorius and Krige 2011), as well as a camp inhabitant Rensche van der Walt (see Section 4.1.1 for more information).
Letters and diaries from this context provide invaluable linguistic data, as the population in general did not have strong ties to the Netherlands and Dutch anymore, but Afrikaans had not been recognized or standardized by then (and often called Cape Dutch or ‘Hollands’ at the time). Furthermore, the writers of these texts were mostly from the Free State and Transvaal, rather than the Cape Colony, as is the case with the Corpus of Cape Dutch Correspondence (Deumert 2004).
An aspect of the grammar that changed quite drastically from Dutch to Afrikaans is the verbal system, with significant regularization and simplification. For this reason, focusing on verbs is a useful way to investigate language variation and how far language change has proceeded at a specific point. This paper will present a case study focused on the use of the past tense in a war diary compared to a corpus of war letters, to determine how far the development of the Afrikaans verbal system was, and what the extent of variation was at the time of writing.
In Section 2, the sociohistorical context of South Africa and the war will be described. Thereafter, in Section 3, the importance of ego-documents in historical sociolinguistics will be discussed, followed by a description of the methodology and data (Section 4). Section 5 presents the case study of the past tense in the diary, contextualized against the larger letter corpus, followed by a discussion in Section 6 and then the conclusion.
2 Sociohistorical context
At the turn of the 20th century, a large part of what is known as South Africa today was involved in a brutal war between the British Empire and the Boer (settler) republics – the South African Republic (colloquially known as Transvaal) and the Republic of the Orange Free State. In England, the war is known as the South African war, but in South Africa it is known as the Anglo-Boer War or the Second War for Freedom. At the center of the war was gold mining on the Witwatersrand (Worden 2012: 30), part of the Transvaal, to which the British wanted more direct access, being willing to go to great lengths to obtain it (Giliomee and Mbenga 2007: 209). The president of the Transvaal, Paul Kruger, declared war on the British, with the support of the Orange Free State as allies, in October 1899 (Ross 2009: 78).
The British and the Boers hoped that the war would not last long, but it lasted for almost three years and extended over a wide area (Worden 2012: 34). In an attempt to break the spirit of the Boer guerilla commandos later in the war, the British followed a two-pronged strategy: firstly, the scorched earth tactics, where farms were burnt down and livestock was taken or slaughtered; secondly, the concentration camps where the families and servants of the farmers were imprisoned (Jewell 2003: 1; Worden 2012: 34). While these tactics initially strengthened the resolve of the Boer commandos, they were eventually forced to surrender, and signed the Treaty of Vereeniging in May 1902 (Worden 2012: 34).
In the concentration camps, the Boer women, children, sick and elderly lived in tents or reed huts, and later, in some camps, in dormitory structures of corrugated iron. The camps were supposed to be managed according to military guidelines, but with smaller rations, and inhabitants were provided with little clothing and medical treatment (Jewell 2003: 2; Pretorius and Krige 2011: 73). Camps were generally under-resourced regarding, for instance, tents, food, wood and water, and their staff struggled often, as they were not well equipped in terms of training or experience (Pretorius and Krige 2011: 75).
The Bethulie camp, which is of particular interest for this article, was one of the camps with the worst circumstances and living conditions from the start (Pretorius and Krige 2011: 75–76). The town of Bethulie is located in the southern Free State province, close to the Orange River which forms the border with the Eastern Cape, about 50 km east of the Gariep Dam. The Bethulie camp, one of the biggest concentration camps during the war, was adjacent to the town of Bethulie. Later in the war, in January 1902, with a new camp manager, the conditions in the camp improved somewhat.
The diary used as primary data source in this article was written mostly in the Bethulie camp, and the letter corpus consists of letters written by many different people in different roles and places during the course of and shortly after the war.
3 Ego-documents in historical sociolinguistics
In recent years, special attention has been paid to studying ego-documents in historical sociolinguistics, specifically with a perspective of ‘language history from below’ (Elspaß 2012: 156; Knüsli 2019: 2). The relevant ego-documents would be especially letters and diaries, and increasingly from the lower social classes (Elspaß 2012: 161; Knüsli 2019: 2) – these sources of linguistic data did not traditionally receive the attention they deserve (Elspaß 2012: 161; Litty 2019: 2).
One of the advantages of using this kind of document as data, is that it moves away from the standard language perspective, investigating language use that is much closer to the spoken varieties of the time (Brown 2019: 3) – it is, in fact, “as close to speech as non-fictional historical texts can possibly be” (Elspaß 2012: 156). This is especially true when using texts written by semi-literate writers – “private letters and diaries written by semi-literate writers seem to have the highest potential to render authentic sources of historical orality” (Elspaß 2012: 159).
There are differences between letters and diaries, however, and it is important to take this into account. Private letters are dialogical, entailing a ‘social practice’ between correspondents (Knüsli 2019: 5). Diaries, on the other hand, especially private or personal diaries, are monologic by nature, so while they may also be informal and unplanned (like letters), they are usually less ‘oral’ (Elspaß 2012: 162). In fact, personal diaries are often more formal and standard-like than personal letters in various contexts and languages (Rutten and Krogull 2021: 312). The monologicity of diaries are in principle closer to the language of distance rather than the language of immediacy, and authors may have an undefined future audience in mind when writing their diaries. These considerations can contribute to the relative formality of diary writing compared to that of private letters (Rutten and Krogull 2021).
In this paper, the primary data source is a private diary, with a letter corpus against which it can be contextualized. While the type of language use in these texts should not differ too much, it is important to keep possible differences in mind when interpreting the data. In particular, the language use in the diary may be more formal than that of the letter corpus.
