T/V pronouns in global communication practices: The case of IKEA catalogues across linguacultures * Journal of Pragmatics

In this paper we investigate how the second person pronominal T-form is translated in IKEA catalogues in a number of different languages. IKEA is renowned for using the T-form as a form of branding: it promotes this form even in those countries where it might not be perceived favourably. However, our examination of a sample of IKEA catalogues shows that there are frequent deviations from IKEA's T-policy. By examining translations of the T-form in IKEA catalogues, and language users' evaluations of the (in)appropriacy of these translations, we aim to integrate T/V pronominal research into the pragmatics of trans- lation, by demonstrating that the study of the translation of seemingly ‘ simple ’ expressions, such as second person pronominal forms, can provide insight into an array of cross- cultural pragmatic differences. The study of translation in global communication is also relevant for research on the pragmatics of globalisation. course of the interviews, we presented excerpts from the catalogues to our interviewees in their native languages. We then conducted semi-structured interviews by asking 4 questions to obtain information, not only on the perceived (in)appropriacy of the T/V choice, but also about local norms of T/V usage in the linguaculture of each interviewee (for details see Section 3.3). We do not believe that interview corpora such as ours are representative of entire linguacultures. However, interviews of this type can substantially enrich our interpretation and evaluation of the translational choices being made in the various IKEA catalogues, and the relationship between these choices and local norms. the Translation and Intercultural include contrastive pragmatics, discourse analysis, politeness research and as a global language. include as Communication across Languages and The


Introduction
In this paper we investigate how the second person pronominal T-form is translated in IKEA catalogues in a number of different languages. IKEA is renowned for using the T-form as a form of branding (cf. Fennis and Wiebenga, 2012): it promotes this form even in those countries where it might not be perceived favourably. IKEA's choice of the 'T-policy' (Norrby and Hajek, 2011) in its branding may be due to the fact that the T-form is associated with positive and egalitarian values in Scandinavian cultures (Hellan and Platzack, 1999;Norrby et al., 2015). However, an examination of the translational choices in IKEA catalogues reveals some interesting deviations from the company's stereotypical T-policy: translated IKEA catalogues do not use the T-form unanimously. By examining translations of the T-form in IKEA catalogues, we aim to integrate T/V pronominal research into the pragmatics of translation, by demonstrating that the study of the translation of seemingly 'simple' expressions, such as second person pronominal forms, can provide insight into an array of cross-cultural pragmatic differences.
The study of translation in global communication is relevant to a recent, important body of research on the pragmatics of globalisation. Globalisation has been studied from a number of angles in sociolinguistics and pragmatics, and other related areas such as globalisation and English as a lingua franca (see e.g. House, 2003;Phillipson, 2008), language education (Byram and Parmenter, 2012), attitudes (D€ ornyei et al., 2006) and language change (Meyerhoff and Niedzielski, 2003). Yet, little research has been conducted on the impact of globalisation on the translational choices of particular linguistic forms. While scholars such as Sifianou (2013), Garc es-Conejos Blitvich (2018), Perelmutter (2018), and Vladimirou and House (2018) have studied the pragmatics of globalisation, translational choices have received little attention in this body of research (an exception is House, 2017).
In this paper, we also investigate the linguacultural perceptions of the (in)appropriacy of the translational choices that we are studying. This investigation is relevant to research on language and globalisation because it reveals the operation of globalisation beyond the macro-level understanding of this phenomenon. A key area of linguistic research has criticised the effect that globalisation has on different local languages, markets, social hierarchies and political systems (e.g. Coupland, 2003;Fairclough, 2006;Blommaert, 2010;Phillipson, 2008;Pennycook, 2009). While some previous research (see e.g. various studies in Coupland ed., 2010;and Sifianou, 2010) have engaged in a micro-level pragmatic analysis of the effect of globalisation on language use, the majority of previous research in this area has pursued a macro-agenda. We believe that further micro-level research is needed on the pragmatics of globalisation, in order to avoid overgeneralising the effect of globalisation on language use.
In this paper we use the term 'linguaculture' to indicate that our research focuses on the language-culture nexus in a transnational context (cf. Risager, 2014). The term 'linguaculture' has also been frequently used in translation studies (see e.g. House, 2016), and as such it is highly relevant to our current investigation.

Objectives
We propose a bottomeup corpus-based contrastive pragmatic approach to translational choices in global communication, by investigating the impact of global communication on expressions that are (meta)pragmatically significant. 2 Our approach is based on the assumption that there are certain expressions in every linguaculture, which are associated with standard situations (K ad ar and House, 2020a, b). As House (1989:115) stated: "The notion of a standard situation involves participants' rather fixed expectations and perceptions of social role." In both institutionalised and noninstitutional standard situations, including service encounters, small-talk between friends, and so on, language users have fixed expectations of rights and obligations and related conventionalised language use. In a pragmatic sense, it is often expressions that indicate language users' awareness of the standard situation holding for a given context. Exactly because of this, if language use changes in a particular standard situation, this change triggers awareness and reactive feeling.
