Addressing, Greeting and Related Gestures in the Opening Sequences of Finnish, French and Hungarian YouTube Videos

This paper compares the opening sequences of Finnish, French and Hungarian YouTube videos. We concentrate on addressing, greeting and related gestures, important elements when YouTubers interact with their imagined viewers, using data consisting of 138 videos in the three languages. This study falls within the field of the pragmatics of social media and interpersonal pragmatics, and data were analysed using multimodal discourse analysis. Shared practices included the frequent use of greetings, a preference for general nominal address forms as well as for iconic and deictic gestures. Cross-cultural differences revealed that Finnish and Hungarian were closer to each other than to French. Shared practices may be connected to the genre of YouTube videos, whereas differences appear related to cross-cultural practices generally.


Introduction
Although YouTube is the second most popular social media platform with more than 2.3 billion monthly users,1 it remains an unexplored field in linguistic studies, despite its suitability "from the perspective of visual and verbal expression" (Johansson 2017: 191). Eight of ten 18-49-year-olds watch YouTube, which can be navigated in 80 different languages.2 Thus, surprisingly, analyses of social media practices have focused primarily on other Internet platforms and English, whereas contrastive studies remain rare (cf. Ren and Fukushima, 2021). For instance, Ren and Fukushima (2021) studied requests in Chinese and Japanese on social media, Vecsernyés (2021) examined Finnish and Hungarian address practices in Facebook comments and Argüelles Alvarez and Muñoz Muñoz (2012) compared the use of Spanish and English on Twitter.
Our study concentrates on the opening sequences of YouTube videos comparing three languages: Finnish, French and Hungarian. The opening sequences in YouTube videos have received little attention (English: Frobenius, 2011;French: Combe Celik, 2014), despite presenting an interesting dataset to study, namely, YouTubers' interactions with imagined viewers. Here, we limit our research questions to addressing and greeting, which are important elements in the monologic performance of vloggers when interacting with their imagined audiences (Frobenius, 2011). Our research questions focusing on Finnish, French and Hungarian opening sequences are as follows: 1) How do YouTubers address and greet their imagined audiences?; 2) What kinds of gestures are related to addressing and greeting?; and 3) What cross-cultural similarities and differences appear in addressing, greeting and related gestures? We hypothesise that, although Finnish and Hungarian belong to the same language group (Finno-Ugric languages) and French does not, Hungarian lies closer to the French than the Finnish lingua culture (cf. address practices: Isosävi and Lappalainen, 2015). Our data contain 138 videos from 69 Finnish, French and Hungarian YouTubers. Our study falls within the field of the pragmatics of social media and interpersonal pragmatics, and we employ multimodal discourse analysis.
In section 2, we summarise previous studies on YouTube videos, while section 3 reviews the research on address forms, greetings and gestures. In section 4, we describe our data and the theoretical framework. Section 5 presents the analysis and results, and, in section 6, we discuss the implications of our findings.

Social Media and YouTube
Social media is part of computer-mediated communication (CMC), defined as "digitally mediated and Internet-based platforms" utilised to "exchange, share and edit self-and other-generated textual and audio-visual messages" (Hoffman, 2017: 4). Social media encompasses different Internet platforms ranging from blogs and discussion fora to Twitter, Facebook, YouTube and instant messaging services (Hoffmann, 2017: 6).
The key notion of second-generation (Web 2.0) Internet platforms is participation (Hoffmann, 2017: 5). Mandiberg (2012: 1) talks about "amateur media" whose sites "are pointless without audience participation". Thus, social media encourages people to "post, comment, evaluate, link and contribute selfselected […] content" (Hoffmann, 2017: 5). The participatory structure to social media differs from traditional platforms, since interaction is asynchronous between speakers and countless unknown hearers; social media has features of both face-to-face and mass media communications (see Chovanec and Dynel, 2015: 6-7).
The focus in CMC studies in general has shifted. The first wave of CMC research was "language focused" -that is, the effects of communications technologies on language were considered more important than contextual factors -whereas second-wave studies have highlighted situated language use (Androutsopoulos, 2008). Although studies seem to be more balanced in terms of technology and situated language use, little attention has been paid to variation across different lingua cultures.
Turning to the object of our study, YouTube is an extremely flexible online video-sharing platform launched in 2005, where ordinary users can upload multimodal content on various topics; the majority of users simply consume content (Johansson, 2017: 175-176). Consumers can subscribe to uploaders' channels and rate or comment on videos. Previous linguistic studies of YouTube videos primarily concentrated on English. For example, Frobenius (2011) studied the common features from vlog (defined in section 4) opening sequences and the strategies vloggers frequently employed to compensate for the absent audience. Frobenius (2014; also examined vloggers' shared tools to involve viewers and the role of pointing gestures (see section 3.4).
Finnish YouTube videos have primarily appeared in Master's theses studies. In French, multiple studies focused on vlogs. Vlogs were examined as a genre (Combe Celic, 2014), primarily from the perspective of activism and participation on YouTube. For instance, Combe (2020) studied a French vlog educating viewers on politics, Balleys (2019) examined French YouTubers using media visibility for sexual liberation and Caron's (2017) study focused on teenagers' vlogs in Canadian French. In Hungarian, Porkoláb (2015) studied status. Yet, no "one" social media exists. In some French discussion forums, T forms were preferred (Isosävi, 2015: 115, 118;Williams and van Compernolle, 2009), whereas, in others, V forms dominated (Havu, 2019). In Hungarian, T forms are common in public chats and forums where users do not know each other's background, whilst addressing in private online communication contexts follows similar rules as those beyond the Internet, whereby V forms can be used (Domonkosi, 2002: 166-167). On Facebook, informal T as well as formal V were found in both Finnish and Hungarian in comments to posts from prime ministers (Vecsernyés, 2021), and V was found in Hungarian also by Domonkosi (2018;.

