Translating a guidebook:
addressing reader expectation
A small-scale corpus study of direct reader address in a Swedish-English translation
Author Carina Sjöberg-Hawke Supervisor Jukka Tyrkkö Examiner Jenny Ström Herold Semester Spring 2018
Subject English Level Advanced Course Code 4EN31E
i
Abstract
This study analyses the comparative frequency of “direct reader address” in English and Swedish walking guidebook texts. Through a quantitative and qualitative analysis of specific linguistic features that constitute “direct reader address”, the study aims to highlight the importance of considering reader expectation of a text, i.e. the target culture’s text conventions when translating. The linguistic features in focus are those which help to establish a familiar relationship between reader and writer and thus set up an imaginary dialogue. These are: second-person pronouns, first-person inclusive plural pronouns and verbs in the imperative mood. A translation corpus of walking-guidebook extracts (10 English originals, 11 Swedish originals and their English translations) and my own Swedish-to-English translation of such a text were analysed in order to a) determine to what extent “direct reader address” is used in Swedish and English original walking guidebooks; b) determine to what extent “direct reader address” is retained and used in English translations of Swedish walking guidebooks; and c) discuss the
implications of this for translators of such texts. The results of the investigation show that although “direct reader address” does appear in both Swedish and English original guidebooks, it is more prevalent in English ones. Imperative verbs are the most common of all the relevant linguistic features. The results also show that the trend is not only to retain in English translations what “direct reader address” existed in Swedish originals but also sometimes to add “direct reader address” for reasons of syntax and idiomatic usage. The implications are that a target culture’s text conventions are consequential when translating a walking guidebook because they relate to reader expectation, in particular in relation to linguistic features of “direct reader address”. To translate well, and where deadlines allow, it is recommended that a translator’s strategy should try to address reader expectation.
Key Words
direct reader address, English, first-person, guidebooks, imperatives, interpersonal, personal pronouns, reader expectation, secondary reader, second-person, Swedish, target reader, text conventions, tourist text, translation, translator strategy
Acknowledgements
Thank you to my supervisor, work colleagues and boyfriend for being so supportive, encouraging and providing valuable feedback during this challenging time.
ii
Table of Contents
1 Introduction ________________________________________________________ 1 1.1 Aim and Scope ___________________________________________________ 3 2 Material and Method _________________________________________________ 4 2.1 Analysis Material _________________________________________________ 4 2.1.1 The translation – text analysis and translation strategy _________________ 4 2.1.2 Building the translation corpus ___________________________________ 7 2.2 Data Collection ___________________________________________________ 8 2.2.1 Counting _____________________________________________________ 8 2.2.2 Frequency and chi-square test ____________________________________ 8 3 Theoretical Framework _______________________________________________ 9 3.1 A Translator’s Strategy ____________________________________________ 10 3.1.1 The guidebook genre __________________________________________ 12 3.1.2 Considering language, culture and the Swedish/English context ________ 14 3.2 Direct Reader Address ____________________________________________ 16 3.2.1 Second-person pronouns _______________________________________ 16 3.2.2 First-person inclusive plural pronouns _____________________________ 19 3.2.3 Imperatives __________________________________________________ 20 3.2.4 Not addressing the reader directly ________________________________ 21 4 Quantitative and Qualitative Analysis __________________________________ 23 4.1 Direct Reader Address in the Corpus _________________________________ 23 4.2 Direct Reader Address in the Personal Translation ______________________ 28 4.3 Implications and Evaluation ________________________________________ 32 5 Conclusion ________________________________________________________ 34 References __________________________________________________________ 36 Primary Sources ____________________________________________________ 36 Secondary Sources __________________________________________________ 38 Appendix A: Profiling Results of ST and TT ______________________________ 42 Appendix B: Detailed Result Tables _____________________________________ 44 Appendix C: Chi-Square Test Results ___________________________________ 48
1
1 Introduction
You may be surprised reading this sentence. This paper is of the academic variety; the reader (you) will likely have expected not to be addressed directly with the second-person pronoun.
This probability is high given personal pronouns tend to rarely occur in academic writing (Biber and Conrad, 2009:116 Table 5.2). However, over the years, linguistic features typical of spoken language, such as the English second-person pronoun you, have increasingly been appearing in written English of many informative texts (Sabater et al, 2001:489). This might in part be a consequence of the digital world we live in, which has resulted in communication styles being more instant and in real-time, characteristic of conversation, but, Sabater et al (2001:490) also suggest that it may be an attempt by many companies and administrative bodies to approach their consumers on a more familiar level. Removing the unequal power distribution in the relationship between the participants allows the writer to involve the reader in the communicative process and perceive the experience as actively participating in the communication. Essentially, the text is a “site of interaction” (Hoey, 2001:13), where the writer produces a text which engages the reader and which the reader responds to, a kind of imagined dialogue that has the atmosphere of a face-to-face conversation (Torresi, 2010:128 cited in Cui and Zhao, 2014).
Texts which invite this kind of interpersonal communication most likely have a conative function (Jakobson, 1960), i.e. the writer is appealing to and/or persuading the reader to act on something. One way to achieve this persuasive element in the communication can be to use a direct form of address, such as with the English personal pronoun you. For example,
Francesconi (2011) noted in her study of a tourism brochure of Malta that personal pronouns were “pivotal verbal items” in the discourse. It was clear that they were used in the brochure to establish a familiar relationship between the reader and writer, with the reader being addressed through the second-person pronoun.
A similar genre within tourism that does this is the guidebook. It not only uses the second- person pronoun you to directly address the reader, but also the imperative mood of a verb (typical of the conative function), which is still aimed at the second-person, but the second- person reference is implicit. Both can be seen in example (1) below, highlighted in italics.
2
(1) Take a trip in an open-top tramcar beside the River Axe on the Seaton tramway, which runs for 3 miles (4.8km) from Colyton to Seaton via Colyford. You should see a wide range of birds, including grey herons, kingfishers,
oystercatchers, curlews and egrets. More than 50 different species have been spotted from the tram in one day. (Viccars, 2013:16)
Example (1) illustrates how guidebook text not only informs a reader about a place of interest, but also directs the reader about how to reach this place of interest and command/recommend to the reader what to see and do there. The informative and directive nature of the text is combined with forms of second-person referencing giving the reader the impression that the writer is addressing them directly, as in spoken conversation, almost sounding like a real-life tour guide.