Pragmatically salient expressions are particularly useful for the study of translational choices and the related perceptions of translational (in)appropriacy in global communication because such expressions migrate across the world in a rather straightforward manner (see Baumgarten and House, 2010). Take, for instance, please and its translational equivalents in the service industry. As we have found in previous research (K ad ar and House, 2020a), please e which is a typical expression that is pragmatically salient e appeared in service encounters in many linguacultures. Chinese is a prime example of how this expression has spread in non-English speaking countries. In Chinese, for instance, the equivalent of please, qing 请, had, until recently, been rarely used as an expression in service encounters e e.g. during interactions in restaurants e and has only recently gained popularity mainly due to the influence of globalisation and the supposed dominance of English as a global language.
In this study, we capture how a particular inventory of expressions operates in global communication. We examine the variation of second person 'T/V' pronominal forms in localised e i.e. covertly translated (House, 2015) e marketing materials, namely, IKEA catalogues. A covert translation is one in which the original text is adapted to the discourse norms of the receiving linguaculture, via a so-called 'cultural filter'. It will be interesting to investigate whether cultural filtering is present in IKEA catalogues despite IKEA's 'T-policy', i.e. whether these catalogues make linguacultural adjustments to local norms (see Section 3.3). If cultural filtering can be observed in IKEA catalogues across different linguacultures, then what does it look like? Also, how do locals evaluate the practice of cultural filtering? Our aim in this study is to pursue the answers to these two questions (see more in Section 2.3).

T/V phenomena: A brief literature review
In order to position the methodology deployed in this paper, it is first necessary to provide a brief overview of previous research on T/V pronominal forms.
Many languages feature a formal (V, from the Latin vos) and an informal (T, from the Latin tu) pronominal form, and in various languages there are more than two second person pronouns (see Braun, 1988). Even in languages such as Japanese and Korean e which do not operate with narrow-sense pronominal forms e pragmatic norms still regulate the use of quasisecond person pronominal forms, and these norms are centred on in/formality (Lee and Yonezawa, 2008). 3 T/V-forms are prime examples of pragmatically salient expressions as their unexpected use in a standard situation may trigger (strong) feelings centred on rights and obligations. When language users choose either the T or the V-form, they display information on who and where they are.
Previous work on T/V pronouns owes much to the classical work of Brown and Gilman (1960), who coined the term 'pronouns of power and solidarity'. They argued that social relationships are defined in a particular way by using either the T or the V-form: the T-form indicates solidarity and familiarity between speakers, while the V-form indicates power, authority and seniority. One of the most insightful early critiques of Brown and Gilman is Kendall (1981), who argued that the same pronominal form can acquire multiple meanings in a certain context, stressing the fact that any conventional meaning of a form can be manipulated to create new meanings. For instance, pronominal forms can be used creatively to indicate humour and sarcasm.
Later studies on address terms e in particular, the T-form e stressed the fact that it "enables the speaker to generalize and personalize at the same time" (Bolinger, 1979:207). Bolinger's distinction between the 'impersonal' and 'personal' uses of the T-form is especially relevant to our study. Bolinger notes that the pragmatic relationship between personal and impersonal T uses can be intricate: The deeper we go into impersonal you, the more personal it seems. If the reference is to a stage on which the speaker has trouble imagining himself, you is proportionately difficult e which is to say that you adopts the viewpoint of the speaker. (Bolinger, 1979:205) The assumption that the T-form can be used in different languages not only as an address form for specific addressees but also to present general information in an impersonal use of the T-form, has attracted much academic attention (Biq, 1991;Kamio, 2001;Stirling and Manderson, 2011;Helmbrecht, 2015;Gast et al., 2015;Hogeweg and de Hoop, 2015;Igaab and Tarrad, 2019). In our data such impersonal pronominal use in the different language versions of IKEA catalogues is likely to occur, given the nature of these texts. However, it should be noted that not all linguacultures afford the impersonal use of the T pronoun, as the seminal study of Kitagawa and Lehrer (1990) demonstrates: Although the use of the 2 nd person singular for an impersonal is widespread, not all languages permit such an extension. A partial pattern appears to be the following: The extension of a 2 nd person pronoun to an impersonal is possible only in languages with small, closed pronoun sets.