4 Methodology and data
The approach of this study is empirical, i.e., reporting on the occurrences of a particular variable. It follows a research design that is similar to that of Ávila-Ledesma and Amadro-Moreno (2023) – a case study design with a mixed methods approach, based on corpus linguistics in a historical sociolinguistic framework, including both quantitative and qualitative analyses. It focuses on the case study of Rensche van der Walt’s Anglo-Boer War diary, and particularly the past tense, against the background of a corpus of Anglo-Boer War letters. In this way, the study uses a combined methodology to investigate a topic that would typically be viewed from a qualitative perspective, again similar to Ávila-Ledesma and Amadro-Moreno (2023: 248).
The quantitative analyses were done with the help of the corpus software WorthSmith Tools (Scott 2022) to extract the data for analysis, which was then further refined in Excel. For example, all instances of het in the diary were extracted from the corpus and exported to Excel. In a first step, all its uses as a definite article (‘the’) were removed. When only verbal uses remained, the concordances were classified according to whether het is used as an auxiliary or as a lexical verb. All the auxiliary instances were then coded for certain further variables. The same process was followed for all the other variants of het, although the other variants were typically not used as an article as well.
The Diary Corpus, the primary dataset for this article, will be described below, followed by a description of the Anglo-Boer War Letter Corpus.
4.1 Corpora
4.1.1 Diary
The author of the diary is Rensche van der Walt, born on 29 June 1878 in the town of Reddersburg, located in the Republic of the Orange Free State at the time. She is the middle of nine siblings, all of whom attended school as pupils of the Reformed Church School in Reddersburg (Hamman 1965: ix). It is unclear what level of schooling Van der Walt (or her siblings) attained, but the norm at the time for small towns and farming communities was attending primary school. Van der Walt never married or had children, but remained involved in the extended family and especially with her nieces and nephews (Hamman 1965: ix).
Van der Walt wrote the diary during the Anglo-Boer War, detailing her experiences with the British troops and the concentration camps, especially Bethulie. It ranges from the family’s first encounter of British troops on the farm, to several months after the war when the whole family was together again. She held on to the diary until her death, after which it went to her sister. Her niece, Kezia Hamman, transcribed her diary and translated it to modern Afrikaans, and published both versions together in a book in 1965. After this, the original was donated to the War Museum in Bloemfontein (Hamman 1965: x). The publication was not well known, and did not receive much public attention.
The existence of the diary came to my attention through the Afrikaans Language Board, who is currently working with the War Museum in Bloemfontein (South Africa) to re-publish the diary. One of the board members generously lent me their personal copy of the original publication, which I used to manually type the original version electronically as a corpus. It was then saved in txt-format for use with corpus software.[1]
The corpus consists of 24,265 tokens, with 2,829 types. It is quite small, even for a historical corpus, which is why a mixed method approach is taken.
4.1.2 ABW letter corpus
The Anglo-Boer War (ABW) Letter Corpus is a compilation of 260 letters written by 129 different people during and shortly after the war, and it consists of 105,922 tokens and 9,814 types.
The letters were written between friends and family members in contexts ranging from prison and concentration camps to commandos in the field and people at home (Nel 2016: 120). All the letters were written by citizens of the Boer republics and the Cape colony, in Cape Dutch / Afrikaans (Nel 2016: 120). The authors were mostly elementary schooled citizens, showing limited literacy, but still being able to communicate effectively in the letters (Nel 2016: 120). While these people weren’t strangers to letter writing, the war’s catastrophic circumstances caused a great deal of letter writing, much more than usual, because friends, family and loved ones were often separated for long periods of time (Nel 2016: 124). At the time, there no written standard for Afrikaans had been established, and it was often called ‘Hollands’; the limited literacy the people had was usually learnt through Dutch.
Copies of the letters were obtained from the archives of the War Museum and National Literary Museum in Bloemfontein, as well as the National Archives in Pretoria (Nel 2016: 121). When one particular author wrote a large number of letters, not all of these were included in the corpus to remain balanced between different authors (Nel 2016: 120). The letters were then transcribed electronically, retaining the language use and content of the originals as far as possible (Nel 2016: 121).
The corpus was compiled by Nel (2016) for her MA dissertation, and generously made available to me. The MA dissertation focuses on the use of health formulas in the letters, as an instrument used by semi-literate writers. The corpus was in Word format when I received it, so I proceeded to remove all formatting as far as possible without losing any content, remove the addresses, and include meta-data in the form of headings above each letter. The letters were then copied into a txt-file to enable corpus analyses.
Unfortunately, not much information about the authors of the letters is available. The names of the authors are included in most letters, and from this and how authors signed off, the distinction between male and female authors can often be made, but not always. About 40 % of the corpus could be confirmed as written by women, which means the corpus is not too unbalanced for such a specific historical corpus. Other information such as their age, where they are from or their education could not be determined with any certainty. This unfortunately does not allow for the corpus data and analyses to be interpreted according to any of these variables, but only as a whole.
5 The case study
5.1 General characteristics of the diary
In this diary, written before the formal standardization of Afrikaans, there is variation, as one would expect – some older forms that are not used in Afrikaans anymore, and some newer forms that are quite different from Dutch.
Some spellings are used consistently throughout the diary, although they are quite different from the standard forms accepted later. The spelling of diphthongs and long vowels are particularly interesting: modern Afrikaans ou [œu] is written oi, as in noi for nou ‘now’; and uu [y] is spelled ui, as in uir for uur ‘hour’. Further examples are eu for ei [əi], so the fifth month is spelled Meu ‘May’, and oiw for ooi [oi], as in goiw for gooi ‘throw’ (see noi and hoit in 1, and uir and noiwe in 2 below).
