1. Introduction

Global tourism has been (and continues to be) both a major catalyst for economic development and the largest industry across international borders (Lyon & Wells, 2012). Based on the latest annual report of the United Nations World Tourism Organization (UNWTO, 2015), international tourist arrivals in 2014 totaled 1.13 billion and generated US $1245 billion in receipts, representing 30 percent of worldwide service exports. The growth of this industry has also surpassed gross domestic product (GDP) growth, accounting for 9% of GDP in 2014. Despite the significant role that tourism plays in today’s world, relatively few empirical studies have been undertaken to explore tourism from a language and discourse perspective.

Unlike other industries, tourism is an information-intensive business, since the discourse of tourism mediates the relationship between tourist and destination by manipulating various kinds of linguistic and audio-visual resources to persuade potential consumers and convert them into actual clients (Jack & Phipps, 2005). The role of discourse in tourism has attracted some research efforts in the past decade, including studies focusing on tourism websites (Catenaccio, 2009; Garzone, 2009; Hallett & Kaplan- Weinger, 2010), TV holiday programs and newspaper travelogues (Thurlow & Jaworski, 2010), and tourist brochures (i.e. Dann, 1996; Santulli, 2011). However, few empirical studies have examined the discourse of hospitality corporations. These corporations are an important part of the tourism industry, as hotel expenditure constitutes one of its major sources of revenue (Law & Hsu, 2006). This study will address this gap by comparing the websites of two luxury resort hotel chains to examine how they represent themselves for promotional purposes.

2. Web-mediated tourism discourse and identity construction

The rise of the internet has radically changed the discursive practices adopted by businesses, and the construction of a strong, favorable brand identity through the semiotic resources offered in the online environment is crucial for tourism firms seeking to increase their competitiveness. A brand represents the characteristics and qualities associated with a firm that differentiate it from rival companies in the target market (Qu, Kim & im, 2011). It is a means by which a company creates values for itself and others (Gnoth, 2007; Költringer & Dickinger, 2015), so that the brand can be identified and selected by consumers. According to Machin and Thornborrow (2003: 454), “the contemporary commercial concept of a brand is realized through specific discursive practices”. They further pointed out that this concept of “brand” is very similar to Kress and Van Leeuwen’s (2001) concept of discourse, as noted in Machin and Thornborrow (2003), a concept which is viewed as contextualized, specific knowledge about a social practice and can be realized by means of various semiotic resources. As “a set of representations and values” (Machin & Thornborrow, 2003: 454), a brand can be examined by means of multimodal discourse analysis.

Although little empirical research has been undertaken to examine brand identity and the use of multimodal resources in the hospitality business, two studies on web-mediated tourism websites and identity construction from a discourse perspective are particularly pertinent to the present research. In comparing the official tourism websites of Spain and Italy, Garzone (2009) concluded that more differences than similarities were found, despite both countries being well-known for their location on the Mediterranean and their popular seaside resorts, as well as for being part of the ancient roman Empire. The analysis of the verbal elements, homepage layout, and types of hyperlinks used indicated that the Spanish website adopted a more interpersonal approach than the Italian one. This illustrates the higher degree of interactivity and more diverse range of images used to project a Spanish identity that is more varied and less universally known than Italy’s. By contrast, Italy’s website tended to reinforce the stereotypical national identity, with a focus on the country’s cultural and artistic heritage, constructing a typical Mediterranean identity without introducing users to less familiar features of the country.

Likewise, another study by Catenaccio (2009) examining the differences and similarities among two official websites and four corporate websites with a focus on health tourism in India also found marked differences in identity construction. Based on the analysis of lexical and visual choices, their results showed that the official websites highlight Indian medicine but downplay the Western medical tradition. In contrast, the corporate websites focus on Western medical practices, with limited representations of “indianness”. Catenaccio suggested that the medical tourism websites differ in their self- representations, with the official websites attempting to communicate an essential Indian identity. These studies thus indicated that the textual and multimodal resources represented in official and corporate websites serve as a means to convey an intended image and identity.