[This example] would place such languages as Chinese, English … among those possibly having recourse to impersonal use of the 2 nd person pronoun; these are all languages with a closed set of personal pronouns. It would place, on the other hand, such languages as Japanese and Korean among those having no recourse to impersonal use of the 2 nd person pronoun; neither Japanese nor Korean possesses a clearly defined closed set of personal pronouns. (Kitagawa and Lehrer, 1990:753) Even though we agree with Kitagawa and Lehrer in that languages such as Japanese do not, strictly speaking, have pronouns, they offer a wealth of pragmatic solutions to express personal versus impersonal quasi-pronominal references to the recipient. 4 To the best of our knowledge, no pragmatic research has been conducted on the use of T/V-forms in translated IKEA catalogues. The majority of existing studies on IKEA are concerned with the marketing and management aspects of this multinational company, and stress IKEA's promotion of Swedish corporate culture (Edvardsson and Enquist, 2002;Arrigo, 2005;Jonsson, 2007;Jungbluth, 2008;Rask et al., 2010;Norrby and Hajek, 2011). A number of dissertations compare IKEA catalogues in different languages (Blancke, 2007;Tesink, 2016;Lathifatul, 2018), but these texts have not analysed pronominal issues in detail.

Methodology
In this paper, we adopt a two-fold analytical approach. Firstly, we examine the ways in which T/V pronominal forms and their linguacultural equivalents are deployed in our corpus of translated IKEA catalogues (see Section 2.4 for details). This analysis aids our investigation into whether the use of second person pronominal forms in IKEA catalogues reflects IKEA's aforementioned T-policy. Secondly, we examine how language users evaluate the appropriacy of the use of second person pronominal forms in the translated versions of the catalogues, by studying interviews that were conducted with cohorts of native speakers of the languages into which the catalogues under investigation were translated.
Regarding the first analytical approach, we compare the use of T/V pronouns in these catalogues with that in the English version of the catalogue, with the latter serving as the 'basic text' (as IKEA terms it) for all covert translations (cf. House, 2015) into the 38 languages in which IKEA markets its products. IKEA's Swedish headquarters produces this 'basic' English text and, obviously, only the pronominal form 'you' is used in this version. The English you is neither T nor V, but considering IKEA's 'T-policy' it is much more likely to be interpreted as T than V. By examining IKEA catalogues, we focus on a corpus which represents a specific domain of language use, which implies that we can safely draw analogies between any of the pronominal forms that occur in our corpus. Thus, the T/V choices that are studied may include both languages with 'proper' T/V-forms and other languages, such as Japanese (see Section 2.2 above). Our current investigation includes both T/V IKEA catalogues, such as the German and Chinese one, and the Japanese IKEA catalogue. As our research will demonstrate, for personal use the Japanese catalogue deploys the honorific form of address okyakusama お客様 ([hon.] visitor, i.e. 'honoured customer') together with other honorific forms (see Section 3), whereas for impersonal use it applies anata あなた ('you'), which is a quasi-pronoun with an honorific origin. Note that anata is not frequently used in service encounters but may occur in public announcements and other written genres (cf. Wetzel, 2010).
In our study, we not only contrast the choice of second person pronominal forms throughout the translated IKEA catalogues, but also engage in a contrastive pragmatic analysis of the use of these pronominal forms in two main sections of the catalogues, namely, 1) general 'product description' and 2) 'instructions for customers'.
These sections represent two different standard situations, with different rights and obligations (see Section 2.1), and as such different default expectations towards T/V use. We assume that the 'product description' is impersonal in tone, i.e. if a Tform is used in this section in accordance with IKEA's policy, it has a more general pragmatic meaning. In contrast, the second section e in which IKEA instructs the customer on how to proceed with the purchase e is personal, i.e. it addresses the customer as an individual.
With regard to our second analytical approach, we conducted interviews with 8 native speakers of each of the following linguacultures that were represented in our corpus: Mainland Chinese Mandarin, Hong Kong Mandarin, Japanese, Hungarian and German (cf. Section 2.4). All interviewees were female and university educated. A total of 40 interviews were conducted which allowed us to examine the perceived (in)appropriacy of the pronominal choice in the translated IKEA catalogues (see House, 2018 on this form of interviewing). It is worth noting that these linguacultures were chosen for both pragmatic research reasons and because they represented the range of preferences as to how the T pronoun is translated in IKEA catalogues (see Section 3.1). Each language group of 8 interviewees was divided into 2 age categories (with 4 interviewees in each): 1) Age group 1: 18e35 years 2) Age group 2: 36e65 years This age classification was made in an attempt to determine whether a generational difference existed in the perception of the (in)appropriacy of the translational choices. During the course of the interviews, we presented excerpts from the catalogues to our interviewees in their native languages. We then conducted semi-structured interviews by asking 4 questions to obtain information, not only on the perceived (in)appropriacy of the T/V choice, but also about local norms of T/V usage in the linguaculture of each interviewee (for details see Section 3.3). We do not believe that interview corpora such as ours are representative of entire linguacultures. However, interviews of this type can substantially enrich our interpretation and evaluation of the translational choices being made in the various IKEA catalogues, and the relationship between these choices and local norms.