In modern Afrikaans, the schwa vowel is written in different ways, depending on the context; most often it is written with either i as in vir ‘for’, or with e as in verby ‘over’. In the diary, e is used much more frequently than i, where the frequently used dit ‘it’ is mostly written as det, niggie ‘female cousin’ as negie, fris ‘sturdy’ as vres, and many more. It is not unusual for the time, when vir was often written as ver, and dit as det (see ver in 2).
The example of vres further illustrates the use of v in most positions, reserving mostly the end position for f in the diary; in modern Afrikaans, f is used much more widely. More examples from the diary include tavelkie for tafeltjie ‘table.dim’, vemielje for familie ‘family’, and vliks for fluks ‘hardworking’ (see vlyt in 2).
Another spelling phenomenon is actually indicative of a phonological process – rounded vowels are often written as their non-rounded equivalents (for example ee [e] for eu [ø]), and non-rounded vowels are written as their round equivalents (for example u for i [i]). The third person plural pronoun hulle is consistently written as hele, and the second person plural julle as jele. Since the orthography had not been standardized by hen, and the author probably did not have a high level of education, the spelling appears to be quite close to how words were pronounced. This phenomenon then suggests that the unrounding of round vowels, a common phonological process in Afrikaans today, already started at that time. Because the distinction between the pronunciation of round and non-round vowels were not clear at the time, certain round vowels are written as non-round, and some hypercorrection occurs with non-round vowels being written as their rounded equivalents (see hele in 1 and 2, and vlyt and deer in 2).
Several further phonological phenomena can be seen in the spelling used in the diary:
devoicing of word-final consonants, such as wort for word [vɔrt] ‘become’, but also hard for hart [ɦɑrt] ‘heart’ (see in 1: hart, gewort, paart and gezont) – the voicing re-emerges with any suffixes added to the word, such as wording [vɔrdəŋ] (‘becoming’ as noun);
assimilation, for example voggende [fɔxəndə] for volgende [fɔlxəndə] ‘following’ and kenners [kənərs] for kinders [kəndərs] ‘children’;
lowering of vowels, in baja [bɑjɑ] for baie [bɑjə] ‘much / many’ (see schrikraag in 1);
develarization, such as laans [lans] for langs [lɑŋs] ‘beside’.
| De 2 Maart op een Zaterdag a v ond gaan loop vrouwens en kenders h oi t haal in de berg dit was een paar weeken vry om te gaan haal. Maar de 2 Maart heef daar achter de berg iets wonderlyk har t geklap en h e le heef schrikr aa g gewor t en daar jaag de polliets Kloete met de p aa r t n oi de vrouwens en kinders en nog paar Camp mans die daar by was gaan vang en die ou Klaas was toen we d er gezon t . |
| ‘The 2nd March on a Saturday evening women and children walked to fetch wood in the mountain it was free to go fetch for a few weeks. But the 2nd March something banged loudly behind the mountain and they became afraid and there the police Kloete was chasing with the horse the women and children and a few Camp men who were there to catch them and old Klaas then got well again.’ |
Many of the intervocalic consonants which have since disappeared from Afrikaans are still present in the diary, though, such as dagen for dae ‘days’, morgen for môre ‘morning’, over for oor ‘over’, and blyven for bly ‘stay’ (see also avond and weder in 1).
Also regarding grammar, there is a mixture of older, more Dutch forms and newer Afrikaans forms. The brace negation was used in some texts during that time, but not in the diary. The preterit form mog (spelled moog) of the modal verb mag ‘may’ is still used, as well as the reflexive pronoun zich, both of which are absent in modern Afrikaans.
The genitive also comes in both newer and older forms. The Dutch genitive is still used, but almost exclusively in religious contexts: “de goedheid des lieve Heere” (‘the goodness of the loving Lord’), and “de wil dus Heeren” and “de wel dis Heeren” (‘the will of the Lord’). A few instances of the Dutch pronominal genitive occur, as in “Commedant Hertzog zyn vrouw” (‘Commandant Hertzog’s wife’), and this is the form from which modern Afrikaans genitive particle se developed. More often in the diary, the form ze is used as the genitive particle, invariant for gender, person, and number. Some fixed expressions still familiar in modern Afrikaans are present: “Affrikaner ze kunt” (‘Afrikaner’s child’), which is used as an identifier (‘Afrikaner’s child’ simply means ‘an Afrikaner / Afrikaners’).
Another grammatical phenomenon regards the first-person plural pronoun – while the Dutch subject form wy ‘we’ is used occasionally, the object form ons ‘us’ is also used in the subject position (see in 2), as is the practice in modern Afrikaans.
| Dien nach om elf ui r gaan ons toen een groot berg d ee r maar ik kan dit nooit beschryf hoe aardig dit is. Zoo land ons die n oi we party die d ee r die berg gaan v l y t die trein zoo stadig. Dit zyn v e r my onvergeetlyk dit was toen de tondel wat h e le ons van vertel heef. |
| ‘That night at eleven we went through a big mountain but I can never describe it how strange it is. So as we go through the narrow passage the train gave a long whistle. It was unforgettable for me it was the tunnel they told us about.’ |
Articles also show variation in the diary: Dutch het ‘the’ is used to an extent, Afrikaans die ‘the’ is used occasionally (see in 2), and de ‘the’ is used most often, regardless of the noun type or gender.
Finally, the past tense auxiliary also varies. The form used most frequently is heef ‘have’, while hebben and heeft (Dutch) and hefe are also used sporadically, and het (Afrikaans) a bit more frequently. The lexical verbs used in the past tense are also more often the regularized Afrikaans forms, although several verbs with Dutch conjugation do occur.
The remainder of the article will focus specifically on the past tense as it is used in the diary and in the ABW-letter corpus. Before diving into the analyses and presenting the findings, it is pertinent to describe the development of the Afrikaans past tense from Dutch briefly.