Despite the paucity of the published literature addressing hotel websites, the findings suggest that the genre of hotel websites can exhibit its own distinctive textual and multimodal features. For example, Suen (2009) attempted to identify common traits among five-star commercial hotel websites, that is, the textual and image features constitute the major characteristics of hotel websites as a genre. By applying Bhatia’s (2004) multidimensional genre model and Kress and Van Leeuwen’s (1996) visual grammar, Suen found typical patterns in the obligatory/optional move structures, the types of hyperlinks, the frequently occurring words, and the three major functions of the images projected in the websites examined. Of particular relevance to the current research are the representational functions of images used by these five-star business hotels, including the concepts of grandeur and elegance, prosperity, relaxation, and hospitality. Another study by Cheng and Hsu (2013) conducted a cross-cultural website comparison on hotels located in Japan and the US. The lexical and image analysis revealed that the hotel websites of the Japanese Princess hotel chain tend to highlight the peaceful atmosphere and quiet activities, such as “enjoy[ing] the romantic seaside view” and “enjoy[ing] the great nature from the hot spring”, whereas the websites of the US-based Sheraton hotels emphasize more active leisure activities, such as “Enjoy[ing] ... a snorkeling adventure” and “Enjoy[ing] a refreshing dip in … the swimming lagoon”. This initial finding may reflect the cultural differences in what local citizens expect from luxury hotels.

The above review shows how the semiotic features of hotel websites reflect either common traits or cultural differences among five-star hotel firms, but they did not address how these features relate to the construction of brand identity. Among the full range of attributes of a hotel – such as location, infrastructure, amenities, dining, and activities – only a limited number can be selected and represented on a given website, and these will contribute to defining the identity of the hotel brand being marketed. The present study will thus focus on the deployment of textual and visual resources and examine how these semiotic resources contribute to constructing each hotel’s brand identity.

3. Visual analysis

The current approach to visual analysis is couched in the framework of multimodality, as proposed in The Grammar of Visual Design, developed by Kress and Van Leeuwen (1996 & 2006) (see Figure 1). Kress and Van Leeuwen (1996) extended and adapted the Systemic Functional Linguistics (SFL) metafunctions of language – ideational, interpersonal, and textual – to the study of images and other visual modes as representational, interactive, and compositional, respectively. The visual systems project the three meanings simultaneously, constituting multidimensional structures. Each meaning can be realized through the choices taken from system networks, as explained below.

Representational meanings refer to how the participants, processes, and relationships are represented in the image. This system includes conceptual and narrative processes. Narrative processes present “unfolding actions and events, processes of change, transitory spatial arrangements”, while conceptual processes present “participants in terms of their more generalized and more stable and timeless essence, in terms of class, structure or meaning” (Kress & van Leeuwen, 2006: 79).

Interactive meanings include three dimensions: contact, social distance, and attitude. “Contact” refers to the ways that images directly or indirectly address their viewers: demand and offer. A demand image engages viewers with a direct gaze from a person in the image, while an offer image addresses the viewers without a direct gaze. “Social distance” refers to the distance of the presented image from the viewers; the further the social distance, the more distant the relationships. These social relations are realized by varying the length of shots from very close-up shots through long or distance shots. The third dimension indicates “attitude” or point of view toward the represented images, and is determined by the angle of the photographic shot. It includes the objective (without perspective) and the subjective (with perspective). The subjective features of the images are realized through various choices of horizontal and vertical angle perspectives. The former can be further divided into frontal and oblique angles, which differ in degrees of involvement. The vertical angle is important in establishing power relations between the viewer and images.

Compositional meanings refer to the ways that the representational and interactive meanings are integrated into a meaningful whole, for instance, the composition of a text comprising one or more visual elements along with written texts.

This framework has been widely applied and adapted in prior research to a variety of contexts (see Ventola & Guijarro, 2009; Bowercher, 2012); it has also received some enthusiasm in recent publications on tourism discourse (Suen, 2009 on Hong Kong five-star hotels; Hallett & Kaplan-Weinger, 2010 on various official tourism websites; Francesconi, 2014 on irish tourism texts and websites). Since the multimodal analysis of tourism discourse has so far been mostly exploratory or based on case studies, the present study will advance the field by conducting a systematic investigation into one major component of the tourism industry, the hospitality business, with a well-controlled corpus design in terms of size and representativity, and the use of a consistent coding criteria to identify prominent semiotic features.