3.2
Nominal Address Forms Nominal address forms appear more frequently in Hungarian and French than in Finnish. In Finnish, by contrast, politeness rules do not require mentioning the name of the person addressed, even in greetings or thanking (Seppänen, 1989: 198, 204). Yet, they are used in Finnish in imperatives and questions requiring a reaction (VISK, 2004 or in two opposing situations: ceremonial and intimate (Yli-Vakkuri, 2005: 220). French nominal address forms enjoy more frequent use and remain more formal than those in Finnish (Havu, et al., 2014: 330). In Hungarian, a wide range of nominal address forms are used in intimate relationships, whereas names and titles are favoured in asymmetrical situations (see, e.g., Domonkosi, 2002: 116-136;2010: 43-45).
Terms of endearment appear in different contexts in the three languages. In Finnish, their use is restricted to intimate situations (family or close relationships), whereas, in French, their use is extended to unofficial formal situations (e.g., work) (Isosävi, 2010: 223-224). In Hungarian, terms of endearment are favoured within friendly and intimate relationships (Domonkosi, 2010: 43-47), as well as in TV shows (Lappalainen and Vecsernyés, forthcoming).
The uses of nominal address forms in social media remain less studied. In French, nominal address forms are used on discussion forums (often with T) to express a conflictual relationship; in comments to press-related blog posts, nominal address forms are used (often with V) in greetings, excuses and thanking (Havu, 2019). In Hungarian, a rich variety of nominal address forms are used on Facebook with different functions (Domonkosi, 2019), and titles commonly appeared in the comments to posts from the Hungarian prime minister (Vecsernyés, 2021). In her study on English vlogs, Frobenius (2014: 61-62) categorised nominal address forms as generalised (e.g., 'you guys'), mediumspecific (e.g., 'Youtubers' and 'vlog fans') and specific (e.g., 'keyboard cowboys'), which we use in our study.

3.3
Greetings and Welcomings The number of greetings vary in different languages (over 200 in Hungarian, Kiss, 2004), and, in each language, different greetings vary in their pragmatic efficacy (Agha, 2007: 260). Furthermore, differences exist between languages in the frequency of greetings (e.g., in Isosävi's 2020 study, French people appeared to greet more than Finns).
Finally, complementary greetings follow an exchange of greetings in the form of a question (e.g., 'how are you'). In French, the value of complementary greetings ((Comment) ça va?) has weakened, benefitting bonjour, and is used amongst interlocutors who have met, but is generally not used with a superior unless the individual is unwell (Kerbrat-Orecchioni, 2001: 112-114). To our knowledge, complementary greetings remain unanalysed in Finnish (Mitä kuuluu?) or in Hungarian (formal Hogy van?; informal Hogy vagy?).
Finally, to our knowledge, welcomings (tervetuloa, bienvenue, üdvözlöm Vsg) have received very little attention. Isosävi (in preparation) found that welcomings were more frequent in Finnish than in French Covid-19 signs; in Hungarian, we found no such studies.