Corpora
Our research is based on two corpora: IKEA catalogues and interviews. The first corpus involves the latest (2019)  It should be noted that the Hong Kong catalogue uses Mandarin, even though the most commonly spoken language in Hong Kong is Cantonese. Unlike its spoken equivalent, written Cantonese is very similar to Mandarin, and most companies (including IKEA) use written Mandarin in their promotional material.
The second corpus is comprised of 40 interviews, with each interview being an average of 15 min in duration. Interviews were conducted on Skype/WeChat, and were audio recorded and then partially transcribed. In storing the audio recorded data, we adhered to the standard ethical processes of the Centre for Pragmatics Research, Research Institute for Linguistics, Hungarian Academy of Sciences. We adopted a simplified form of transcription when transcribing what we considered to be analytically significant sections of the interviews.

Data analysis
Data analysis is divided into three sections. In Section 3.1, we investigate the translational choices in terms of the pronominal forms in our catalogue corpus. In Section 3.2, we focus on two different translational strategies that are used in selected IKEA catalogues. Finally, in Section 3.3, we present the results of our interviews.

Analysing T/V choices in the two standard situations in translated IKEA catalogues
Some IKEA catalogues lack the ambiguity of the T-form: there are no personal or impersonal T uses in them because they only deploy the V-form, i.e. they openly defy the IKEA T-policy. This is the case in the Hong Kong Mandarin and Belgian French catalogues, as the following excerpts from the 'product description' sections of these catalogues show. Although we also examined the French version of the catalogue, we were unable to include it in our contrastive analysis because this catalogue is structured differently from its counterparts in our corpus, all of which have the same structure and content. Thus, the text used in the French catalogue does not afford a proper comparison.
In the following, any text in languages other than English will be provided with a 'back translation' (BT), which is an informative gloss for the reader unfamiliar with these languages.
The Hong Kong Mandarin catalogue (example 1) uses the V-form nin 您, and the Belgian French (example 2) the V-form vous consistently, in both standard situations in the catalogue. This finding is noteworthy in itself because the pronominal translational choice in these catalogues apparently flies in the face of IKEA's T-policy. The fact that some catalogues defy the IKEA T-policy becomes particularly evident when we compare the Hong Kong Mandarin catalogue with the Mainland Chinese version; the latter uses the T-form ni 你 only in the standard situation 'product description', as the following extract illustrates: Similarly to the Mainland Chinese and Hong Kong Mandarin catalogues, in the case of the Belgian French versus the Dutch/ Belgian Dutch versions, it is more likely that the choice of V and T-forms represents an alignment to the Dutch and/or Francophone linguacultures (see Martiny, 1996;Vandekerckhove, 2005). It is worth mentioning that the French catalogue consistently deploys the V-form, a translational choice which, in our view, reflects pragmatic preferences in the Francophone world. The following are excerpts from the Dutch, Belgian Dutch and Belgian French IKEA catalogues, featuring the same 'product description' section that we have examined thus far: Extracts (4) and (5), from the Dutch and Belgian Dutch catalogues, both use the T-form je. Example (6) e from the Belgian French catalogue e deploys the V form vous.
If we examine those catalogues that use only the T-form e i.e. Dutch, Belgian Dutch, German and Hungarian in our corpus e we discover a sense of unresolved ambiguity in the use of the T-form. That is, if we examine the standard situation 'product description', which represents the main part of the IKEA catalogue (even though some 'product descriptions' also include instructions for customers at the bottom of the respective pages), it is impossible to decide whether the T-form is used in a personal or an impersonal way. The standard situation 'product description' might never be completely impersonal because it is basically a persuasive advertisement addressed to the customer. However, given Bolinger's (1979) paradox described above, the impersonal nature of these T-forms makes them perlocutionarily even more personal (see Kramsch, 2020 forthcoming). Consider the following excerpts (of the same 'product description' featured in the previous examples) from the German and Hungarian catalogues, respectively: Note that the Hungarian version in Example (8) does not include a single use of the T-form. This is because, in Hungarian, T/V use can adroitly be expressed by inflection (see Domonkosi, 2010). The entire Hungarian catalogue is translated with T inflection. From the reader's perspective, there is no pragmatic demarcation between the impersonal address form used in the standard situation 'product description' and the personal address form used in the standard situation 'instructions for customers'. This lack of pragmatic demarcation is resolved in the Mainland Chinese and Japanese versions of the catalogue, as examples (9) and (10) below illustrate. Both these examples include the Mainland Chinese and Japanese versions of the 'product description' excerpt that we have analysed thus far, as well as an excerpt from the 'instructions for customers' section of the respective catalogues: The reader does not have the opportunity to interpret the T-form as personal in either the Mainland Chinese or the Japanese catalogues. This is because in the standard situation 'instructions for customers', in which the customer is addressed in a personalised way, the T-form is systematically replaced with either the V-form in Mainland Chinese Mandarin, or with the quasi-V-form and other honorific forms in Japanese. With regard to the Mainland Chinese version, the catalogue consistently deploys the V-form nin, in a similar fashion to the Hong Kong Mandarin catalogue, in the standard situation 'instructions for customers'. Thus, in this catalogue we can observe what various experts of East Asian languages, such as Cook (2008), describe as 'style shift'. The aforementioned pragmatic ambiguity created by the T-form in a number of catalogues is thus disambiguated in the Mainland Chinese version, as the two standard situations in this case are clearly demarcated in the translation. There is even less ambiguity in the Japanese version. The 'product description' uses anata, which is approximately equivalent to the T-form. It is worth noting that the Japanese catalogue deploys a translational strategy that we will analyse in Section 3.2 below: it attempts to use this T-form as little as possible. As far as the standard situation 'instructions for customers' is concerned, the Japanese catalogue not only stops using this T-form, but also suddenly switches to an honorific communication style. For instance, in example (10) above, the text uses the quasi-V-form okyakusama when referring to the customer, and also deploys other honorific forms, including otodokeshimasu お届けします ([hon.] 'respectfully deliver') go-kibou-no-shoohin ご希望の商品 ([hon.] 'respected chosen product') and go-jitaku ご自宅 ([hon.], 'noble dwelling').