5.2 Past tense in Afrikaans from Dutch
In the development of the Afrikaans verbal system from Dutch, inflection was reduced to a minimum, and the order of the verbs became fixed (Conradie 2006: 86). The tense system was also simplified to a great extent: the past perfect disappeared; the preterit was replaced with the present perfect (functioning as past tense), along with a functionally extended historical present, and the use of the adverb toe ‘then’ in the present and past tense (Conradie 1999: 21–22). As mentioned, in the development of the Afrikaans verbal system, significant simplification took place, and this includes the perfect, which became a convenient and regularized replacement for the preterit (Conradie 2006: 91). The perfect simplified in different ways:
the Dutch auxiliaries hebben ‘have’ and zijn ‘be’ with all their different forms were reduced to the forms het and is;
hebben (or het) replaced zijn (or is) as the auxiliary used with mutative verbs;[2]
the verbal past tense was regularized to the base form of the verb and the prefix ge- (optional in some contexts), so the strong and weak past participle distinction and the use of ablaut were also eliminated (Conradie 1981: 275, 1999: 22).
The morphological distinction between the infinitive, person and number forms were eliminated, then, with lexical verbs invariant for person and number, and no separate infinitive form (Scholtz 1958: 63) apart from two specific exceptions (wees ‘be’ as infinitive for the copula is ‘is/are’, and hê ‘have’ as the infinitive for main verb het ‘have’).
The choice of het as replacement for hebben and its variants may seem somewhat unexpected, as it is not one of the usual Dutch variants of hebben. The form het was originally used as the third person singular form in the present tense in a Dutch dialect which found its way into Cape Dutch, where it was generalized to the singular in general and later to the plural as well, becoming the dominant and later only form used in all past tense constructions (Conradie 2006: 89–96).
The description of the past tense in the diary corpus, against the background of the ABW-corpus, will proceed in two parts: first, the use and forms of the past tense auxiliary het and its variants will be investigated, after which attention will be paid to the lexical verbs used in past tense constructions and to what extent Dutch inflection is still present.
5.3 The past tense auxiliary het and its variants
In the diary, het and its variants occur 245 times as past tense auxiliaries. Van der Walt uses a number of different forms (see Table 1) but strongly favors heef, although het is also used more than the other variants.
All variants of het in the diary.
| Variant | Occurrences | Proportion |
|---|---|---|
| hadden | 1 | 0.4 % |
| hebben | 3 | 1.2 % |
| heef | 215 | 87.8 % |
| heeft | 1 | 0.4 % |
| hefe | 1 | 0.4 % |
| het | 24 | 9.8 % |
The past tense hadden ‘had’[3] and perfect heeft are both only used once, the former according to Dutch grammar (3a), the latter not (3b).
| Doch | wy | hadden | dit | niet | verkeert | geraai. |
| ‘Though | we | had | it | not | incorrectly | guessed.’ |
| ‘Though we had not guessed it incorrectly.’ | ||||||
| Maar | ons | kon | dit | niet | anderste | verwag | heeft … |
| but | we | could | it | not | differently | expect | have |
| ‘But we could not have expected it differently…’ | |||||||
The form hebben from the Dutch perfect is used three times, each with plurals, such as (4).
| …toch | hebben | wij | het | reeds | klaar | bemerk… |
| although | have | we | it | already | done | notice |
| ‘… although we have noticed it already…’ | ||||||
The auxiliary form hefe could be an error, either by the author or in the original transcription, as it is not a known variant, but it is difficult to be certain.
The most frequently used variant in the diary is heef (5).
| Hy | heef | gehoor | dat | ons | gekomen | heef. |
| He | have | heard | that | we | came | have |
| ‘He has heard that we have come.’ | ||||||
| Die | heef | hele | niet | in | die | hande | gekryg. |
| this | have | they | not | in | the | hands | got |
| ‘This, they have not gotten their hands on.’ | |||||||
| Ik | heef | niets | gezeg. |
| I | have | nothing | said |
| ‘I said nothing.’ | |||
| … | de | 12de | toen | heef | ons | tent | omgewaai… |
| the | 12th | then | have | our | tent | over.blown | |
| ‘… on the 12th, our tent was blown over…’ | |||||||
It is clear that heef is used regardless of person and number. Although Van der Walt does not use the more modern het as frequently, it seems that heef fulfills the same purpose as het in her language use.
While het is not used nearly as frequently as heef in the diary, it is the variant with the second highest occurrence. Similar to heef, het is also used regardless of person and number:
| Toen | moet | ek | omdraai | voor | dat | ik | toen | my |
| then | must | I | around.turn | before | that | I | then | my |
| ze | ge ze | het . | ||||||
| say | said | have | ||||||
| ‘Then I had to turn around before I could say my say.’ | ||||||||
| Maar | ons | het | geen | taal | gehad | te | zeggen. |
| but | we | have | no | language | had | to | say |
| ‘But we had no language to say it.’ | |||||||
| Jele | het | ons | gevang, | zoo | moet | jele | ons | maar |
| you.pl | have | us | caught | so | must | you.pl | us | but |
| goet | verzorg. | |||||||
| good | care | |||||||
| ‘You caught us, so you must take good care of us.’ | ||||||||
| … | want | hy | het | nie | goet | gewerk | met | ons. |
| because | he | have | not | good | worked | with | us | |
| ‘… because he did not work well with us / treat us well.’ | ||||||||
From the examples it becomes clear that the parameters of use of heef and het are similar in Van der Walt’s usage, even though heef is used much more frequently. To determine whether this is the case more generally as well, the use of het and its variants in the ABW-corpus will now be briefly discussed.