The linguistic features involved in addressing the reader can vary from language to language and situation to situation. Depending on these, a writer will make careful linguistic decisions about exactly who and how many readers they are addressing. For example, if personal pronouns are chosen, one language’s system may only have one lexeme to choose from to address the recipient but another may have a variety of words because gender, number and level of formality need to be indicated. Essentially, the features used can be related to the language used, the situation and the text typology.
A translator has the same decisions to make as a writer, but the context of the target culture could make decisions more strategic. The translator may decide that they focus mostly on the linguistic level (like Nida’s (1964) dynamic equivalence) or maybe more on the functional level (like Reiss and Vermeer’s (2014) skopos theory or Nord’s (2005) translation-oriented text analysis), or maybe a combination of both. Alternatively, a translator may take a more dynamic approach through a combination of these while incorporating other extratextual factors which take into consideration the target culture, e.g. profiling the reader, “reader characterisation” (Sager, 1997). Whichever translation strategy a translator chooses is likely to be partly governed by factors outside their control such as the commission deadline; a tight deadline would limit the time a translator has to analyse the source and target profiles
involved, let alone the reader profile. However, the fact remains that a reader will have expectations about the translation product as a text, the conventions of its genre and its
3
readability and fluency, and therefore, a translator should try to meet those to produce a successful text.
1.1 Aim and Scope
This study aims to highlight the importance of considering a target culture’s text conventions when translating, in particular in relation to linguistic features of “direct reader address”, such as personal pronouns, which help establish a familiar relationship between reader and writer, in a guidebook context. Material for this investigation includes a translation corpus of
walking-guidebook extracts (of English originals, Swedish originals and English translations) and my own personal Swedish to English translation of such a text. There are three primary goals with this investigation:
1. To determine to what extent “direct reader address” is used in Swedish and English original walking guidebooks;
2. To determine to what extent “direct reader address” is retained and used in English translations of Swedish walking guidebooks;
3. To discuss the implications for a translation strategy of such texts.
In order to fulfil these goals, both a quantitative and a qualitative research approach have been adopted. This includes an analysis and discussion of frequencies of “direct reader address”
occurrences in the material. Furthermore, observations of translation choices and strategy are made on instances of English and Swedish correspondences of “direct reader address” in the personal translation. Due to practical constraints, this study will only focus on linguistic features of “direct reader address”.
The remaining part of the paper is organised as follows. Chapter two summarises how the material, the corpus and translation, was produced and how the relevant linguistic features were counted. Chapter three examines a translator’s strategy in relation to text conventions and target culture, and defines “direct reader address”. The results of the investigation follow in chapter four, with an examination and discussion of the frequencies of “direct reader address” and the translation strategies involved. The final chapter draws some provisional conclusions.
4
2 Material and Method
This chapter explains how this investigation was conducted, starting with a description of the analysis material used, followed by an explanation of how the data was collected and
compiled.
2.1 Analysis Material
The material for investigation consists of two parts; a personal Swedish-to-English
translation, and a translation corpus developed specifically for this study. The corpus contains Swedish and English original texts of the walking guidebook variety and also English
translations of the Swedish originals. This section first summarises the source and target text of the personal translation and the translation strategy that was used. This is followed by a description of the building of the corpus including how texts were selected and obtained.
2.1.1 The translation – text analysis and translation strategy
Four chapters (Olskroken och Redbergslid pp.21-27, Älvsnabben pp.50-55, Spårvagn 5 pp.70- 76 and Donsö och Styrsö pp.128-133) from the Swedish walking guidebook Göteborg till fots: trottoarturism och spårvägssafaris (Hansson, 2016) were translated into English. Before the translation was performed, a profile was generated for both source text (ST) and target text (TT) to determine the text types, the target readers and the commission requirements of the texts. Both Nord’s (2005) translation-oriented text analysis method and Biber and Conrad’s (2009) analysis of situational characteristics were applied. A summary of those results follows (an exact breakdown can be found in Appendix A).
The ST, published in Sweden in 2016, is part of a printed book with text and pictures interwoven throughout. As its foreword indicates, the aim of the book is to guide Swedish tourists around Gothenburg in a friendly manner, “Den här guideboken vill vara din föjeslagare [sic] genom många olika sidor av Göteborg – på tvären och på djupet.” (“This guidebook wants to be your companion through the many different facets of Gothenburg – in all directions.”) (Hansson, 2016:2). It is a guidebook with a focus on walking. The ST’s intended functions are informative, instructive and conative. The text informs the reader about aspects of Gothenburg, its history, geography, architecture, culture, etc., instructs the reader in some parts by directing them which routes to take on foot, and appeals to them (conative
5
function) to see and do what has been informed and instructed about. Based on the translation brief, the TT should function in the same way.
The guidebook itself indicates who the addressees, or target readers, of the ST are, that they are either curious visitors or locals of Gothenburg (Hansson, 2016:2). The TT’s readers would be the same. However, the target readers of the ST are further defined by the fact that the information is given in Swedish and thus giving rise to the readers having an expectation about the language used, the style, then text conventions, based on Swedish language/culture expectation. The readership of the TT might be much wider since the addressees will be English-speaking tourists, and English is more widely spoken than Swedish. They are more likely to be tourists from outside Gothenburg and Sweden and are unlikely to share the same amount of knowledge about the city and country as the Swedish target reader. This might require the translation to be more explicit in places, such as when explaining a specific cultural aspect or Swedish word. However, the English-speaking tourists may also be locals/residents of the city.
In order to fulfil the function of the book, which clearly places the needs of the target readers first, it seemed logical to aim at producing what Nord (2005) calls an instrumental translation.
This is a functional translation where the message is reproduced through a text that reads like an original target language text, rather than a documentary one, where the reader is aware that it is a translation, typified by literal or word-for-word translations, which, although it may be awkward, would give the TT reader access to the source language culture while accessing the ST’s ideas. These two translation products are very similar to House’s (1997) two, but she labels the instrumental type covert and the documentary one overt. The latter does not try to be an original and the former should have the status of an original source text. A documentary or overt translation might appeal to some, but for those readers who want the text to only function as its function of guide rather than anything else, and want coherency, then a translation that is more like texts they are used to, whether termed instrumental or covert, would appeal more. The message of translations which are unlike an original source text may result in misunderstandings, such as not reaching the destination.
In endeavouring to produce an instrumental translation, Nord’s (2005) three-step approach was adopted. First, the translation brief was considered, comparing intended text functions, participants, time and place of text reception, medium and motive of ST and TT, as
6
summarised above. Secondly, an analysis of the role of the ST was performed, examining feasibility of translation, decision on what ST linguistic, structural and even content items were relevant in achieving a functional translation and what strategy to fulfil the brief. The final step in Nord’s (2005) translation approach is to establish a functional hierarchy of translation problems (working top-down).