The analysis so far has demonstrated that, in some IKEA catalogues, the two standard situations may be clearly demarcated if both the T and V-forms are deployed. Certain catalogues e such as the Mainland Chinese and Japanese versions e appear to be skillfully designed to downtone and thus neutralise the effect of IKEA's T-policy in instances where this policy may go against local linguacultural norms. In other words, translational choices are made in order to culturally filter the translated text and, potentially, make it more acceptable to local conventions. In the following section, we will further analyse this phenomenon. We will devote particular attention to the strategies that the Mainland Chinese and Japanese IKEA catalogues deploy, to model the translational complexities that surround the cultural filtering of IKEA's policy in these linguacultures. In our view, the Mainland Chinese and Japanese IKEA catalogues represent a 'third way' to avoid both the strict renunciation of the T-form and the complete adoption of this culturally alien form. It should be noted that, in the following analysis, we will only contrast examples of Mainland Chinese and Japanese (T/V use) with Hong Kong Mandarin (V use), English, German and Hungarian (all T use).

Strategies for translating T in Mainland Chinese and Japanese IKEA catalogues
The Mainland Chinese IKEA catalogue and, to a greater extent, the Japanese version both deploy two types of translational strategy. In our interpretation, the goal of these translational strategies is to avoid using the T-form in the standard situation 'product description'. Our data reveals that the Mainland Chinese catalogue adopts various deviations from how the English 'basic' text deploys 'you', in an attempt to make the frequent use of the T-form ni more acceptable to native readers. While in the Chinese linguaculture the T-form ni is preferred in certain contexts (Kuo, 2006), in marketing contexts it can appear overly intimate, as the Mainland Chinese IKEA catalogue reveals. The Japanese text shows even more drastic translational solutions, possibly because few Japanese companies use the quasi-T-form anata when addressing their customers; this form would be acceptable in many settings but certainly not in marketing practices (Kuroshima, 2010; see also example (21) in Section 3.3).

Translational strategy no. 1: Avoiding the T-form
Two catalogues in our corpus e namely, the Mainland Chinese and the Japanese ones e afford the translational strategy of avoiding the T form e presumably in an attempt to counter possible objections to the T form. In Mandarin Chinese, this strategy can be neatly implemented because, in Mandarin, clauses can often be adeptly formed without the use of pronouns. It is worthwhile comparing the Mainland Chinese and the Hong Kong Mandarin equivalents of the following excerpt from a product description: While the Hong Kong Mandarin version of this excerpt consistently uses the V nin form (this is in accordance with the overall practice of the Hong Kong Mandarin catalogue, see Section 3.1), in the Mainland Chinese text the T-form ni is not used at all. In total, 17 of the 95 product descriptions display such a discrepancy between the Mainland Chinese and Hong Kong catalogues. Note that when the Mainland Chinese catalogue uses this strategy of avoiding the second person pronominal form, the translation does not become 'vague' because the text clearly addresses the reader, even if it does not explicitly use the T-form. For instance, the phrase suixin-changwo 随心掌握 lit. 'follow your heart to decide on the pattern' in the Chinese text above represents a clearly personalised instance of language use. However, it should be noted that the removal of the T-form often coincides with a sense of ambiguity, as the analysis of the second translational strategy below will illustrate.