In the ABW-corpus, het and its variants occur 1,287 times as auxiliary verbs, taking on many different forms (see Table 2).
Variants of het in the ABW-corpus.
| Auxiliary | Occurrences | Proportion |
|---|---|---|
| had | 32 | 2.5 % |
| hadden | 23 | 1.8 % |
| hadt | 3 | 0.2 % |
| hat | 4 | 0.3 % |
| heb | 300 | 23.3 % |
| hebbe | 1 | 0.1 % |
| hebben | 205 | 15.9 % |
| hebt | 55 | 4.3 % |
| heef | 242 | 18.8 % |
| heeft | 259 | 20.1 % |
| heefte | 1 | 0.1 % |
| hep | 14 | 1.1 % |
| het | 148 | 11.5 % |
The proportion of uses of het is slightly higher than in the diary, just over one percent, which is not significant (log-likelihood value of 2.13, indicating a p-value of more than 0.05).[4] A clearer difference with the diary is that there is not a dominant variant in the same way, as none of the variants come to a quarter of all the auxiliary uses together. There are four fairly popular variants: heb, hebben, heef and heeft, and a few variants only used once or a few times, as well as a group in the middle. What this shows is that the choice of variant had not been adopted in the general speech community yet, even though specific speakers or writers may have had a dominant variant, like Van der Walt using heef.
Similar to the diary, het is used with different persons and numbers:
| Di | Engelse | het | dit | laaste | nag | Verlaat… |
| the | English | have | it | last | night | leave |
| ‘The English left it last night…’ | ||||||
| … | veel | tente | het | om | gewaai… |
| many | tents | have | over | blown | |
| ‘… many tents were blown over…’ | |||||
| … | u | perzend | wat | u | gestuur | het | het |
| you.sg | gift | that | you.sg | sent | have | have | |
| my | zoo | heerlyk | Gesmaak… | ||||
| me | so | delicious | tasted | ||||
| ‘… your present that you have sent me tasted so delicious…’ | |||||||
| O | my | liefste | ik | het | u | zagte | handjes | gemus… |
| oh | my | love | I | have | your.hon | soft | hands.dim | missed |
| ‘Oh my love, I missed your soft litte hands…’ | ||||||||
| Ons | het | mooi | regens | gehad… |
| we | have | beautiful | rains | had |
| ‘We had good rain…’ | ||||
| Hij | het | zij | plig | gedoen… |
| he | have | his | duty | did |
| ‘He did his duty…’ | ||||
It is clear that het was not the dominant variant yet by any means, but that it was already used as freely in terms of context as in modern Afrikaans. Not long after this diary and these letters were written, in the 1910s, het became the overwhelmingly dominant variant as past tense auxiliary while losing its use as definite article (Kirsten 2016: 120). This demonstrates how quickly variation can be reduced by standardization – the first official Afrikaans spelling rule book was published in 1917 (Le Roux et al. 1917), and a grammar was published in the same year (Malherbe 1917).
5.4 Lexical verbs in the past tense
It was already mentioned that most of the Dutch inflection was lost in the formation of the Afrikaans verbal system, with only the prefix ge- added to otherwise invariant verb forms, if at all, in the past tense.
In order to investigate the relevant past tense verb forms in the diary, every lexical verb used with a het-variant was manually coded in the data. For each, it was noted whether any Dutch inflection (that is not present in modern Afrikaans) is present or not. All instances of the same verb were grouped together, including spelling variation that does not change its inflection or meaning, such as the use of spaces or different letters for the same sound.
In the diary, Van der Walt uses 135 different verb forms, and 39 (or 29 %) of these have Dutch (or Dutch-looking) inflectional markings or forms.
| De | Heer | heef | gegeven , | de | Heere | heef | genomen , |
| the | Lord | have | given | the | Lord | have | taken |
| de | naam | dus | Heere | zy | geloof. | ||
| the | name | gen | Lord | be | praised | ||
| ‘The Lord has given, the Lord has taken, the name of the Lord be praised.’ | |||||||
| Die | heef | hele | niet | in | de | hande | gekryg. |
| this | have | they | not | in | the | hands | got |
| ‘This they have not gotten a hold of.’ | |||||||
| Toen | hy | uit | was | hoor | ik | dat | hy | ons | kwaam | tel | heef. |
| when | he | out | was | hear | I | that | he | us | came | count | have |
| ‘When he was out, I heard that he had come to count us.’ | |||||||||||
| Nu | heef | hy | van | zyn | huisgezen | hy | met | een |
| now | have | he | of | his | family | he | with | one |
| dochterje | overgeblyf . | |||||||
| daughter.dim | over.stayed | |||||||
| ‘Now he remained of his family, one he and on little daughter.’ | ||||||||
| Ons | heef | hem | goet | sleggezeg . |
| we | have | him | good | bad.said |
| ‘We have insulted him properly.’ | ||||
In modern Afrikaans, gegeven ‘gave / have given’ in example (8a) would be written as gegee, without the final n; genomen ‘took / have taken’ would be geneem, without ablaut and the word final -en. In (8b), gekryg ‘got / have gotten’ would be gekry in Afrikaans, and in (8c) the auxiliary use of kwaam ‘came’ would be present tense kom without ablaut. The word final f in overgeblyf ‘remained’ (8d) would be absent in Afrikaans as -gebly, and sleggezeg ‘insult’ in (8c) would be sleggesê, without the -g and with a lengthened vowel instead (indicated with ê). Some of these forms were not used in Dutch either, illustrating the intermediate stage of the development of the Afrikaans verbal system, when speakers did not know the Dutch forms and used some forms and affixes unsystematically and ‘incorrectly’.