The differences between the ST and TT amounted to differences between the target readers, not only of language but also of culture, such as genre conventions and presuppositions. This meant that items such as reader address, which can vary interlingually (see section 3.2), and place and street names, which can be easily understood in the source language (SL) but not in the target language (TL), were some items to be tackled. Decisions on such items were based on personal experience of the TL, guiding and guidebooks, thereafter supported with
referencing in corpora such as BNC and COCA, as well as Google search. For official terms, Gothenburg’s tourist website (www.goteborg.com) was utilised. Based on the commission brief, some content which possibly could be left out in the English version was left in for editing later. Additionally, the context of some content was made in previous chapters and thus such content was not explained in the translation where could have been the case.
Translation strategies included: being word-for-word, where the language system allowed for that, as example (2) illustrates:
(2) Redbergslid är en av Göteborgs livligaste stadsdelar.
Redbergslid is one of Gothenburg's most vibrant boroughs.
using more or fewer words in the TL depending on syntactic needs, and adjusting word order where appropriate too, as can be seen in example (3):
(3) Här låg en stadsdel uppbyggd på 1890- talet med rader av
landshövdingehuskvarter och mängder av butiker och småföretag…
Built in the 1890s, this was once a borough with rows of governor house neighbourhoods and many shops and small businesses…
And being explicit when, for example, a road was not clearly a road, as in example (4) below:
7
(4) Stanna vid krysset med Åsenliden Continue until you reach the cross on the road Åsenliden.
This translation, which endeavoured to be an instrumental translation with a focus on the target readers’ needs, constituted one part of the investigation material. The other part was the corpus created for this study.
2.1.2 Building the translation corpus
Since part of the goal of this investigation involves obtaining the frequency of certain linguistic features in a certain genre, additional texts were required to provide more
comprehensive observations. A common practice in translation studies and research today is to incorporate the use of corpora (Munday, 2012:283ff.). A corpus is a collection of authentic texts which can provide “access to large quantities of real world data” (Peters, Picchi, and Biagini, 2000:73); in other words, access to language in use (Lindquist, 2009:1), a helpful tool for a translator when making natural target language choices of a lexical, grammatical or structural nature. Consequently, to be able to analyse the actual language in use for “direct reader address” in this study, a Swedish-English translation corpus was created, consisting of extracts from Swedish guidebooks and their English translations as well as extracts from English-written guidebooks. Criteria for selecting the texts were based on the ST and TT profiles of the personal translation, and are summarised below:
• Genre: guidebooks with a walking focus
• Languages: Swedish originals with English translations and English originals
• Addressees: respective Swedish/English tourists
• Intended functions: informative, instructional, conative
• Motive: to accompany the tourist on a guided tour of the destination
Due to practical constraints, local libraries were the access point for the corpus material.
However, this was limited, in particular regarding the English translations and as a consequence, it was necessary, to utilise inter-library loans. This proved more time-
consuming than expected and limited the number of and the length of the extracts added to the corpus. The final count was 11 Swedish originals (20,345 words), 11 English translations (of those Swedish originals) (23,424 words) and 10 English originals (18,637 words).
8
2.2 Data Collection
Having created the corpus, the data then had to be collected, which began with defining aspects of the method, followed by counting words and features and then calculating frequencies. Due to the small-scale nature of this study, the data was collected using non- computational methods.
2.2.1 Counting
Since word counts were being carried out, what counted as a “word” needed to be clarified.
Based on Biber and Conrad’s (2009) work, contractions were counted as one word rather than two, as were hyphenated words, and individual numbers. Telephone numbers, for example, also seen as one unit, were counted as one word. This was applied across all analysis material.
The linguistic features which constitute “direct reader address” for this study were also to be counted. However, first they also needed to be identified and defined (see section 3.2 for details). Occurrences of a single feature could be realised in more than one way, so features were counted per feature, whether they were one word or more.
Due to the medium (printed books), and resources available, the counting was performed manually and each instance of “direct reader address” recorded in a notebook.
2.2.2 Frequency and chi-square test
To be able to compare results, the problem of higher or lower counts in different lengths of text needed to be addressed. This was done by calculating an average hit count or, as is more conventionally known, a standardised frequency. Based on Biber and Conrad’s (2009:62) recommendations, the actual counts of “direct reader address” instances were divided by the total word count multiplied by the fixed amount of text (in this study 1,000 words). This can be visualised as follows:
(instances count / total word count) x 1000 = instances per 1,000 words
9
So, for example: if 24 instances of “direct reader address” were found in a text with a total word count of 1,876, then the standardised frequency would be 12.8 (all frequencies rounded up to one decimal place):
(24/1876) x 1000 = 12.8
To determine whether, based on the amounts of data and the sizes of the differences observed, it was reasonable to claim that the difference was ‘real’, that is, that the two types of texts (ST and TT, or English originals and English translations, etc.) really differed (or did not) in the real world, a statistical hypothesis test, known as the chi-square test (based on guidance from Lindquist, 2009 and Tyrkkö, 2018) was performed. This p-value (probability value) was calculated using an online chi-square test calculator (Stangroom, 2018).
Before the results are presented, some of the principle ideas of a translator’s strategy, of text conventions and of reader expectation need to be outlined. An explanation of the term “direct reader address” must also be provided.
3 Theoretical Framework
Texts are traditionally divided up into the following three categories: text types, genres and registers. What distinguishes one from the other is a point of discussion for many scholars.
However, this paper has already referred to the guidebook as a genre. Therefore, it is necessary to clarify what this means. This investigation acknowledges the following text typology definitions: that text type is related to its function (Reiss, 1977/1989; Nord, 2005);
genre to its purpose in a particular situation (Trosborg, 1997); and register to its language features for its function (Biber and Conrad, 2009). Register experts Biber and Conrad (2009) distinguish between genre and register by the former having conventional features which are representative of the whole text (like the opening and closing salutations of a letter) and the latter varying in the use of linguistic features. As part of their register analyses, they examine situational characteristics, which are similar to Nord’s (2005) translation brief variables (see section 2.1.1). This perspective is not new and can be traced back to Michael Halliday’s systemic functional linguistics (SFL), a discourse analysis model which developed in the 1970s (Munday, 2012:136f.). Essentially, his model is concerned with the lexico-grammatical choices that a text producer makes since it is these that form the text’s meaning. The meaning
10
can be ideational, interpersonal or textual. Each of these are connected to three register variables, respectively: field (topics/actions of language), tenor (participants/relationships) and mode (form). The functional focus of Halliday’s model is appealing to translation studies and has been utilised by many scholars over the years, including Nord (2005) and Baker (2011). This translation study with a focus on reader address finds the interpersonal and tenor elements of particular interest.