3.4
Gestures Addressing and greetings can be accompanied by gestures. McNeill (2005: 38-43) distinguished four dimensions of gestures -iconicity, metaphoricity, deixis and beat -but emphasised that they are not categorical; that is, a gesture can encompass elements from various dimensions. Iconic gestures represent images of a concrete entity or action. Metaphoric gestures present abstract meanings as a form. Deictic gestures include pointing with a hand (or other body part), and are either abstract or concrete. Beat gestures involve temporal highlighting, with the hands moving up and down or back and forth, in relation to the speech rhythm. Frobenius (2013) distinguished different pointing (deictic) gestures in vlogs: pointing toward a referent in the environment/at the camera (referencing viewers) or on the Web site (e.g., comment section), holding a referent or presenting emblematic gestures and shifting the camera toward a referent. She concluded that pointing gestures are important resources in videos where the audience is absent and imagined, and by adopting the point of view of viewers vloggers aim to interact with their viewers. Frobenius (2014) analysed amongst others the manner of speaking (e.g., whispering or speaking in a low volume) and gaze behaviour as a part of the audience design in vlogs. To our knowledge, gestures in YouTube videos have not been cross-culturally analysed.

Data, Theoretical Framework and Method
Next, we present our data (4.1.) as well as our theoretical framework and method (4.2).

Data
The data consist of Finnish, French (France) and Hungarian videos (primarily from 2019-2020) (see Appendix). The YouTubers' ages ranged from 9 to 85, and their videos span a range of topics such as beauty and fashion, lifestyle, Table 1 Overview of dataset challenges, gaming, hobbies and science popularisation. Table 1 provides an overview of our data. We examined 23 YouTubers from each language (69 YouTubers in total). We analysed the opening sequences of two videos from each YouTuber, or 46 videos per language (in total, 138 videos in the three languages).5 Analysing two videos from each YouTuber allowed us to observe recurrences. The range of followers varied, reflecting the differences in the populations of the three countries.
We refer to the videos as simply "YouTube" videos. We examined a range of different YouTube videos, choosing not to concentrate on vlogs specifically like many previous studies (cf. section 2), because the definition of a vlog may differ across time and in specific vlogger communities. For instance, one definition from Frobenius (2011)6 ten years ago regards almost any personal YouTube video as a "vlog", although meta-linguistic comments in our data found that some YouTubers considered only some videos from their channel vlogs, typically those related to their day activities.7 Furthermore, we selected videos filmed by famous and lesser-known YouTubers who at least partly created video content themselves. In most videos, the YouTubers speak, at least to some extent, about their own lives.
We concentrate on the opening sequences of the YouTube videos, which we define as follows. An opening sequence can be preceded by a short teaser containing, for instance, images, video clips, music or text. The actual opening sequence often establishes a connection with viewers and outlines the video content. Video openings differ from openings in face-to-face interactions; they can include some conversational elements, but negotiation between participants (e.g., on social status or availability for interaction) is impossible with an absent audience (Frobenius, 2011: 814-815). In our data, opening sequences varied in length, from less than ten seconds to over two minutes. An opening sequence ends with an implicit or explicit (e.g., words, cut, music or text) transition to the primary video content.

4.2
Theoretical Framework and Method Our study forms a part of research on the pragmatics of social media (Hoffmann, 2017), and is situated within interpersonal pragmatics, referring to studies of the "relational aspect of interactions", which "affect and are affected by their understandings of culture, society, and their own and others' interpretations" (Locher and Graham, 2010: 2). Interpersonal pragmatics is ideal for analysing different types of CMC (Locher and Bolander, 2017: 411). In the context of YouTube videos, interpersonal pragmatics can refer to interactions with other potential YouTubers (if the chain has at least two participants) and the audience, but we limit our analysis to concentrate specifically on the audience. Interpersonal pragmatics is not a single methodology nor does it represent one particular theoretical framework; it is rather a perspective where the use of mixed methodologies is encouraged (Locher and Graham, 2021: 569-570, 587-588).
In our qualitative analysis, we apply multimodal discourse analysis (MMDA) referring to the study of "materially diverse textual 'threads'", such as speech, images and gestures; the whole meaning of a speech or written text can be understood only by also analysing gestures and images (Kress, 2012: 36-37). YouTube videos are, par excellence, a genre in which speech is only one means of representation and meaning-making; thus, MMDA is a suitable method for the analysis of data. In examining verbal addressing and greetings, we concentrate on social deixis (or attitudinal deixis, Verschueren, 1999: 20-21), distinguishing informal and formal address forms (see sections 3.1 and 3.2); the same distinction between formality and informality applies to greetings (see section 3.3). Finally, in the analysis of gestures, we apply the dimensions of McNeill (2005: 38-43) (see section 3.4).