The data from the diary indicate that many different forms of inflection are used, but full Dutch inflection is not the norm for this author, and it is not used consistently. For example, gehoor ‘heard’ is used with first person plural subjects and a third person singular subject as it would be used in Afrikaans, and gehoort is used with a third person singular subject as well. The Afrikaans form gekom ‘came’ is used once with a third person singular subject, but gekomen is used several times with subjects of different persons and numbers. Interestingly, the most frequently occurring inflectional suffix is -en, regardless of whether the verbs are regular or irregular in Dutch. This demonstrates that the Dutch inflectional system does not function as a system anymore, and its remnants are used in fairly idiosyncratic ways by Van der Walt.
In the ABW-corpus, 420 verb forms are used with het and its variants, and 275 or 65 % of these include Dutch inflection which is no longer used in Afrikaans. The most frequently used inflectional suffix is -t/d, which is not unexpected as the ending of the past participles of regular verbs. The other frequently used suffix is -en, which is also not unexpected for past participles, as it frequently occurs with Dutch past participles of strong verbs. What this indicates is that some measure of Dutch inflection is still used widely at this time, although not necessarily according to Dutch norms.
So far in this section, the focus has been on past participles as they are used in the perfect tenses, as the present perfect had replaced the preterit to a large extent in Afrikaans, and only remnants of the past perfect remains. However, the preterit was still used during this time, and should also receive attention when looking at the past tense.
Historical Cape Dutch / Afrikaans corpora are not tagged for part of speech, and no software exists currently to do this; as historical corpus linguists would know, tagging historical documents automatically, especially those by semi-literate writers, is notoriously difficult. Another way to investigate the preterit had to be found then.
The first step was to take the most frequently used past participles in the data sets above, in which het and its variants are used as past tense auxiliaries. Next, the corpora were searched for Dutch preterit forms of these verbs, carefully taking possible spelling variations into account. The reason for this approach is to compare the usage of a specific set of verbs in their past participle and their preterit form. This should give a rough indication if preference is already given to the Dutch perfect / Afrikaans past tense, or the Dutch preterit.
The ten[5] most frequently used lexical verbs in the diary data set above, their frequencies, and the frequencies of their preterits, are reported in Table 3.
The 10 most frequently used past participles in the diary and their preterits.
| Past participle | Frequency | Preterit | Frequency | Translation |
|---|---|---|---|---|
| gekomen | 18 | kwam(en) | 48 | came |
| gehad/t | 11 | had(den) | 0 | had |
| gedink | 10 | dacht(en) | 15 | thought |
| gevoel | 7 | voelde(n) | 0 | felt |
| gezien | 7 | zag(en) | 1 | saw |
| gezeg | 6 | zei(en) / zegde(n) | 0 | said |
| ontvang | 6 | ontving(en) | 0 | received |
| gehoor | 5 | hoorde(n) | 0 | heard |
| geslaap | 5 | sliep(en) | 0 | slept |
| gekryg | 4 | kreeg / kregen | 0 | got |
Of the ten verbs, only three occur in the preterit. There is one occurrence of zag ‘saw’, compared to seven occurrences of gezien ‘saw’. Then there are 15 occurrences of dacht(en) ‘thought’, compared to only ten occurrences of gedink ‘thought’. However, this particular case is slightly more complex than it appears – the modern form dag and an alternate form dog is still in use today in Afrikaans, but with a shift in meaning to indicate ‘mistakenly thought’ (see example 9a). Of the 15 uses of dacht(en), eight are used with this meaning, and the remaining seven to indicate the simple past (see example 9b).
| Ons | dach | toen | dat | ons | nu | kan | rus. | O | nee, |
| we | thought | then | that | we | now | can | rest | oh | no |
| de | Camp | moet | nu | verzit | wort… | ||||
| the | camp | must | now | moved | become | ||||
| ‘We thought then that we could rest. Oh no, the Camp must be moved now…’ | |||||||||
| Maar | ik | dach | toen: | Wach | jele | beert | kom | aan. |
| but | I | thought | then | wait | your.pl | turn | come | on |
| ‘But I thought: wait, your turn will come.’ | ||||||||
The shift in meaning seems to be underway already in this dataset, and this together with the maintenance of the form in modern Afrikaans could explain why it is used more frequently than the past participle.
Finally, kwam ‘came’ is used 48 times, and gekomen ‘came’ is only used 18 times. While kwam is sometimes used as a linking verb (see example 8c) (kom ‘come’ is used in this way in modern Afrikaans), only the instances where it is used as a main verb are taken into account here.[6] It means that for this specific verb, the preterit is still the preferred form for the past tense for Van der Walt, although the past participle is also used.
Overall, it seems that the modern Afrikaans past tense, based on the Dutch present perfect, is the preferred form for the past tense in general in the diary, apart from specific verbs with a high frequency of the preterit.
Regarding the ABW-corpus, it paints a somewhat similar picture of the past tense (see Table 4).
The 10 most frequently used past participles in ABW-corpus and their preterits.
| Past participle | Frequency | Preterit | Frequency | Translation |
|---|---|---|---|---|
| ontvangen | 113 | ontving(en) | 13 | received |
| gehad | 71 | had(den) | 153 | had |
| gehoord | 57 | hoorde(n) | 10 | heard |
| geschreven | 53 | schreef / schreven | 27 | wrote |
| gezien | 42 | zag(en) | 13 | said |
| gedaan | 30 | deed / deden | 10 | did |
| vernomen | 27 | vernam(en) | 3 | heard |
| gekregen | 24 | kreeg / kregen | 44 | got |
| gestuur | 23 | stuurde(n) | 3 | sent |
| gemaakt | 21 | maakte(n) | 7 | made |
The preterit forms of all ten[7] of the most frequently used past participles in the ABW-corpus dataset used with het and its equivalents occur in the corpus as well. This is not surprising, as the corpus is much larger than the diary corpus, and many different people contributed to it. This means that individual idiosyncrasies do not skew the data as is typically the case with a single authored corpus.