4. Methods

This study aims to examine how two five-star resort hotels represent themselves through hotel-owned websites to develop their brand identity. To do so, brand identity is investigated by identifying and comparing core traits of verbal and image construction that stretch across the hotel chain websites of each hospitality firm. The textual analysis explores how such a composite corporate identity is linguistically codified at the lexical level while the image analysis is undertaken to uncover how it is manifested in the representational and interactive meanings of the image choices.

4.1. Data sources

The present study focuses on examining two major luxury hotel chains: Banyan Tree Hotels and resorts and Aman resorts. Both were initially established in Phuket, Thailand. Aman resorts was established in 1988 and Banyan Tree Hotels and resorts followed in 1994. They are typical Asian companies that have grown into one of the world’s leading operators in the hospitality industry, with each firm operating more than 20 luxury resorts and hotels worldwide. Both are listed in the Five Star Alliance, an organization providing reservation services for a broad range of the world’s best resorts and hotels. Although international brands are numerous, there are few successful Asian brands in this sector with the exclusion of the giant brands of Japan. So far, both hotel companies have established their locations primarily in Asia, but they aim to expand geographically outside of the region as part of its growth strategies. Both face the challenge of transforming from a niche Asian brand into a global brand in the travel industry. These two hotel brands were chosen for study since Asia is considered to be the major economic player in the 21st century, and thus how Asian companies gain access to many global markets and manage to create and maintain powerful global brands warrants further attention.

The primary sources for textual and image analyses are limited to the homepage (the first or main web page of a website, see Figure 2 for illustration) of each hotel-owned website, because the first page of each website is the most important element in terms of its promotional effect on website visitors. This involved accessing only the page in the first layer of depth inside each website without taking account of hyperlinks. All of the major texts and images accessible with a single click from the homepage of these hotel websites were downloaded and collected into four subcorpora: Banyan Tree texts, Banyan Tree images, Aman texts, and Aman images.

As can be observed, the upper half of the homepage is occupied with a large frame photo and promotional texts introducing the distinct charms of each hotel. In the lower part of the homepage, three more photos in smaller frames along with brief messages, consumer quote, and reviews and awards are found throughout each hotel homepage. The Aman resorts sites are very similar to the format in the upper half of the Banyan Tree resorts homepage. The major difference is that all four photos are foregrounded with a large frame, with each displayed as a series of flash photographs with a five-second pause between each one.

At the time of data collection, from January to April, 2014, there were 24 Banyan Tree Hotels and resorts and 27 Aman resorts. Each company adopts the same homepage layout for its websites for hotels in different locations, as shown above, but uses different texts and images to highlight the features of specific locations. To compile comparable data, none of these brief accompanying messages and quotations in the Banyan Tree websites is included in the data, except for the major texts shown below the large-frame photo. The data compiled from 24 Banyan Tree resorts homepages consists of 24 texts and 96 photos, while that from the homepages of 27 Aman resorts includes 27 texts and 108 photos (see Table 1). Although Aman resorts has more locations than Banyan Tree resorts, the corpus size of the homepage texts is fairly similar: 3,069 words are in the Banyan Tree resorts subcorpus compared with 3,092 words in the Aman resorts subcorpus.

4.2. Textual analysis

The textual analysis starts from a quantitative overview of the lexicon, relying on the tools of corpus linguistics, in order to identify the characteristics and the main themes that appear more dominantly in the homepages of each hotel website. Following the methodology adopted in prior literature on tourism and company identities (Catenaccio, 2009; Degano, 2009; Garzone, 2009), keywords and the most frequent word lists in the two text subcorpora were generated with WordSmith Tools 5 (Scott, 2010). Keywords refer to those terms that are particularly frequent on one of the two text subcorpora, and these are identified by comparing the frequency lists of a study corpus with a reference corpus. With regards to the frequent word lists, the top ten lexical items were selected from the lists of each subcorpus, in line with Degano (2009). These were achieved by excluding proper nouns, such as Aman or Banyan Tree, and all items that are not nouns, adjectives, and verbs. Since these selected words occurred roughly above 0.32% of the time, they were considered as valid sources to observe the most frequent topics on each hotel homepage. To look more closely at the discursive choices, concordances for the keywords and the ten high- frequency lexical items were also retrieved electronically and analyzed using extracted texts.