Analysis and Results
We begin with a cross-cultural comparison of addressing, greeting and related gestures in the three languages (5.1) and, then, concentrate on individual languages: Finnish (5.2), French (5.3) and Hungarian (5.4).

Cross-Cultural Comparison of Finnish, French and Hungarian
Opening Sequences First, we present a cross-cultural comparison of verbal addressing and greeting, and, second, of related gestures. Table 2 summarises addressing and greeting/welcomings in Finnish, French and Hungarian (Table 2).
Comparing Finnish, French and Hungarian opening sequences of YouTube videos revealed both similarities related to the genre (CMC, YouTube) and cross-cultural differences. First, if more than one YouTuber appeared in a video, they interacted more with each other and less with the audience, whilst the opening sequence functioned as a performance during which the audience was regarded as viewers of a show. When the YouTuber was alone on screen, they engaged more with viewers and viewers were considered participants activated from the beginning of the video.
Second, whilst the greetings were frequent, they were not categorical in any language (in 31-36 videos), representing another similarity across languages. In other words, a video can begin without a greeting, possibly due to the interactional nature of greetings. A greeting initiation requires a reply from the interlocutor (Agha, 2007: 260) and marks the co-presence of the other person (Pillet-Shore, 2018: 217), but the paired structure of greetings cannot be accomplished verbally with an audience behind a screen. Greetings were more Moving to cross-cultural differences, in French, viewers were most explicitly addressed using pronominal and nominal forms, whereas, in Finnish, such forms were least common. Pronominal address forms appeared most in French (36 videos), followed by Hungarian (31 videos) and Finnish (25 videos). Because Hungarian distinguishes between the singular and plural T forms (expressed by 2nd person forms), and the 3rd person is reserved for V forms, it is clear that the 2nd person plural forms are T forms. In Finnish, the 2nd plural form can be either the T or V form. Although based on previous research and the rarity of V forms (see section 3.1), we can say that in these videos most employ the plural T forms (for singular T, see section 5.2). In French, the plural of T is the same as the V form, but since the use of T forms is typically restricted to close relationships in French (see section 3.1), we assume that the pronominal forms mostly employ V forms. Furthermore, French was the only language where references to the health of viewers were frequent (e.g., J'espère que vous allez bien, 'I hope you're doing well').9 Like pronominal address forms (see above), nominal address forms also appeared most frequently in French (25 videos) and less often in Finnish (15 videos) and Hungarian (15 videos). The latter finding was surprising considering that nominal address forms are more common in Hungarian than in Finnish (see section 3.2), whereas traditionally Finnish has avoided reference to others (Hakulinen, 1987;Yli-Vakkuri, 2005). Thus, we expected to find more nominal address forms in Hungarian than in Finnish. The nominal address forms remained primarily quite general (for example, 'everyone' , 'viewers') in the three languages examined, similar to Frobenius's study of English (2011), likely because YouTubers cannot be certain of their viewers, although they hope to address large audiences. Yet, a few other nominal address forms were found in the three languages (see sections 5.2, 5.3 and 5.4).
Next, we examine gestures related to addressing and greeting. Frequent in all three languages, they appeared in 90 of 138 videos (65%),10 and seemed to make "direct" contact with "imagined" viewers. As to cross-cultural differences, these gestures were more frequent in Finnish (34 videos) and in French (30 videos) than in Hungarian (26 videos). This result is surprising, since Finnish has a reputation as a lingua culture in which the gestures are used less often than in many other cultures.11 We noticed that Hungarian YouTubers over approximately 55 years old and some middle-aged Finnish YouTubers employed fewer gestures than younger YouTubers. Therefore, the age range of YouTubers may affect the results. Table 3 summarises the number of gestures and the relationship of gestures to addressing and greeting. Table 3 shows that gestures appeared more frequently with greeting (78) than with addressing (34), but cross-cultural differences emerged. In Finnish and Hungarian, gestures accompanied greetings most often, whereas, in French, they were almost equally related to addressing and greeting. The analysis based on McNeill's four dimensions -iconic, deictic, metaphoric and beat (see 3.4) -showed that the majority (64 of 78) of iconic gestures were related to greeting, whereas the majority (24 of 34) of deictic gestures were related to addressing. Beat gestures did not appear in our data, but we identified two additional dimensions: clapping and performative gestures. With the latter, we refer to gestures which create a performance for viewers, unlikely to occur in face-to-face interactions. 10 In four French videos, one Finnish and one Hungarian video, respectively, the YouTuber did not appear in the opening sequence. 11 See, e.g., https://businessculture.org/northern-europe/finland-businessetiquette/business -communication/ (accessed 10 January 2022).   Table 4 shows that iconic gestures were most frequent (67), followed by deictic gestures (26); these two account for 93 (83 %) of 112 gestures.
Iconic gestures generally consisted of waving or other hand movements imitating greetings, typically appearing at the beginning of the opening sequence (see 5.2, Extracts 1 and 2). Deictic gestures involved pointing with the fingers or hands usually when addressing. Such gestures appeared at different points of the opening sequence, often at the end of it (e.g., see 5.4, Extract 9), such as when YouTubers pointed to viewers with their finger, asking 'are you ready': oottekste valmiita (Lakko) or vous êtes prêts (Tibo InShape). Infrequent metaphoric gestures (e.g., forming a heart or V sign with their fingers) related to addressing or greeting. Although infrequent, clapping gestures using their hands, which may help get viewers' attention, appeared in all three languages, accompanying addressing or greeting (see 5.3, Extract 4 and 5.4, Extract 7). Performative gestures occurred infrequently, but were most common in French (see 5.3, Extract 5).