Consequently, the following chapter reveals what research states about such writer-reader elements. Firstly, a presentation of past and current approaches to a translator’s strategy in the context of text typology is provided. This is subdivided into an examination of the guidebook as a genre and in translation, and a discussion of aspects of language and culture relevant to the interpersonal and guidebooks, with a focus on the Swedish-English context. The second half of the chapter is devoted to defining the term “direct reader address” and identifying the features relevant to it for this study.
3.1 A Translator’s Strategy
According to Sager (1997), a translator’s strategy can be divided into two approaches: the static and the dynamic. The static approach is the one typical of the past which focussed on equivalence – a concept which focuses on changing linguistic form but content remaining the same with vague criteria for the translation such as Reiss’ ‘adequacy’, Toury’s ‘acceptability’, Nord’s ‘loyalty’ or Baker’s ‘pragmatic equivalence’ (Sager, 1997:26). A common term in the latter years was functional equivalence, a term Nida (Nida and de Waard, 1986:36) used to distance himself from his earlier term dynamic equivalence, which he had coined in 1964, an equivalent effect concept that highlighted a translator focusing on producing a natural
translation rather than having literal accuracy. With functional equivalence, Nida (Nida and de Waard, 1986) wanted to include the importance of how people interact in cultures. This idea was furthered by Reiss and Vermeer’s Skopos Theory (2014), which was concerned with having a translation strategy where purpose, skopos, and function of the TT in the target culture were at the forefront. This approach moved away from a focus on linguistic features and instead focussed on communication theory and translation as “an act of intercultural communication” (Munday, 2012:133). Nord (2005) developed this functional approach through her two types of translation product: documentary and instrumental (see section 2.1.1 for details). Despite many aspects of both ST and TT being very carefully considered, Sager
11
(1997:26) argues that this approach is still part of the vagueness of equivalence, since Nord (2005) often alluded to ‘loyalty’ of a translation, which Sager (1997:26) claims is part of vague criteria. Additionally, to be loyal to the writer can be difficult since the writer’s intention cannot always be known.
Sager (1997) instead advocates translating using a more dynamic approach, whereby the translation is considered as only “a step in a communicative process between two cultures.”
(ibid.:26). The translator is a mediator between writer and reader, trying to pass on the message as the writer intended in a form and language that the reader recognises (the genre culture). When composing, the writer will likely have made linguistic, content and structural choices based on the usual style of that kind of text and will likely be assuming that their target reader’s intertextuality is in line with this as it will help communicate the message more easily. For example, a scientific report writer will assume their reader is familiar with the basic textual conventions of a scientific report, such as avoidance of conversational linguistic features such as personal pronouns, contractions and non-specific vocabulary (Monash University, 2018), and inclusion of text organisation features such as the report being divided into introduction, theory, method, results, discussion, etc. (University of Wisconsin-Madison, 2018) If the writer deviates from any of this, it could affect communication, since the reader will possibly be confused by the deviation, it not meeting their expectation. Ideally, writer intention would be equal to reader expectation but Sager (1997:27) says this is not possible without the writer knowing the reader and vice versa.
In the translation context, reader can actually involve two types: the primary reader, the receiver of information of the source text, and the secondary reader, the receiver of
information of the target text (Sager, 1997:28). Even if the target text has a similar type of audience as the source text, with a translation, the readers are always going to be different in regards to their language and cultural background. Both writers and readers know the forms and conventions of texts based on what they have encountered over the years (Hoey, 2001).
However, with a translation, those forms and conventions are unlikely to be that similar for writer and secondary reader. Additionally, a number of actions and players could arise between completed source text and the beginning of a translation.
For example, someone other than the writer may request the translation (a commissioner).
This party may have a different intention to what the writer had had. Additionally, the text
12
may have undergone machine translation before reaching the translator, giving the translator both a translator and editor role. However, whatever the situation is by the time the
commission is entrusted to the translator, in order to produce a text which the target reader recognises in regards to text conventions and can read with ease, the translator would need to be provided with specific text product details. Sager (1997:26) argues that this would, for example, mean job specifications that not only include the usual delivery details and price, but also “a characterisation of the future readers of the translation, their expectations of the
translated text and the use they are likely to make of it.”. This is in a way an extension of Nord’s (2005) and Biber and Conrad’s (2009) text analysis with a stronger focus on the addressees.
Hogg, Liao and O’Gorman (2014) underline, in relation to tourist texts, that if cultural norms of the text, part of the reader’s expectations, are not considered, consequences are two-fold;
not only may the translator fail with their product but the tourism experience could be affected negatively. As Hogg, Liao and O’Gorman (2014) argue, a dimension of culture/genre
expectation is needed in order to translate effectively. A translator’s strategy should thus include meeting the genre norms and the readability and fluency of the target language, which can be summarised as Chesterman terms “product/expectancy norms” (2016). For this
investigation, it is the guidebook norms that are in focus, and in particular those of the English and Swedish context.
3.1.1 The guidebook genre
A guidebook can be defined as a type of tourist text that is a resource for tourists and functions as a mediation between the destination and the tourist (Bhattacharyya, 1997). The perception of a place can be dependent on the information provided and the package that it is in. A tourist text can be delivered through a variety of media, e.g. brochures, websites, books, and apps. The conventions of these can differ for each, in relation to layout, language used, content included, etc. With the main function of a guidebook, like most tourist texts, being to attract and inform their reader (Hogg, Liao and O’Gorman, 2014), the communicative
function is a combination of persuasive and referential. A persuasive text has the receiver of information (the tourist) in focus (try to persuade the tourist to visit somewhere and follow directions) and the referential text, the realities of the world (the information about the destination) (Trosborg, 1997). Guidebooks are also directive (guiding the reader to and
13
around the destination). Walking guidebooks, which this study is particularly focused on, has a primary focus of guiding the reader to and around the destination on foot.