4.3. Image analysis

The image analysis was undertaken based on Kress and Van Leeuwen’s Grammar of Visual Analysis (2006), and mainly focused on representational and interactive meanings. Although the representational meanings in their framework can be further categorized into two types, the current study is not interested in making distinctions between narrative or conceptual images. Rather, it focuses on the symbolic meanings represented in relation to some sense of the mood or atmosphere of the scene. Given that the representational meanings of visual elements is industry-/discipline- specific, Suen’s (2009) set of hotel image categories for representational meanings, which draws on Kress and van Leeuwen’s (1996 & 2006) framework, provided the theoretical grounding for category formation. After the initial analysis of collected photos, Suen’s scheme was expanded into a set of seven categories. Coding criteria for each category were established recursively, with schemes being created, tested, and refined until clear and consistent criteria were developed. Due to space constraints, a brief introduction of the categories along with the pictorial examples is given in the results section. Each photo was considered as a single unit of content, and thus classified into only one category (Krippendorff, 2004).

A formal reliability study was conducted to assess the soundness of the image categories through intra-rater and inter-rater analyses. The intra-rater analysis involved the researcher analyzing the data at two different points in time, and in the current case a time interval of 6 months was used. In terms of representational meanings, a second rater, a senior manager in an international 5-star chain hotel who was also an MA student applying Kress and van Leeuwen’s framework to study other hotel websites, was recruited to conduct the image analysis. The rater training included introducing the categories of representational meaning developed by the researcher and analyzing 24 photos with the researcher’s guidance. After the rater training, the second coder independently analyzed 120 photos (60 from each company). These reliability analyses were undertaken using the Statistical Package for Social Sciences (SPSS) version 19.0 to calculate the kappa statistic K, which measures the degree of agreement between raters. The kappa values were .889 and .795 for intra-rater and inter-rater agreement, respectively, which indicated “almost perfect” (0.81-1) and “substantial” (0.61-0.80) agreement between the coders (Landis & Koch, 1977). A detailed discussion between the raters with regard to the specifics of certain representational meanings was carried out until any disagreements were resolved.

The analysis of the interactive meanings focuses on contact (demand vs offer), social distance (close, medium and long shot), and subjective attitude, including horizontal (frontal vs oblique angle) and vertical angles (above, at and below eye level). Some pictorial examples of this coding scheme were also offered in the result section. Although this analysis was relatively straightforward, it was also validated through intra-rater analysis. There were only two disagreeing cases, which were resolved through the researcher’s reflections.

5. Results

5.1. Lexico-grammatical analysis

Each hotel subcorpus was analyzed with a view to exploring how a corporate brand image is linguistically construed at the lexical level. The keyword analysis of the Banyan Tree subcorpus shows that the only keyword is “you”, apart from “Banyan” and “Tree”, with 21 entries and a keyness index of 29.34 against none in the Aman resorts data. It is evident that the interactive personal pronoun “you” was utilized frequently in this context, revealing that Banyan Tree resorts aimed to establish a higher degree of interactivity with reader involvement, creating a sense of hospitality. The deployment of “you” often acts to place readers in the position of the beneficiaries of some specific action related to their personal enjoyment, inviting them to share a journey of exploration that is guided by the text. “You” tends to colligate 13 times at N-1 or N-2 with verbs such as “enthral”, “invite”, “greet”, or “cater to”, and 7 times at N+1 or N+2 with verbs such as “sink into”, “dream”, or “imagine”. Here N refers to the node word whose salient pattern of co-occurrence with other lexical items is under examination. The number indicates the span of collocations or colligations to the left and right of the node word. In contrast, in the Aman subcorpus no keyword can be identified except “Aman”, the hotel name.

The frequency lists of the two subcorpora were further analyzed to identify the dominant topics. In Table 2, the left column presents the occurrences of the top ten most frequent content words in each subcorpus, while the right column shows the occurrences of the same words in the other hotel subcorpus.