All of the past participles occur more frequently than their preterit counterparts, except for two. The preterit kreeg / kregen ‘got’ occurs 44 times, and the past participle gekregen ‘got’ just more than half that at 24 times. The form had(den) is used 153 times as a main verb, with gehad at 71 (less than half). In this case again, there are specific verbs with a clear preference for the preterit, and for the rest the past participle is used much more frequently. Looking at the total frequencies of the past participles and the preterits, the preterits’ frequency is less than two-thirds that of the past participles. This further suggests that the modern Afrikaans past tense is already the preferred form, on its way to push out the Dutch preterit. This process was probably sped up by standardization which started a decade later, during which the preterit was almost completely eliminated (apart from specific remnants still in use today; see Conradie 1999).
6 Discussion
The trends in the use of the past tense in the diary and the ABW-corpus do not differ all that much, but the differences are interesting to note.
The author of the diary, the primary corpus for this article, is a young woman, although not with a high level of literacy. Nevertheless, young women are known to lead in the adoption of new variants during language change, especially with changes from below (Labov 2010: 197). The expectation would be, then, that she would make more use of the Afrikaans past tense – based on the Dutch perfect, rather than the preterit – compared to the ABW-corpus written by many different authors.
Regarding the use of past tense auxiliaries, Van der Walt does not predominantly use the Afrikaans variant het ‘have’, but rather heef, although het does feature more than the remaining variants. Her choice of auxiliary is not the new Afrikaans one, but it is still important to note that she has a clearly preferred variant used in all grammatical contexts. In this way, her use is not too far off from modern Afrikaans.
In the ABW-corpus, there is more variation in the choice of past tense auxiliary – in order of frequency: heb, heeft, heef, and then hebben, and finally het, with other variants used infrequently. What this indicates is that the speech community had not made a clear choice for any of the variants yet at this time, and het would become the dominant variant shortly after this, and the only variant in due time.
In terms of the lexical verbs used in the past tense, Van der Walt’s choices are more clearly ahead of the ABW-corpus. Only 29 % of the verbs used with variants of het have clear Dutch inflection, and the preterit is used quite infrequently apart from specific verbs. The verbs used with variants of het in the ABW-corpus have more than double the proportion of Dutch inflection, at almost two thirds, indicating that it was still better maintained in the speech community in general. The Dutch perfect with the past participle is mostly preferred in the ABW-corpus dataset, rather than the preterit, as well. This points towards the fact that the perfect was already the main past tense form at the time – the biggest difference between the diary and the ABW-corpus is the amount of Dutch inflection still being used with the past participles in the past tense.
The text types of the two corpora are different: the main corpus consists of a diary, and the reference corpus consists of letters. Although the nature of the texts is different in certain respects (as discussed in Section 3), they are still personal documents, written spontaneously, by mostly semi-literate writers. In terms of the past tense investigated in this study, there were no differences that could be attributed to the difference in text types, as both of these text types have a significant focus on reporting events from the past, although the diary typically does not discuss possible future events and prospects like letters do. This would, however, affect the use of future time reference more than the use of the past tense. There is no indication of greater formality in the language use of the diary, as is often the case, but it may be due to both the low level of education received by the author, and the fact that there was no standard yet established for Afrikaans (even though it differed from Dutch a great deal).
7 Conclusions
At the turn of the twentieth century, the development of modern Afrikaans from Dutch was quite far along, but there was still a great deal of variation in language use in general. The formal standardization and recognition of Afrikaans were still to come, and the people in the Free State and Transvaal were very much focused on another matter – the Anglo-Boer war. Language was simply a means to communicate with loved ones through letters, and record events in diaries, while trying to survive the war.
The letters and diaries written during the war are important sources of linguistic data, and the analyses in this article reveal the state of the past tense in particular during this time. While the Afrikaans past tense, based on the Dutch perfect, was already the primary past tense form at the time, the preterit was still used to an extent. There is further also variation regarding the past participles used in the past tense – while the Dutch inflectional system is still employed to an extent, it is somewhat idiosyncratic and, in the diary, not systematic any longer. This spontaneous development paved the way for the elimination of the preterit during standardization, and the regularization of the past participle used in the past tense, leading to the Afrikaans past tense as it is used today.
Funding source: National Research Foundation
Award Identifier / Grant number: RA191204495962-PR-2023
Acknowledgments
Thank you to the attendees of the HiSoN conference in Brussels in 2023 for their valuable questions and feedback on my presentation. Further thanks to the Afrikaans Language Board and the War Museum in Bloemfontein for working on the publication of the diary and involving me in the process. Thank you to the South African National Research Foundation for the Incentive Funding for Rated Researchers (Grant 132110). Opinions, findings and conclusions or recommendations expressed in any publication generated by the NRF supported research is that of the author alone, and the NRF accepts no liability whatsoever in this regard.