5.2
Finnish Opening Sequences A typical Finnish opening sequence contained a greeting accompanied by an iconic gesture, with addressing remaining minimal (Extract 1).
Extract 1 [0:25] Beginning of opening sequence 01 PIN *moikka::* *opening her hands wide* (( Figure 1)) 'hi' 02 *mä täytän tänä vuonna kahdeksantoista* *puts her hands together and moves them* 'I will be eighteen this year' 03 *↑minä?* *points at herself* 'me' The opening sequence of Pinkku Pinsku's video (Extract 1) begins with a greeting moikka:: ('hi') (line 01). The informal moi greetings -either moi (12 videos) or moikka (5 videos) -appeared most frequently in Finnish videos, whereas the neutral hei ('hello') appeared in only three videos. As Pinkku Pinsku utters moikka:: (line 1), she lengthens the vowel (emphatic vowel lengthening), which, to our knowledge, remains unstudied in Finnish. Emphatic vowel lengthening was also used by other young female YouTubers in our dataset (Nelli kniivila, Deria Dincer, Ella and Helmi12 and Nelli Orell13). Thus, we argue that the use of emphatic vowel lengthening may be connected to a younger age and perhaps the female gender. In terms of the function of vowel lengthening, it may give more weight to a short greeting, which passes quickly and creates a personal style. Yet, further studies are needed to analyse its sociolinguistic value. The opening sequence of Extract 1 did not include a nominal address form, similar to most other Finnish videos, although when greeting, Pinkku Pinsku uses an iconic gesture, opening her hands wide (Figure 1), as if she was hugging and including all viewers in her greeting. Thus, we argue that her wide hand movement serves a function similar to the nominal address form "everybody" 12 Ella and Helmi lengthened the middle vowel (mo::i). 13 Nell Orell lengthened the vowel of the address form: heippa kaikki:: ('hi everybody').  In addition to generalised nominal address forms, Finnish YouTubers employed several other types of nominal address forms as well. The gamer Alffa pelaaja addressed his viewers with contextual and personal address forms. The first was a medium-specific form referring explicitly to players (no niin kaikki innokkaimmat Clash Royalen pelaajat, 'okay all the most eager players of Clash Royal'), whereas the second was a more surprising specific nominal form (tervetuloa kaikki karvanaamat, 'welcome all hair faces'). Based on these two videos, it is difficult to determine if karvanaamat ('hair faces') is an address form he uses regularly and if it represents a term of endearment. Nevertheless, terms of endearment remained rare in Finnish, found, for instance, in a video from Deria Dincer, referring to her cold, uttered: se ei silti estä mua tekemästä teille rakkaille mussuille videota ('it still doesn't prevent me from doing a video for you all dear sweethearts').
If an explicit verbal ending appeared in the opening sequence, it often included a directive, expressed, for instance, using a second person imperative (Extract 3). (line 03), kuunnelkaa ('listen' (line 04) and toivottavasti nautitte ('I hope you enjoy')). Address pronouns were used in more than half of the Finnish videos, primarily through plural T forms, similar to Extract 3. After the imperatives, Pinkku Pinsku uses multimodal resources to refer to her viewers, sending them kisses, an iconic gesture adding intimacy to addressing.
Although the plural T form was the most common address form in Finnish, singular T forms appeared in four videos, whilst both singular and plural T forms appeared in six videos. Singular T forms were used when YouTubers promoted their channels or products (cf. French: section 5.3): mulla on teille kampanjakoodi eli kampanjakoodilla Joona Hellman sä saat kuunnella bookbeatia yhden kuukauden ajan ilmaiseksi ('I have a campaign code for you (T, plural) that is with the campaign code Joona Hellman you (T, singular) are allowed to listen to bookbeat for one month for free').
Unlike French (see section 5.3), no popular utterances marked the end of an opening sequence, but some YouTubers seemed to use repetitive endings. For instance, the hiker Ali Leiniö showed his viewers that the "journey" was beginning, and involved them using a deictic gesture, by pointing to the imagined audience with a finger, which compensated for the lack of address pronouns in his videos: tervetuloa mukaan reissuun ('welcome along to the journey').