In both corpora, the words “resort”, “island”, and “beach” appear frequently, with the term “resort” used to refer to the hotel itself and “island” and “beach” to describe the location. Except for the deployment of these three terms, specific differences can be observed in the ways each hotel represents themselves through textual means.

The subcorpus of Banyan Tree resorts gives marked prominence to its famous spa services, with 20 entries for the word “spa” in its corpus. it also shows the high frequency of the word “sanctuary” (16 entries) to construct the Banyan Tree brand, and a number of words relating to sensual experience, such as “experience” (with 18 entries), “offer” (17), “senses” (11), “discover” (10) and “romance” (10). interestingly, the Aman resorts websites give more attention to the role of each hotel’s location, featuring higher occurrences of words such as “situated” and “located”, as well as general geographical terms, such as “city”, and “sea”. Another theme that emerges from the high-frequency words is the use of evaluative adjectives, such as “peaceful” and “private” to highlight the characteristics of Aman resorts, with 12 examples for the word “peaceful” and 10 for the word “private”. On the Banyan Tree resorts websites, the word “private” also has eight entries but the word “peaceful” is not present at all. This indicates that private space or privacy denoting exclusivity is a common trait across both luxury resorts.

The above observations based on the top ten lexical items are further substantiated and expanded by taking into account the co-occurring words and their functions. The concordances of these words are discussed along with their collocations or colligations to show the major differences in the brand construction for both hotel chains (the full list of these concordances will be provided upon request). Among the common top 10 words, the concordance lines of “resort”, “island”, and “beach” in the Banyan Tree resorts subcorpus have strong colligations with evaluative adjectives. “resort” colligates 10 times at N-1, N-2, or N-3 with evaluative modifiers (i.e. “this world-class resort”, “a pampering resort stay”, “this enchanting vacation resort”, “luxury resort living”, or “the exclusive hot spring resort”). “island” colligates six times with evaluative adjectives such as “private” or “exclusive” at N-1, or with “sanctuary”, “paradise”, or “romance” at N+1. Likewise, “beach” occurs five times with evaluative terms such as “idyllic”, “excluded”, or “premier” at N-1, or with “retreat” or “paradise” at N+1. All these evaluative terms not only indicate how each Banyan Tree resorts hotel positively assesses its location, amenities, or surroundings, but also express and negotiate the hotel’s ideological positions. They are deployed to strategically present Banyan Tree resorts as one of the best luxury hotels and to evoke in readers a favorable attitude toward the resort. Readers are positioned to take the explicit values of evaluation expressed in the text.

By contrast, these evaluative terms are only minimally featured in the Aman resorts subcorpus. “resort” colligates four times with evaluative expressions at N-2 or N-3 (i.e., “a secluded seaside resort” or “a tranquil all-season resort”, while “island” colligates with these expressions once at N-1 (i.e. “peaceful island”) and “beach” with evaluative terms three times at N-2 (i.e. “a beautiful white-sand beach”). “Resort” and “island” often collocate with the definite article “the” at N-1. This occurs most often when the writer describes the hotel’s locations, design, and geographical characteristics, such as “The resort is only 15 km from the Forbidden City”, “The resort features the original village layout” or “the island of Bora Bora”. “Beach” colligates seven times at N-1 with classifiers such as “sand”, “pebble”, or “crescent- shaped”. The above distinction is further strengthened and elaborated in the following concordance examples of other frequent content words.

In the Banyan Tree subcorpus, the colligations of the most frequent verb items, “experience”, “offer”, and “discover” co-construe the sensory experience. “Experience” mainly acts as a verb, indicating the sensual activities that the guests can enjoy (i.e. “experience a sense of enchantment & romance”, “experience Banyan Tree’s romance of travel”), and co-occurs with these activities 12 times. It also occurs 6 times as a noun, highlighting the special experience that one can have at this hotel (i.e. “an Arabian wildlife experience” “water-village resort experience”). Likewise, “offer” and “discover” often colligate with certain activities or experiences, such as “offer awe-inspiring views”, “offer a signature blend of romance and travel” or “discover the charm, romance and dramatic sunsets” and “discover what your senses have been seeking”.