The Japanese text also adopts the translational strategy of T-avoidance described above. The quasi-T-form anata tends to be avoided in the Japanese catalogue: it is used only 6 times in the entire catalogue to indicate the standard situation 'product description'. Thus, a major discrepancy exists between the Japanese and English texts. In the 'basic' English text, the T-form you is used 299 times in the 'product description' section. Note that even the quasi-V-form okyakusama is used more frequently e 16 times e than the quasi-T-form anata, in the 'instructions for customers' section of the Japanese catalogue.
The following excerpt e the Japanese version of example (12) above e illustrates this translational strategy: What one can observe regarding the strategy of T-form avoidance is that the Japanese translation prefers the plural imperative mashou ましょう, as can be seen in example (13) above, which makes the text less individualistic in style. The reason why the imperative fulfils such a deindividuating pragmatic function (see e.g. Garc es-Conejos Blitvich, 2015) in the Japanese linguaculture is that companies and vendors tend to use honorifics towards their customers, and so the imperative implies that the text is not addressed to a specific customer.
In sum, avoiding the T-form altogether is an obvious strategy to decrease pragmatic ambiguity. However, this translational strategy is not an option for many languages. For instance, the following English, German and Hungarian equivalents of the above studied excerpt (see examples 11 and 12) illustrate that neither of these catalogues avoid using the T form or related inflection, despite the fact that they could deploy impersonal pronouns: The Hungarian translation of the above excerpt is even more heavily T-loaded than its English and German counterparts, due to the aforementioned fact that, in Hungarian, inflection (e.g. v alassz e BT: 'you choose [T]') is bound to T/V use.

Translational strategy no. 2: Making the product description vague
The Mainland Chinese and Japanese translations of the catalogue tend to not only remove the T-form, but also often make the product description vague, in that the customer e who is being targeted by a particular product e is not mentioned at all. In the previously analysed strategy of removing the T-form, the reader is addressed indirectly. However, in 8 cases in the Mainland Chinese catalogue and 19 cases in its Japanese counterpart, the standard situation 'product description' becomes completely 'reader free' in the sense that the translated texts depict an abstract scenario in which the marketed product is foregrounded without the customers themselves ever entering the picture. This, in our view, represents a translational strategy of making the product description vague. The contrastive analysis of the Mainland Chinese and Japanese versions of an excerpt and their English, German, Hungarian and Hong Kong Mandarin counterparts illustrates this point: The Mainland Chinese and Japanese version of this excerpt in example (14) are void of personal references to the customers, in the sense that they do not address the customers even in an indirect way. In contrast, the Hungarian, German and Hong Kong Mandarin texts follow what we have previously observed: the Hungarian text uses the T-inflection and its German and Hong Kong Mandarin counterparts consistently use the T du and V nin. For instance, while the Hong Kong Mandarin text formulates the initial part of the message as Buyong banjia, ye neng chuangzao da-de shenghuo kongjian! 不用搬家,也能創造更 大的生活空間! (BT: 'There is no need [for you] to relocate, it is also possible to create a larger living space even without that'), the Mainland Chinese text uses the words Heli-guihua, bu banjia ye neng yongyou gengduo kongjian. 合理规划,不搬家也能拥有 更多空间。(BT 'It is a reasonable plan to create more living space without moving home.'). The Japanese text is equally impersonal, and e as a recurrent pattern in the translated catalogue e it even changes the structure of the information, by foregrounding information about the product that IKEA is proposing to the customer. This foregrounding makes the promotional activity even less personal, as it changes the standard logic 'the customer wants x' / 'product y fulfils this wish' of the narrative.
In this section, we have examined the translational strategies used by the Mainland Chinese and Japanese catalogues to cope creatively with IKEA's T-policy, which might be at odds with the pragmatic norms of these linguacultures. By applying a cultural filter (House, 2015), the translations make the text more palatable to local readers while at the same time maintaining a certain 'IKEA style'. For instance, while there are few instances of anata in the Japanese IKEA catalogue, the small number of this form that do exist differ from the default T-style of Japanese marketing practices (see also our interview corpus in Section 3.3 below). As such, the few anata forms that the Japanese catalogue deploys are an indication of IKEA's global T-policy. The analysis has illustrated that certain catalogues do not simply refuse to use the T-form e like the Hong Kong Mandarin and Belgian French catalogues e but rather absorb the T-form by employing a cultural filter. This points to the fact that the relationship between global communication and localised language use is more complex than meets the eye. In the following section, we investigate the evaluations by language users regarding the (in)appropriacy of the T/V translational choices in the catalogues being studied.