5.3
French Opening Sequences A typical French opening sequence contained a greeting, occasionally accompanied by a gesture, a nominal address form and an address pronoun (Extract 4).

Extract 4 [0:00] Beginning of an opening sequence 01 ENJ *bonjour* tout le monde/
*claps her hands together* (( Figure 3)) 'good morning/afternoon everyone' 02 j'espère que vous allez bien/ on se retrouve aujourd'hui 'I hope you're doing well we meet again today' 03 pour une nouvelle vidéo/ *donc avec ma SOEUR/* *opens her hands* 'for a new video thus with my sister' In Extract 4, the YouTuber utters a time-related greeting bonjour, the most frequent greeting in our dataset (16 videos), followed by the informal salut (13 videos15), and the greeting is accompanied by a clapping gesture (Figure 3). She claps her hands in both of the videos analysed, demonstrating that clapping 15 In two videos salut appeared with bonjour and in one video with the English hello. Some French YouTubers had their own personal opening utterances and styles. Two child YouTubers, Swan and Néo, had a dialogue with their mother (not shown on screen), which included an exchange of greetings and complementary greetings, after which the mother referred to the health of viewers. The YouTuber from L'atelier de Roxane had a personal style in the opening 16 Ça va ('how are you') appeared only in a video from Studio Danielle.
sequences in her videos during which she made extensive use of multimodal resources (Extract 5).

Figure 4 L'atelier de Roxane, 18 February 2020
Extract 5 [00:00] Beginning of the opening sequence 01 ROX BON:::*JOUR mais comment dire j'espère que* ((singing voice)) *extends the fingers of her raised right hand* (( Figure 4)) 'good afternoon but how to say I hope that' 02 *VOUS allez bien* ((singing voice)) *closes the fingers of her right hand* 'you're doing well' 03 *moi ça va super méga giga bien* ((singing voice)) *the right hand raised up again* 'me I'm doing super mega giga well' 04 *je suis au (.) TAquet:::* ((smiling and singing voice)) *lowers her hand and moves her fingers* 'I'm on fire' The beginning of Roxane's opening sequence represents a performance (Extract 5). She utters the opening sequence with a marked prosody, smiling and singing, and uses performative gestures related to greeting and addressing, which differ from gestures in face-to-face communication. In the greeting, the performative gesture consists of extending the fingers of her right hand (line 01). She lengthens the nasal vowel BON:::jour, emphasising the greeting 10.1163/26660393-bja10042 | Contrastive PragmaticS (2022) 1-34 (cf. section 5.2, Extract 1). When Roxane refers to the health of viewers, she emphasises the address pronoun V, and uses another performative gesture, closing the fingers of her right hand (line 02). Then, she refers to her own feelings using extremely positive expressions, combined with more gestures (lines 03-04). With performative gestures and extreme expressions, Roxane aims to establish a good feeling amongst viewers, showing her own excitement to motivate viewers. Furthermore, some distinctive French nominal address forms emerged. A gendered address form appeared in only one video from Tibo InShape: mesdames ou messieurs cette vidéo est pour vous ('ladies or gentlemen this video is for you'), accompanied by a deictic gesture. Instead of the traditional mesdames et messieurs ('ladies and gentlemen'), this YouTuber used the conjunction ou ('or'), perhaps indicating uncertainty regarding who watches his videos. Most of the other distinctive address forms referenced a wide audience. McFly et Carlito extended their address form to the French-speaking world: bonsoir la francophonie ('good evening French-speaking communities'). The YouTuber Explique-moi encore used a medium-specific address form: salut les gens à l'écran ('hi people on the screen'). Only Nota Bene used a term of endearment: mes chers camarades bien le bonjour ('my dear friends good afternoon'). Thus, showing camaraderie was likely considered too intimate by most French YouTubers.
Despite the dominance of the V form, in seven videos, variation in T and V forms emerged when YouTubers promoted something (cf. Finnish: section 5.2). YouTuber L'atelier de Roxane changed to T when she encouraged her viewers to subscribe to her channel: abonne-toi à ma chaîne YouTube active la petite cloche ('subscribe [T] to my YouTube channel by activating the small bell'). Some YouTubers used T only in writing (e.g., text in a video from Explique-moi encore: suis-moi sur Instagram ('follow [T] me on Instagram').
If an opening sequence explicitly ended, YouTubers often addressed their viewers using directives in the form of imperatives. YouTuber Enjoy Phoenix prepared her viewers for a long video: on espère que ça va vous plaire préparezvous un thé un café tout ce que vous voulez ça va être long ('one hopes that you're going to like it prepare [V] a tea or a coffee anything you [V] want it's going to be long) (cf. section 5.2, Extract 3). Furthermore, the utterance c'est parti ('let's go') occurred frequently (Extract 6).