With the reader so much in focus in guidebooks – they are being guided – it is logical that a promotional strategy of “ego-targeting” (Dann, 1996) is adopted. In other words, in order to appeal to the reader, addressing them directly could help single the reader out and make them perhaps feel unique and different to the wider readership that was intended. To fulfil that kind of function, “pivotal verbal items” such as personal pronouns (Francesconi, 2011) could be used. Second-person personal pronouns single out the recipient involving them in the action (event/task), and thus provide them with a sense that they have a familiar relationship with the sender of information. In a guidebook, this can help the reader believe they are being advised and attended to, but Cui and Zhao (2014:26) claim that it is also a way to help with the processing of information, performing the task and remembering the message.
According to Bhattacharyya (1997), guidebooks have not been a particularly well analysed part of tourism. This was, however, addressed to some extent in Thurlow and Jaworski’s study in 2010, where they did a study involving guidebooks, observing that guidebooks are a quintessential feature and typical genre of tourism. They also highlighted that within the guidebook genre there is a wide variety of styles, mentioning the different series of guidebooks that exist with their different but recognisable styles:
…authoritative tone of Baedeker, the picture-book-cum-encylopaedic style of Eyewitness, the ‘alternative’ off-the-beaten-track ethos of Lonely Planet and Rough Guides… (Thurlow and Jaworski, 2010:192).
The recognisable style helps to communicate the message. The styles mentioned above will have influenced the language used by the writers of the books. An authoritative tone is likely to prompt the writer to use an active voice and authorial presence (either singular first-person personal pronoun or plural first-person personal pronoun together with verbs of advice such as recommend or should (see section 3.2). The “picture-book-cum-encyclopaedic style” is likely to have the writer focus on the facts and avoiding addressing the reader. The “off-the-beaten- track ethos” is probably the widest category but probably allows the writer to be very friendly – to engage with the reader – and the communication is probably infused with passion in
14
order to make the “off-the-beaten-track” appealing. The style, though, is not the only aspect of a guidebook that can be recognisable, the content can be too.
The information conveyed in a guidebook includes practical information on how to get to the destination and how to get around there, facts about the place/sight in relation to its history, geography architecture and other cultural aspects, social activities and itineraries are also commonplace, including tips about places to eat, sleep and shop. Some guidebooks include some phrase help. With many guidebooks containing most of these communicative stages (Van Leeuwen, 2005), the reader is helped to recognise the type of text they are reading. This familiarity helps communicate the message to the reader with some ease. The reader is usually helped along the way as well with the information provided at the beginning, usually about the aim of the guidebook and the intended readers who were in mind when the text was composed, as with the foreword of the personal translation’s ST (see section 2.1.1).
The general target readers of such texts, which was touched upon in section 2.1.1, tend to be foreign visitors to the destination. These do not necessarily have to be foreign to the country but just to the city or area. In relation to the latter, the visitors might just be locals of the city where the destination is but have not visited the particular sight before. The readers of these texts could also have a different native language to that of the text they are reading. Reader expectation is thus both language- and culture-related.
3.1.2 Considering language, culture and the Swedish/English context
Linguistic features used for “direct reader address” are culture-specific and thus can even be sensitive areas for translation. The cultural factor could exist on many different levels
(individual, family, locally, nationally and globally, including publishers’ style). In English, if a writer chooses to address the reader directly, the writer does not need to make a choice about which pronouns to use. The lexeme you is used for both the second-person singular and plural subject pronoun and the second-person singular and plural object pronoun (see section 3.2.1 for more details). In other languages, however, this is not the case. In Spanish, for example, the lexemes used can depend on gender, number and level of formality (Sabater et al, 2001; and Estling Vannestål, 2007:317). This is similar in Thai and Indonesian with status also being relevant (Munday, 2012:153). In Swedish, just as English, there is no gender marking, but unlike English, Swedish uses different lexemes for singular, plural subject and object pronouns (Estling Vannestål, 2007:318). Depending on language, then, a text producer
15
will make careful linguistic decisions about exactly who and how many readers they are addressing.
With some cultures being more interpersonal than others, certain scholars, such as House (1997), consider that this should be reflected in a translation. House (ibid.:115-117) gave the example of business communication in Germany where it tends to be preferred that the focus is on the content, whereas in English the focus is more on relationships and communication between people, the interpersonal. There has been a vast amount of research into intercultural communication, which this study certainly leans into, and the interpersonal, which certainly relates to this study’s reader address focus, is claimed to be part of high-context cultures (Gibson, 2002:33). Low-context cultures, in contrast, are task-oriented instead. High- and low-context cultures was a concept created by the anthropologist Edward Hall (1976). His research was agreed with in House’s (1997) portrait of Germans in business since they are considered very low down the scale. Sweden (Scandinavia) is considered slightly less low- context and the UK even more so. According to Hall’s scale, the UK is more interpersonal than Sweden (Gibson, 2002:33). These categorisations are not absolute and there will always be some individual deviation. However, to improve communication between individuals of differing cultures, it can be helpful to have some understanding of the way certain cultures tend to be.
These culture categorisations could also be applied to textual culture. Hall (1976) presented a simple example of chemistry and physics scientists tending to have lower-context cultures, i.e.
they are task-oriented, since chemistry and physics are considered hard science fields. This can be compared to scientists working in the field of ecology where communication is
unlikely to be as explicit since the field is considered a softer science field.. The guidebook is about the interpersonal, suggesting high-context, but a considerable amount of the language is straightforward, and explicit, just as in marketing – another persuasive text, suggesting low- context. Thus, the explicit nature of “direct reader address” would suggest that guidebooks have a low-context rather than a high-context culture. This could also indicate that if Sweden is considered more low-context than the UK, as mentioned earlier, then Swedish guidebooks should be more explicit in nature of “direct reader address”.
In summary, meeting a reader’s expectation is dependent on genre norms of content and style and a reader’s language and culture (Kelly 1998:36). In the context of translation, this implies
16
that a translator would need not only to profile the typology of the source and target texts, but also to profile the target reader, and then make decisions on how to translate in relation to meeting those specifications. With Swedish guidebooks, this could involve correlating the recognisable relaxed style of Lonely Planet with a Swedish reader’s low interpersonal cultural manner. Once these aspects have been established, then the codes of the register, the linguistic features used to communicate the function (“direct reader address”), need to be unravelled for translation.
3.2 Direct Reader Address
Oxford English Dictionary defines address as “a speech directed at or appealing to a particular audience; a written copy or account of this” (Oxford University Press, 2018).