The following Table 1 summarises the two strategies studied in this section, applied by the Mainland Chinese and Japanese catalogues:

Language users' evaluations of T/V forms in IKEA catalogues
For the interview section of our study, we recruited cohorts of expert speakers e 8 speakers in each cohort e belonging to two different age groups in the following linguacultures: Mainland Chinese, Hong Kong, Japanese, Hungarian and German. The selection of the interviewees' linguacultural backgrounds was due to a) pragmatic research reasons (we had personal acquaintances in these cultures who in turn recruited further interviewees), and b) the fact that these linguacultures were of particular importance in our contrastive analysis of the IKEA catalogues. As has been shown in Section 3.2, catalogues that have been translated into these languages cover the whole spectrum of translational choices that have featured in our IKEA corpus. During the 40 interviews that were conducted, participants were shown 3 excerpts which displayed the translational choices of pronominal forms from the same two parts of the catalogue that were previously examined. Subsequently, the interviewees were asked the following 4 questions: 1) Do you find the choice of pronouns in this catalogue to be appropriate? 6 2) If yes, why, and if not, what is the problem with their use? 3) Is this use of second person pronoun common in marketing materials in your culture? 4) Can you elaborate on this?
Questions 1 and 3 are simple 'yes/no' questions, which provided us with quantitative evidence, whereas Questions 2 and 4 are open-ended questions that provided data for the ensuing qualitative analysis. Interviews were conducted in English and sensitive information about the identity of the interviewees was removed from the data. Table 2 provides a summary of the responses given to Questions 1 and 3: The table reveals the following: 1) There are significant differences in the way in which German and Hungarian interviewees and their Hong Kong, Mainland Chinese and Japanese counterparts responded to Question 1. Many Germans and Hungarians e 5 of the 8 Germans and 7 of the 8 Hungarians e felt that the use of the T pronoun in the catalogue was inappropriate. This is in stark contrast to the other 3 cohorts of interviewees in the corpus. In response to Question 3, 6 of the 8 Germans and 5 of the 8 Hungarians felt that the pronominal use in the respective catalogues was not in accordance with the general marketing conventions of their countries. Whilst the size of our interview corpus is somewhat modest and, as such, cannot 'represent' the entire population of the linguacultures being investigated, the differences between the evaluative tendencies in the dataset are Table 1 Summary of the translational strategies in the Mainland Chinese and the Japanese IKEA catalogues.

Strategy Description
Avoiding the T-form Avoiding the T form & indirectly addressing the customer Making the product description vague Not mentioning the customer at all & Making the description abstract still noteworthy: these tendencies indicate that the translational choice of the T form in the respective catalogues is problematic for both Germans and Hungarians, but is less controversial for the interviewees in the case of the East Asian catalogues. The responses given to Question 2 reveal that Germans and Hungarians provided a negative T-form evaluation because they felt that IKEA's use of the T form was intrusive. As an example, let us refer to the response given by a German interviewee: In response to Question 4, a number of German and Hungarian interviewees rationalised their negative evaluations by arguing that IKEA's T-style is somewhat 'jarring' and 'foreign'. In the following excerpt, the same German respondent expresses this view: It is worth considering the sociopragmatic reasons behind such negative linguacultural evaluative tendencies. In Germany, the pronominal style of the service sector has traditionally been the V-form, and even IKEA used this form when it first opened stores in Germany in 1974. IKEA only changed its translational choice of the V-form in 2005. 7 In Hungary, IKEA first opened stores in 1990, immediately after the end of Communism, and the company used the T-form right from the beginning of its operation in that country. However, this policy has met with resistance (see also Cameron, 2003), as the following newspaper extract illustrates: Even after 29 years operating in Hungary, strong feelings continue to surround IKEA's T-policy, as Ildik o's response below to Question 2 illustrates: In response to Question 4, another Hungarian interviewee, Enik} o e in keeping with other German and Hungarian interviewees e rationalised her evaluation by arguing that the spread of the T-form in the business sector is part of the world-wide colloquialisation and personalisation of language use. Such folk-theoretical comments (see K ad ar and Haugh, 2013) are, interestingly, in accordance with what scholars such as Leech et al. (2010) have discovered about the global use of English. Our East Asian respondents were a great deal more satisfied than their European counterparts with the translational choices in their respective IKEA catalogues. Question 1 triggered an unanimous positive evaluation. In response to Question 3, only 1 of the 8 Hong Kong and 2 of the 8 Mainland Chinese interviewees felt that the translational choices made by the respective catalogues were different from what they were accustomed to in their own linguaculture. For instance, Linglang from Mainland China noted the following: A surprising number of Japanese respondents, 7 out of 8, felt that the pronominal choices of the Japanese catalogue deviated from the general style of Japanese marketing. These evaluations are centred on the infelicity of the quasi-pronoun anata, as Noriko's response below illustrates: In summary, our analysis has so far revealed that the translation of seemingly 'simple' expressions, such as pronominal forms, can contribute to fundamental differences in the evaluation of the (in)appropriacy of IKEA catalogues.