5.4
Hungarian Opening Sequences A typical opening sequence in Hungarian contained an informal greeting and/or a welcoming often with iconic hand gestures and addresses with T (Extract 7).
Some individual choices emerged in the Hungarian opening sequences both in address forms and in gestures. For example, Cresser greeted his audience in both videos using the rarer form halihó18 ('hi') accompanied by a performative gesture (turning around) in his first video, and both deictic and iconic gestures in his second video. These gestures appear aimed at getting the attention of viewers and were accompanied by a nominal address form (lányok és srácok 'boys and girls'). 17 Both YouTubers uttered a meta-linguistic comment. The YouTuber from Emcy beauty said: jó reggelt kívánok nektek, hamarosan 11 óra lesz és még nem reggeliztem ('good morning it's almost 11 o'clock and I haven't eaten my breakfast yet'). Miklós Schnitzer uttered: sziasztok jó reggelt hajnali olyan fél öt körül van ('hi good morning it's about half past four'). 18 Kiss (2004: 132) suggests that halihó is a humourous greeting used amongst students.
In a few Hungarian videos, the opening sequence began with in medias res (Extract 8) (cf. section 5.1) (cf. Frobenius, 2011: 815 on unnecessary opening sequences in vlogs). In terms of pronouns of address, V forms did not appear in Hungarian opening sequences. Videos from four YouTubers over roughly 55 years old have some specific features, since they seemed to avoid using pronominal address forms. To determine if these YouTubers avoid choosing between T or V only in the opening sequences (8 videos), we examined their entire videos. These YouTubers did not use T or V forms before the very end of their videos, but used constructions allowing them to avoid direct addresses. However, three of the YouTubers used a pronominal address at the very end of both of their videos, when they encouraged viewers to like their channel, post comments or when thanking them for watching.19 All of these speech acts would be difficult without direct addresses. Interestingly, two of these eight videos included T forms in their titles. Because older adults typically use V forms in many contexts in Hungarian (cf. section 3.1), an avoidance strategy can indicate an uncertainty in choosing between T or V on YouTube, where T forms represent the norm, at least based on our study.
Nominal address forms were surprisingly rare in Hungarian (cf. section 3.2), appearing in only 15 videos. The most frequent nominal address was the generalised form mindenki(nek) ('(to) everybody'), similar to Finnish and French (cf. sections 5.2 and 5.3), followed by the gendered (lányok és) srácok '((girls and) boys'), which did not appear in Finnish or French. Two YouTubers used specific address forms. Ilonka néni, an 85-year-old YouTuber, used an informal greeting and a nominal address form, YouTyukák20 in both of her videos, a personalised playful version of the word YouTuberek ('YouTubers'). In her beauty and fashion videos, Alexy Vivien used a term of endearment, more specifically, a traditional, gender-based form, sziasztok szépségeim! ('hello my beauties'). Whilst some appeared to use more specific and individual nominal address forms in all three languages (cf. Finnish: section 5.2, and French: section 5.3), the generalised and medium-specific nominal address forms were preferred (cf. the categories of Frobenius, 2014: 61-62).
The ends of Hungarian opening sequences were mostly explicit (Extract 9). The second part of YouTyukák resembles the Hungarian word tyúkok ('hens'), adding a funny connotation. 21 The normative written form of the word is kíváncsiak.