Reader address in this paper then refers to written statements directed at a particular
readership. There are various language options that a writer could use to address their readers:
using proper nouns, personal pronouns, impersonal pronouns, “people” nouns (see 3.2.4), particular verb choices, particular verb forms, and so on. As identified in the previous section, some of these, in particular Francesconi’s (2011) “pivotal verbal items” such as personal pronouns, can create a direct form of communication with the recipient; they involve the reader in an imaginary conversation (Sabater et al, 2001; Torresi, 2010 cited in Cui and Zhao, 2014). Those that do not involve the reader in this way but still convey the same information are not directly addressing the reader. This particular study is concerned with how the reader is directly addressed in text and therefore termed “direct reader address”.
The following subsections identify two sets of linguistic features, made up of five types, which constitute “direct reader address”. As mentioned previously, features which indicate that the reader is not addressed directly, but in effect could have been, are beyond the scope of this study. However, the final subsection below will briefly summarise some relevant
information related to this to give a broader insight to the whole and to help understand some aspects of the results later in this paper.
3.2.1 Second-person pronouns
Essentially, a pronoun replaces a noun or noun phrase to avoid repetition of the noun or noun phrase (Estling Vannestål, 2007:62). It can either function as the head or as a determiner of a noun phrase (Smitterberg, 2013). Personal pronouns are used to identify someone or
17
something that is already familiar, so-called anaphoric reference. The lexemes used can indicate role in speech (speaker/addressee/other), number (singular/plural), gender (male/female/inanimate/non-human), and even the level of politeness and whether the pronoun is the subject or object of a verb (Wiese and Simon 2002:3). Some languages have more features than others to distinguish these characteristics. Spanish, for example, marks gender in first- and second-person plural pronouns (Estling Vannestål, 2007:317). Swedish and English do not, though in Swedish, subject and object personal pronouns are
distinguishable by different lexemes. In English, this is less so (see Table 1 below).
The personal pronouns relevant for this study are those that identify the receiver of
information, i.e. those in the second person. Possessive and reflexive pronouns are closely related in syntax and appearance to personal pronouns (Eppler and Ozón, 2013:23), and thus should also be a part of the second-person reference features in this study. Table 1 below summarises the relevant forms for both English and Swedish.
Table 1. Second-person personal pronouns in English and Swedish
personal possessive reflexive
EN/SE subject object dependent independent
2nd person singular
EN you you your yours yourself
SE du dig din, ditt, dina dig
2nd person
plural
EN you you your yours yourselves
SE ni er er, ert, era er
Note. Data for English from Eppler & Ozón (2013:24 table 2.1) and for Swedish from Larsson (n.d.) and Smitterberg (2013)
As can be observed, in English, the lexeme you is used for more than one function. Swedish, on the other hand, has a much clearer distinction for each, though the reflexive pronoun and personal pronoun object forms do use the same lexeme. Swedish has more lexical items than English for referring to their recipient in the different situations which allows a speaker/writer to make a clear distinction about their recipient, facilitating more “ego-targeting” (Dann, 1996). The English pronoun you is semantically ambiguous, as Estling Vannestål’s (2007:357) example illustrates below:
18
(5) Although liver contains a lot of iron, you should avoid eating it while you’re pregnant.
…bör bör [sic] man/en/du undvika…
though syntax could help identify, and an inclusion of a noun phrase could be used, as underlined in example (6) below.
(6) It’s time for you, the city visitor, to experience a less concrete environment.
This specificity highlights the singular nature of the pronoun and thus an individual recipient, thus targeting the reader slightly more. However, a writer may choose to avoid options such as example (6) since readers of guidebooks are many and could be singular or plural at any one time, and using an ambiguous pronoun eases the writer’s decision since it can cover both options.
In Swedish, there would appear to be no ambiguity since the language uses different words for each of the singular and plural nominatives and accusatives. Consequently, a Swedish writer has to make a choice: to address an individual or a group. With the specific nature of the former, this could be considered more direct in its address than the generic English you.
However, Swedish is not without ambiguity. Plural pronouns can be unclear in their reference. For example, addressing people with ni can be a polite form of du. This though, claims Wide (2016:41), is now rare. This is supported by what Swedish authorities advise their personnel to use in writing: du is preferable when addressing individuals or particular groups, but ni can be used when addressing many or judicial individuals (Språkrådet, 2014) (and both without capitals as has sometimes been the case for a more ‘intermediate’ sign of respect). However, usage changes and this has actually been slightly adjusted from the previous edition (Stadsrådsberedningen, Regeringskansliet, 2009) where it specified that addressing people with ni could be a polite form of du. In contrast, an article in 2013 in one of Sweden’s leading newspapers observed that there seemed to have been a revival of the use of ni (Lerner, 2013), for example, with people using it in shops. This illustrates how studying language in use can be particularly important in regard to the expectation of the target reader, which, in relation to translation, means the secondary reader (Sager, 1997:28).
19
With regards to the possessive and reflexive pronouns, there are some differences between the two languages: with possessives, English has different forms for heads and determiners of noun phrases, but Swedish does not; additionally, English tends to use possessive pronouns in expressions with body parts, clothes, etc., while Swedish prefers the definite form of the noun; some verbs in Swedish are reflexive when their English equivalent is not (Smitterberg, 2013), such as röra sig ‘move’ and förbereda sig ‘prepare’ (Estling Vannestål, 2007:520).
However, it is not within the scope of this study to examine these pronouns at any depth.
Consequently, although they are included in the linguistic features that constitute “direct reader address” in this study because they identify the receiver of information, so are counted, they are not included in the analysis but instead reserved for future study.
3.2.2 First-person inclusive plural pronouns
Another linguistic feature that should be included within “direct reader address” are first- person inclusive plural pronouns. First-person pronouns allow the sender of information to refer to themselves. This self-mentioning allows for two functions: firstly, to exert their authority on the subject matter (when plural, exclusive) and secondly, to involve the reader in the experience (inclusive plural) (Suau-Jimenez, 2016). Since this study is focussed on the recipient being addressed, their involvement in an imagined dialogue (Torresi, 2010:128 cited in Cui and Zhao, 2014), it is the latter which is relevant. As Cysouw (2002:45) explains, “A minimal inclusive is a category that includes the present speaker and addressee (i.e. inclusive) but nothing else (i.e. minimal)”. Table 2 below summarises the relevant first-person pronouns for English and Swedish.
Table 2. Plural first-person personal pronouns in English and Swedish
personal possessive reflexive
EN/SE subject object dependent independent
1st person
plural
EN we us our ours ourselves
SE vi oss vår, vårt, våra oss
Note. Data for English from Eppler & Ozón (2013:24 table 2.1) and for Swedish from Larsson (n.d.) and Smitterberg (2013)
Since each language uses one lexeme for inclusive and exclusive meaning, only the context and syntax can help determine which meaning it has. In other words, the collocation of two or
20
more lexical items can help determine the pronoun’s identity. For example, if a verb of advice such as recommend is used after the plural personal pronoun we, it is clear that the pronoun is exclusive because the recipient is being advised (the recipient is the object of the verb).