2) In addition to linguacultural variation in the evaluation of translational choices, there also seems to be generational variation in these evaluations. In terms of generational differences, it would appear that in both the German and the Hungarian cohorts, it is members of the older generations who are more dissatisfied with the use of the T-form in IKEA catalogues. In other words, younger interviewees appear to be more tolerant, at least as far as our rather small corpus is concerned, as Judit's response below to Question 2 illustrates: It is particularly relevant to note that all the older German respondents in our small corpus evaluated the T-use to be different from general German marketing conventions, while their younger compatriots were divided on this point. The aforementioned difference between German and Hungarian catalogues on the one hand, and their Hong Kong Mandarin, Mainland Chinese and Japanese counterparts, on the other hand, is also true of the evaluations given by respondents from different generations. However, as Table 2 illustrates, a noteworthy tendency occurs in the Mainland Chinese interview corpus. Interestingly, in this cohort, members of the older generation felt that the catalogue's T-style is compatible with the norms of Chinese marketing, whereas younger people were more divided on this issue. An excerpt from Meifang's response to Question 4 illustrates this point: We have chosen this interview excerpt because Meifang's response provides a sociopragmatic insight into a generational issue in Mainland China where, following economic advancement, the use of the V-form in services has become more widespread (cf. Pan, 2000).
Note that while the present paper is anchored in contrastive pragmatics and the pragmatics of translation, the above findings also have implications for historical pragmatics and its interface with sociopragmatics. In historical pragmatics, extensive research has been dedicated to the relationship between major socio-ideological changes and the related change of the use of T/V forms in China, post-communist countries in Central Europe, and other areas (see K ad ar and Pan, 2011; see also various studies in Hickey and Stewart, 2005). It is likely that generational differences in our interview corpus are related to Ni is ideological changes, but it is beyond the scope and objective of our paper to further investigate this matter. Another historically-related area that might be considered in future research is the role of technologies such as Google Translate in the generational development of people's T/V pronominal preferences. Fig. 1 below provides an overview of the translational strategies that were adopted in the various IKEA catalogues studied: Several catalogues e including the German, Hungarian, Dutch and Belgian Dutch versions e simply adopt IKEA's T-policy. Another simple translational solution, that of openly defying IKEA's T-policy, can be seen in V-only catalogues (Hong Kong Mandarin and Belgian French). A 'third way' is used by the Mainland Chinese and Japanese catalogues, which adopt two different translational strategies in the two standard situations featured in the catalogue. This contrastive study of IKEA catalogues in different languages has thus shown that seemingly 'simple' expressions can be difficult to translate if they are pragmatically salient. In global communication, translation plays a major role, and cultural filtering is often employed to take account of local conventions. This is because the perceived inappropriacy of a pragmatically salient expression in a standard situation may trigger strong feelings, in particular if it perceived that a global player flouts the convention holding for the particular standard situation by 'importing' a foreign norm of language use.

Conclusion
In order to test the acceptability of the T/V translational choices in the IKEA catalogues studied, five cohorts of respondents were asked to examine excerpts from the catalogues. The responses have revealed that the translational strategies of resisting IKEA's T-policy and resolving it strategically were more favourably received than the translational choice of just accepting this policy, according to the corpora used in this paper. The T-policy is IKEA's global form of branding, and it would be easy to condemn it as 'colonising', as a body of previous discourse analytic research has done (see Section 1). At the same time, evaluations provided by the interviewees have revealed that such a negative description of the effect of globalisation in this context would be an overgeneralisation. While the T-form does indeed tend to be received critically by many in those cultures where this form is 'forced' on local customs, there appears to be an important generational difference in the evaluation of this translational choice. It is also worth noting that while many German and Hungarian respondents evaluated the T-form negatively, in linguacultures where the catalogues handle the T/V distinction in a more differentiated manner, evaluations were more centred on the 'foreignness' than the unacceptability of the T use.
Note that research on the acceptability of T/V uses in IKEA catalogues could be further investigated by providing respondents with manipulated examples of such uses. By experimenting with various types of example manipulation, we could obtain further information regarding linguacultural motivations for accepting or refusing certain translational strategies, in particular in the context of globalisation. In this study, we have only investigated the global marketing practices of one particular multinational company. It is interesting to note that many other global players, such as McDonalds, Lidl and H&M, follow a similar practice, as our preliminary research has confirmed. While the 'T-policy' is stereotypically associated with IKEA, it is in fact a policy adopted by many other multinational companies and, therefore, the research presented in this paper may be replicable. For instance, the two standard situations 'product description' and 'instruction to customers' exist, albeit in different forms, in the marketing practices of many other companies, such as Zara (e.g. in Hungary). It would be worth considering whether a generalised conclusion can be obtained with regard to the translational choices of pronominal forms across different multinational companies.
By adopting a two-fold analytical approach, we hope to have presented a new understanding of translational choices and their linguacultural perceptions in the context of global communication.