Finnish, French and Hungarian YouTube Videos
Contrastive PragmaticS (2022) 1-34 | 10.1163/26660393-bja10042 Figure 7 Herczeg Hajni, 30 January 2020 02 akkor *mindenképpen maradjatok velem,* *closes her hands and points to the viewers* ( (Figure 7)) 'then definitely stay (T-2pl) with me' 03 *öt fő ilyen dolog lesz amin végigmegyünk,* *shows five fingers of her right hand and waves her hand* 'there will be five main things we will go through' In Extract 9, Herczeg Hajni motivates the audience (lines 01-02) and, then, she states that she will address five main topics (line 03). The use of an imperative (maradjatok, 'stay-2pl') (line 02) at the end of the opening sequence is a characteristic feature in Hungarian: 17 videos contain directives, expressed by 1pl or 2pl imperative forms, such as vágjunk bele ('let's [1pl] get started') tartsatok velem ('come [T-2pl] with me') and írjátok meg ('write [T-2pl] it'). The imperative forms function as a request or an order, and maintain the audience's interest. Addressing by an imperative is strengthened by a deictic gesture (Figure 7) (pointing to the viewers of the video).

Conclusions
This study contributes to research on the pragmatics of social media. First, we examined YouTube, a second-generation Internet platform, which remains unexplored in linguistic studies. Second, our study falls within comparative studies of computer-assisted communication (CMC) which remain rare, despite the multilingual nature of the Internet. Our study sheds light, on the one hand, on interactional practices shaped by features of the genre (YouTube videos) and, on the other hand, on cross-cultural differences influenced by different lingua cultures. We examined how Finnish, French and Hungarian YouTubers interact with their imagined audiences using addressing, greeting and related gestures. Our data consisted of 138 videos from 69 YouTubers in the three languages (46 videos per language), in which we applied interpersonal pragmatics, and we analysed the data using multimodal discourse analysis. Shared verbal practices included the frequent use of greetings and a preference for generalised nominal address forms ("everybody"). In contrast to our original hypothesis, cross-cultural analysis revealed that Finnish and Hungarian were closer to each other than to French, which used the most explicit and formal means of addressing and represented the only language with frequent inquiries about viewers' health. Although previous research on addressing identified diverging address practices in Finnish and Hungarian, our YouTube data showed that both languages used informal T forms and fewer nominal address forms. This finding may indicate a change in Hungarian address practices, although it remains unclear if this extends beyond the YouTube genre.
Turning to shared practices on gestures related to addressing and greeting, we found that iconic and deictic gestures occurred most frequently in the three languages; iconic gestures generally related to greeting and deictic gestures to addressing. We suggest two interpretations. First, the iconic gesture -that is, the waving of hand(s) -has similarities with face-to-face communication, where it is often used when participants are far from each other (like the YouTuber is "far away" from the audience). Second, the deictic gesture -that is, pointing with the finger -plays a different role in YouTube-specific communication from face-to-face conversation, where it can be considered impolite in many cultures.22 Yet, cross-cultural differences emerged in the frequency of gestures. Surprisingly, gestures appeared most frequently in Finnish, typically regarded a lingua culture with a restricted use of gestures, a finding possibly stemming from the influence of English videos. In terms of the types of gestures, iconic gestures occurred more often than deictic gestures in Finnish and in Hungarian, whereas iconic and deictic gestures were almost equally frequent in French.
Whilst this qualitative study relied on a fairly small dataset, a larger corpus would enable a quantitative analysis. In addition, we limited this study to the opening sequences, although research on closing sequences would complement our understanding of addressing, greeting and related gestures in YouTube videos. If future studies address different languages and lingua cultures, a more complete picture would emerge. Generally, more information is needed on different genres of social media, and the interplay between shared practices influenced by the technological platforms and the cross-cultural differences. Finally, future research should examine the relationship of language on social media and in face-to-face communication in different lingua cultures to determine how they influence each other.

Finnish and Hungarian
Finnish and Hungarian speech was transcribed according to conventions commonly used in conversation analysis. .
strongly falling pitch at the end of a prosodic unit , flat pitch at the end of a prosodic unit ,?
slightly rising pitch at the end of a prosodic unit ?
strongly rising pitch at the end of a prosodic unit moikka:: lengthened vowel ↑ segment produced on a higher pitch level than the surrounding speech >joten tota< accelerated speech rate ((laughter)) transcriber's comments (.) micropause (duration of less than 0.2 sec) French French speech was transcribed according to the orthographic conventions developed for French language by the ICOR group (ICAR lab, Lyon): http://icar .univ-lyon2.fr/projets/corinte/documents/2013_Conv_ICOR_250313.pdf. / rising intonation of the preceding segment .
short pause (less than 0.2 s) VOUS emphasis : stretching of prior syllable ((singing voice)) transcriber's comments non-standard elision