References
Ávila-Ledesma, Nancy E. & Carolina P. Amadro-Moreno. 2023. The seas was like mountains’: Intra-writer variation and social mobility in Irish emigrant letters. Journal of Historical Sociolinguistics 9(2). 243–261. https://doi.org/10.1515/jhsl-2022-0042.Search in Google Scholar
Brown, Joshua R. 2019. Civil war writings of the Pennsylvania Dutch. Journal of Historical Sociolinguistics 5(2). 20180032. https://doi.org/10.1515/jhsl-2018-0032.Search in Google Scholar
Conradie, C. Jac. 1981. Die ontwikkeling van die Afrikaanse voltooide deelwoord I: 1652–1875 [The development of the Afrikaans past participle I: 1652–1875]. Tydskrif vir Geesteswetenskappe 21(4). 275–284.Search in Google Scholar
Conradie, C. Jac. 1999. Preterite loss in early Afrikaans. Folia Linguistica 33(1–2). 19–38. https://doi.org/10.1515/flin.1999.33.1-2.19.Search in Google Scholar
Conradie, C. Jac. 2006. Defleksie en grammatikalisasie: die Afrikaanse hulpwerkwoord het [Deflection and grammaticalization: the Afrikaans auxiliary het]. Journal for Language Teaching 40(1). 86–98.Search in Google Scholar
Deumert, Ana. 2004. Language standardization and language change. The dynamics of Cape Dutch. Amsterdam: John Benjamins.10.1075/impact.19Search in Google Scholar
Elspaß, Stephan. 2012. The use of private letters and diaries in sociolinguistic investigation. In Juan Manuel Hernández-Campoy & Juan Camilo Conde-Silvestre (eds.), The handbook of historical sociolinguistics, 156–169. Malden: Wiley-Blackwell.Search in Google Scholar
Giliomee, Herman & Bernard K. Mbenga. 2007. Nuwe Geskiedenis van Suid-Afrika [New History of South Africa]. Cape Town: Tafelberg.Search in Google Scholar
Hamman, Kezia. 1965. Dagboek van ’n Bethulie-kampdogter [Diary of a Bethulie camp daughter]. Kezia Hamman.Search in Google Scholar
Jewell, James R. 2003. Using barbaric methods in South Africa: The British concentration camp policy during the Anglo-Boer War. Scientia Militaria 31(1). 1–18. https://doi.org/10.5787/31-1-140.Search in Google Scholar
Kirsten, Johanita. 2016. Grammatikale verandering in Afrikaans van 1911–2010 [Grammatical change in Afrikaans from 1911–2010]. Vanderbijlpark. North-West University PhD thesis.Search in Google Scholar
Kirsten, Johanita. 2019. Written Afrikaans since standardization: A century of change. Lanham, MD: Lexington Books.Search in Google Scholar
Knüsli, Ladina. 2019. “Sollte dies mein Geschreibsel meine theure Heymath erreichen”: Linguistic variation in the diary of a nineteenth-century Swiss German migrant. Journal of Historical Sociolinguistics 5(2). 20180033. https://doi.org/10.1515/jhsl-2018-0033.Search in Google Scholar
Labov, William. 2010. Principles of linguistic change: Cognitive and cultural factors, vol. 3. Malden: Wiley-Blackwell.10.1002/9781444327496Search in Google Scholar
Le Roux, Thomas H., Daniël F. Malherbe & Johannes J. Smith. 1917. Afrikaanse Woordelijs en Spelreëls [Afrikaans word list and spelling rules]. Bloemfontein: Het Volksblad-Drukkerij.Search in Google Scholar
Litty, Samantha. 2019. Letters home: German-American Civil War soldiers’ letters 1864–1865. Journal of Historical Sociolinguistics 5(2). 20180030. https://doi.org/10.1515/jhsl-2018-0030.Search in Google Scholar
Malherbe, Daniël F. 1917. Afrikaanse Taalboek: praktiese wegwijser bij die vernaamste moeilikhede in verband met die Afrikaanse grammatika [Afrikaans Language Book: practical guide to the most important difficulties with the Afrikaans grammar]. Bloemfontein: Nasionale Pers.Search in Google Scholar
Nel, Antoinette. 2016. Ons is nog alen Vres door Den Zegen Des Heeren. Gesondheidsformules in Afrikaans-Nederlandse persoonlike briewe uit die Anglo-Boereoorlog [Health formulas in Afrikaans-Dutch personal letters from the Anglo-Boer War]. Vanderbijlpark: North-West University MA dissertation.Search in Google Scholar
Pretorius, Engela & Daleen Krige. 2011. Die dagboek van ’n Boeremeisie in diens van die vyand – siekte en sorg in die Bethulie-kamp [The diary of a Boer girl in service of the enemy – illness and care in the Bethulie camp]. Journal of Contemporary History 36(2). 70–89.Search in Google Scholar
Rayson, Paul. 2016. Log-likelihood and effect size calculator. https://ucrel.lancs.ac.uk/llwizard.html (accessed 12 December 2023).Search in Google Scholar
Ross, Robert. 2009 [1999]. A concise history of South Africa, 2nd edn. Cape Town: Cambridge University Press.Search in Google Scholar
Rutten, Gijsbert & Andreas Krogull. 2021. The observee’s paradox: Theorizing linguistic differences between historical ego-documents. Neuphilologische Mitteilungen 122(1-2). 284–318. https://doi.org/10.51814/nm.101806.Search in Google Scholar
Scholtz, Johannes du P. 1958. Die ontstaan van die Afrikaanse werkwoordelike vormstelsel [The origin of the Afrikaans verbal sytem]. Tydskrif vir Wetenskap en Kuns 18(1). 61–99.Search in Google Scholar
Scott, Mike. 2022. WordSmith tools version 8. Stroud: Lexical Analysis Software.Search in Google Scholar
Virtuele instituut vir Afrikaans [Virtual institute for Afrikaans]. 2023. VivA Korpusportaal [VivA Corpus portal]. http://korpus.viva-afrikaans.org/whitelab/explore/corpus (accessed 16 April 2024).Search in Google Scholar
Worden, Nigel. 2012 [1993]. The making of modern South Africa: Conquest, apartheid, democracy, 5th edn. Hoboken: John Wiley.Search in Google Scholar
© 2025 the author(s), published by De Gruyter, Berlin/Boston
This work is licensed under the Creative Commons Attribution 4.0 International License.