However, even the ‘inclusive’ can be unclear. The pronoun could be referring to all humans or just the sender and receiver, but again, the context, as Estling Vannestål’s (2007:357) example below shows with the adverbial ‘In Sweden’, can help identify the meaning (the pronoun’s clusivity).
(7) In Sweden we dance around the midsummer pole.
I Sverige dansar man/vi…
However, just as the ambiguous nature of the English you and the fact that it covers all recipients, it perhaps does not matter that the inclusive we is also ambiguous. The greater inclusivity includes the lesser inclusivity, and so the reader is included either way. The reader is likely to interpret it as referring to them.
In summary, first-person inclusive plural pronouns together with the second-person pronouns mentioned in 3.1.1 constitute the second-person referencing features of the term “direct reader address”. The final feature involved is detailed in the next section.
3.2.3 Imperatives
An imperative is a type of grammatical mood which signals a command or request (Eppler and Ozón, 2013). It is therefore typical for giving directions, one of the communicative functions of a guidebook. It is a natural component of texts with a conative function since the imperative requires a receiver – the imperative used by the sender is commanding/requesting something of someone, the receiver – and thus a very relevant linguistic feature for this study.
The imperative in English varies from Swedish in that it does not have any inflection forms (Eppler and Ozón, 2013:74). However, this is not relevant for this study, as the focus is only on whether it is being used, not the form it takes. In this respect, there is little difference between the two languages, and both only use it in the present tense (in some languages, such as Latin, there are other tenses too) (Gill, 2018).
21
Typically, in an imperative statement, there is a lack of a subject, but when the subject is included as in Eppler and Ozón’s (2013:201) example below, the result is a more emphatic tone:
(8) You(,) pipe down.
In summary, the feeling of being spoken to, provided by the use of second-person pronouns is amplified with the inclusion of imperative use because the recipient is expected to respond (to visit the sight, to try the food, to turn left at a certain point).
3.2.4 Not addressing the reader directly
Investigating to what extent “direct reader address” is used and retained in a text, which includes the examination of source text and target text correspondences, results naturally in needing to be aware of what language is used when a writer chooses not to address their reader directly. Although a guidebook writer may not want to address the reader directly, they will still have to fulfil the purpose of informing and giving directions. Alternatives to using
“direct reader address” features could be the use of impersonal pronouns such as one in English and man in Swedish, the use of certain “people” nouns, e.g. “visitors”, or the use of the agentless passive (Altenberg, 2004:94).
Using an impersonal pronoun avoids specifying a particular person. This explains why the English second-person pronoun you, an ambiguous pronoun, as pointed out in section 3.2.1, is also considered an impersonal pronoun. English also uses the pronoun one for generic person referencing, but, these days, it is considered very formal, particularly in spoken language (Carter et al, n.d., Estling Vannestål 2007:357). The Swedish counterpart to English one is the pronoun man. Unlike the English one, the generic man is very common in Swedish, and is often translated into you (Altenberg, 2004), as example (9) shows:
(9) En så mycket ”finare” brygga att bada från hittar man nämligen inte i Skåne.
(Sandbring, H. and Borg, M. 2006a:22)
You will not find a better jetty to swim from anywhere in Skåne. (Sandbring, H. and Borg, M. 2006b:20)
However, man is not the only neutral pronoun used in Swedish. According to Dahl (2013), man also has a number of alternatives in some areas of Sweden, such as en, han ‘he’ and du
22
‘you’, the latter being ascribed to the influence of English. Moreover, as part of the debate of only using gender-neutral pronouns, the word hen has now entered the language as an
alternative to man (ibid.). Whichever is used though, the function is the same, to provide a generic sense rather than specific. English does not have an equivalent to hen, but plural pronouns such as they and them are becoming increasingly more common with gender-neutral options becoming popular alternatives (Estling Vannestål, 2007:360). They and even people can also be useful vague words in English to generically refer to those not nearby in time and space, similarly, the case of the English generic phrase those who and the Swedish equivalent de som.
As a less formal option, just like English you, Swedish du is commonly used for advice and instructions (Estling Vannestål 2007:357; Smitterberg, 2013) – appropriate for current study – which was also indicated in section 3.2.1 in relation to advice given to authorities about which pronoun to use. The plural second-person ni in Swedish could also function generically, since it can indicate many rather than one specific person, and in that sense is similar to the English you. The inclusive pronouns Swedish vi and English we could also be alternative impersonal pronouns since they could encompass everyone. However, if any of these are used, the reader is likely to interpret them as referring to them directly since they are the receiver and these words encompass them. Therefore, should a writer wish to avoid addressing the reader directly, man and one would be the better options.
However, as mentioned, English one is often considered too formal and therefore, it could be more apt for the writer to use the passive voice, which is a useful linguistic structure when the person doing the action is not known, common in academic writing where the person tends to be irrelevant with the focus on processes and results instead (Estling Vannestål, 2007:446):
(10) [ibid.]
The study was performed in 1984-1985 on a series consisting of the total population born in 1923 or earlier.
A guidebook writer is unlikely to know exactly who they are informing or providing directions for, so there would naturally be a generic sense. According to Estling Vannestål (2007:358), it is more common for sentences containing man to be translated into passive sentences to avoid the use of the formal one.
23
In summary, “direct reader address” is an umbrella term for this study which applies to linguistic features used by writers to address their recipients directly, and this term will now be acknowledged and used as an approved term for the remainder of this thesis. The term includes second-person referencing features second-person personal, possessive and reflexive pronouns and first-person inclusive plural pronouns and verbs in the imperative mood. As stated though, due to practical constraints, only second-person personal pronouns, the first- person inclusive plural pronouns and imperatives will be included in the following analysis.
4 Quantitative and Qualitative Analysis
The following chapter incorporates the idea and concepts from the previous chapter into the results of this investigation. The empirical results are presented in the first part together with a few specific examples. The second part of the chapter analyses the personal translation in relation to the results of the corpus results, observing whether the TT is successfully an instrumental translation and typical of an original English guidebook. This will be done through an examination of some of the correspondences of direct reader address in the ST and TT. The full results – instances and frequencies – of the corpus and personal translation can be accessed in Appendix B. Appendix C provides the chi-square test results.