It is noteworthy that the word “spa” occurs 20 times in the Banyan Tree subcorpus, compared to its marginal presence in the Aman resorts subcorpus. It collocates six times at N-2 with Banyan Tree and four times at N-1 with Phuket. It also occurs with the ostensibly promotional adjectives “renowned”, “acclaimed”, and “Asian-inspired” to emphasize the resorts’ superior quality and exotic sensuality. Banyan Tree resorts has established a high status for their spa facilities, and the explicit highlighting of this service in the homepage text indicates that this can likely contribute to the promotion of the hotels.

The word “sanctuary” in the Banyan Tree subcorpus is primarily employed to designate the hotel as a safe haven for travelers to relax their mind, body and soul. Not only does the connotative meaning of this word project this image, but this word also often collocates six times at N+3 with a phrase involving the “senses” (another frequent word), such as “sanctuary for the senses”. This is designed to be the marketing tagline used to distinguish the Banyan Tree brand from its competitors (Wirtz, 2012).

As noted above, Banyan Tree endeavors to conjure up a luxurious image associated with palpable sensations to appeal to busy and successful professionals; yet, Aman resorts seems to emphasize a subdued and restrained image. The Aman subcorpus features a higher number of top ten frequent words referring to geographical descriptions, in addition to the above-mentioned words “island” and “beach”. The verbs “situated” and “located” show no strong collocations or colligations with any evaluative expressions or special geographical characteristics. They are deployed, instead, to neutrally depict the location of each resort. Likewise, “sea” is often capitalized and used to refer to a specific sea, such as “the East Sea” or “the Andaman Sea”. “City” does not show any particular collocations or colligations either. There are also few evaluative terms that co-occur with “city” and “sea”, except “this legendary city”, “the ancient city”, and “remarkable sea view”.

In this subcorpus, the two most prominent evaluation terms are “peaceful” and “private”, which are undoubtedly qualities that travelers associate with luxury hotels. However, a closer look at the use of “peaceful” shows that, except for three instances in which it is employed to describe the resort, it is used mainly to explain the meaning of each specific resort’s name. indeed, the word “Aman” means “peaceful” and this enterprise aims to name each of its hotels by using “Aman” plus a label derived from its surrounding geographical features, such as forest, hill, or island, or words with spiritual connotations, such as soul or paradise. The way each hotel is named is further explicated in the collocations and colligations of the term “word”. “Word” collocates 6 times at N-1 with “Sanskrit” and colligates four times at N-1 with other language origins, while it collocates 6 times at N+2 with the word “peace”. Aman resorts explicitly promotes its peaceful image through the explanation of each hotel’s name and the origins of their names. Another frequently occurring evaluation term associated with the image of luxury resorts is “private”, which has strong colligations with the resorts’ facilities, such as “private courtyards”, or their surroundings, such as “private white sand beach”. Aman resorts thus aims to highlight the peaceful and private characteristics associated with typical luxury hotels on its homepages.

Overall, the above textual analyses reveal that both luxury resort hotels show some common traits by highlighting their exclusive locations associated with a beach and island, along with their emphasis on privacy. However, such luxury images are represented with distinguishing characteristics. In the subcorpus of the Banyan Tree texts, the marked themes are the diverse kind of sensory experiences to be consumed by tourists in pursuit of physical and psychological well-being. The language used offers visitors vivid depictions through a wide range of evaluative lexis, inspiring their imagination of the marvels to be anticipated. Further, the texts are constructed interpersonally with a high degree of interactivity through the frequent use of “you”, indicating the hotel’s eagerness to convey friendly and warm attitudes. By contrast, the dominant themes constructed in the subcorpus of the Aman resorts texts are the factual descriptions of the hotels’ geographical locations and the origins of its hotel naming practices, with a much lower occurrence of evaluative terms. it seems that Aman resorts intends to project a peaceful and exclusive image through the connotative meanings of its naming practices and the identification of the locations, but the tangible characteristics of a peaceful scenery, rich landscape, or restful atmosphere are not highlighted. Sensory experience is thus connoted, but not thoroughly and obviously embellished and promoted.

5.2. Visual analysis

The corporate identity is not only linguistically codified at the lexical level, but also created and enriched visually through the accompanying images, as illustrated below.