4.1 Direct Reader Address in the Corpus
The total number of direct reader address instances per 1,000 words (the standardised frequencies) in each of the three subcorpora – ENGLISH ORIGINALS, SWEDISH ORIGINALS and ENGLISH TRANSLATIONS – are shown in Table 3. Direct reader address occurs most frequently in the subcorpus ENGLISH ORIGINALS (33.5).
Table 3. Direct reader address standardised frequencies (instances per 1,000 words) in each of the subcorpora.
FREQUENCY
ENGLISHORIGINALS 33.5
SWEDISH ORIGINALS 17.2
ENGLISHTRANSLATIONS 17.6
The difference between the occurrence frequency in ENGLISH ORIGINALS and the others is considerable; double the amount. This would indicate that direct reader address is common in English walking guidebooks in comparison to Swedish ones. The almost identical frequency
24
of the occurrences in SWEDISH ORIGINALS and ENGLISH TRANSLATIONS suggests little change has taken place during translation; this could be viewed as a sign that the translators remained loyal to the STs, being rather word-for-word, and consequently the translations must be more documentary than instrumental (Nord, 2005). With half as many occurrences of direct reader address in ENGLISH TRANSLATIONS compared with ENGLISH ORIGINALS, the evidence suggests that translators focus on producing a ST-close rendition rather than original-language-like text. The fact that the translations are more similar to the source texts than originals in the same language is supported by a chi-square test of independence, which shows the
disproportionate correlation between ENGLISH TRANSLATIONS and ENGLISH ORIGINALS to be significant at the p < 0.00001 level (see Appendix C for details). This is further supported by the similar occurrence frequencies between the source and target texts (SWEDISH ORIGINALS
and ENGLISH TRANSLATIONS) not being significant1. To gain a more comprehensive insight into why this might be, an examination of some Swedish and English correspondences can help. However, before that, it is necessary to look further into the frequency results for the individual features of direct reader address in each of the subcorpora as well as the results of direct reader address for individual texts within the whole corpus.
Figure 1, below, distinguishes between the different linguistic features that constitute direct reader address in this study in each of the subcorpora: first-person inclusive plural pronouns (1st PP), second-person pronouns (2nd PP) and imperatives (IMP).
Figure 1. Occurrence frequency (per 1,000 words) of each linguistic feature of direct reader address in the three different subcorpora.
1 p < 0.703301
0 5 10 15 20 25 30
ENGLISH TRANSLATIONS SWEDISH ORIGINALS ENGLISH ORIGINALS
Occurrence Frequency
Subcorpora
1st PP 2nd PP IMP
25
Of the three features, imperatives occur frequently in all the subcorpora, but are particularly abundant in ENGLISH ORIGINALS (26.1), just as was the case for this subcorpus having the most overall direct reader address occurrence frequency. However, whereas the occurrence frequency of direct reader address is double the others, the frequency of the imperatives is approximately three times higher than in the other two subcorpora. It is therefore reasonable to say that as well as direct reader address being very common in English walking
guidebooks, it would also seem that imperatives are the preferred choice for addressing the reader directly, perhaps not unexpectedly for a text that gives on-foot directions to a destination. In contrast, Figure 1 also illustrates that although second-person pronouns are used slightly more in SWEDISH ORIGINALS (8.9 vs 7.9) and imperatives more in ENGLISH TRANSLATIONS (9.3 vs 8.0), the differences are minor and the picture suggests to a greater extent that there is more of a balance of the two linguistic features.
For all subcorpora, the use of first-person inclusive plural pronouns is extremely low, almost non-existent, indicating that the focus is more directly on the reader than on having the writer along for the journey in the imagined dialogue (Torresi, 2010:128 cited in Cui and Zhao, 2014). The highest frequency of these features occur in the translation of a Malmö city guide (Agrelin, 2007b). This is quite interesting as its source text contained no instances of direct reader address at all, suggesting that the translator felt some interaction was necessary in the English, that the genre required it, though not a specifically chosen recipient indicator like you.
In ENGLISH ORIGINALS, personal pronouns do not seem to be the “pivotal verbal items” as Francesconi (2011) had observed in her tourist brochure, at least not explicitly. Imperatives are a form that implicitly contain a second-person personal pronoun (see section 3.2.3).
However, whether the features used are imperatives or pronouns, there is still a high
frequency of direct reader address, indicating that “ego-targeting” (Dann, 1996) is transpiring.
In SWEDISH ORIGINALS, where second-person pronouns are used more often than imperatives (8.9 vs 8.0), the realisation of these are almost completely in the form of the singular du. Only in Lindberg (2015a) do the plural versions appear (and curiously intermixed with the use of the singular).
26
Turning to the occurrence frequencies of each text in the corpus, there is clear evidence of direct reader address being prevalent in this study’s genre focus. Figure 2 below illustrates clearly how many of the texts in ENGLISH ORIGINALS have a high frequency of direct reader address (many have over the whole subcorpora’s total frequency of 33.5).
Figure 2. Occurrence frequency (per 1,000 words) of direct reader address in 10 ENGLISH ORIGINALS.
As mentioned previously, imperatives are the preferred linguistic choice in this corpus, and do constitute the majority of the direct reader address figures. In fact, Buckley and Buckley (2007) only have instances of imperatives (63 instances – 57.4 frequency per 1,000 words).
Nine out of ten texts have higher occurrence frequencies of imperatives than second-person pronouns (see appendix B for details), though one of these (Duncan, 2006) has much more of an equal distribution of imperatives and second-person referencing (13.2 to 12.3 respectively).
The exception is Quinn and Felix (2007), which actually has a slightly higher frequency of second-person references (9.6 vs 6.2), but their overall frequency of direct reader address is also quite low in comparison to many others (16.5) (though this is fairly similar to the overall result of the other two subcorpora – SWEDISH ORIGINALS (17.2) and ENGLISH TRANSLATIONS
(17.6)). These different results will have affected the high imperative frequency (26.1) in the English Original subcorpora overall, but this is a consequence of the small size of the corpus.
A different impression is given below in Figure 3. The bar chart presents the occurrence frequencies of direct reader address in SWEDISH ORIGINALS and ENGLISH TRANSLATIONS.
0 10 20 30 40 50 60 70
Berry (2012) Clark et al (2016) Turnbull (2015) Wilson (2014) Buckley and Buckley (2007) Duncan (2006) Filou et al (2016) Viccars (2013) Hancock and Foster (2014) Quinn and Felix (2007)
Occurrence Frequency
English originals