5.2.1. Representational meanings

The types of representational meanings include relaxation (relaxation activities and facilities, see Figure 3), romantic atmosphere (romantic activities or atmosphere, see Figure 4), grandeur and elegance (the design of the hotel building or interior, see Figure 5), seclusion (the presentation of the isolated location of the hotel surrounded by mountains or sea, or a solitary person manifested as a tiny spot in an immense natural landscape), spectacular natural beauty (a panoramic overview of the hotel with its surrounding landscape showing a magnificent outlook, see Figure 6), prosperity (an overview of a city’s prosperous appearance), and local views (images of indigenous people’s activities or the exotic landscape).

As shown in Table 3, the images on the Banyan Tree resorts homepages highlight the visual appeal of relaxation (see Figure 3), grandeur and elegance (see Figure 5), and romantic atmosphere (see Figure 4), whereas those of Aman resorts emphasize seclusion, relaxation, and grandeur and elegance. Both hotels’ homepages share the common luxury hotel intent of capitalizing on the grandeur and elegance of the buildings and the wide range of relaxation services and facilities on offer to induce a sense of relaxation, pleasantness, and upscale exclusivity. it is notable that the dominant focus in these relaxation images is the recreational services and facilities that are available, such as spa-related components (i.e. spa facilities or massage), or sitting in a beach chair to savor stunning views. There is relatively marginal inclusion of vigorous physical activities, such as swimming, scuba diving or water skiing. Apart from these similar features, Banyan Tree resorts also emphasizes the romantic sense of its hotels’ surroundings or facilities, whereas Aman resorts connotes seclusion and a tranquil atmosphere, exotic and unpretentious beauty derived from its hotels’ isolated locations, and the exotic experiences that can be experienced with relish in the local areas.

5.2.2. Interactive meanings

Interestingly, both hotels share great similarity in establishing interpersonal relationships with the viewers through the use of images across the three dimensions of interactive meanings. As shown in Table 4, there is only one demand image occurring in the Aman resorts homepages, in which the represented participants address the viewer with a direct gaze. Few images of people are presented in the homepages of both firms. Where images of people do appear, such images generally involve no direct contact with the viewers but feature, instead, the averted gaze of the represented participants. The great majority of the photos are images of goods and services, or information about the hotels’ buildings and surroundings, presented to the viewers for their own scrutiny (see Figures 3 to 6). These images invite the viewers to “enter the represented natural or artificial space, which seems to be waiting to be visited” (Francesconi, 2014: 82).

Likewise, the frequencies of close, medium, and long shots are similar across both hotel homepages. Close shots are featured more frequently than medium range shots, which are in turn shown more frequently than long shots. The majority of the images are displayed at close distance, mainly depicting the hotel’s facilities. At close distance, these facilities or services are shown only in part (see Figure 3), as if viewers could actually enjoy these facilities or services physically. By projecting the images of these facilities or services as within the viewer’s reach, the hotel can encourage viewers to visualize what they can afford to stay in the hotels.

Photos presented at a social distance can illustrate the depicted objects in full, but without much space around them (see Figure 7). They are represented as still within the viewer’s reach, but the implication that the viewer can immediately use the services or facilities that are shown is lost. This type of image tends to portray the hotel’s outdoor facilities or services, and requires a small frame of the object to provide a larger field of vision so that consumers can clearly see the focal object.

A relatively small number of photos are taken from a long distance (see Figure 6) to show the hotel and its surroundings “from a lookout position, a place not itself in the landscape but affording an overview of it” (Kress & van Leeuwen, 2006: 108). Such a panoramic gaze “invites a reassuring act of contemplation” (Giannitraphani, 2010, as cited in Francesconi, 2014: 94) and is conducive to the appreciation of and aspiration for picturesque scenery.

In addition to the image act and the selection of social distance, as discussed above, the angle selected to express the producers’ subjective attitude toward the objects and participants depicted is also very similar across both hotel homepages. In terms of the horizontal angle of the photographic shot, almost two-thirds of the photos are taken from a frontal angle (see Figure 5), projecting a message of inclusion (Kress & Van Leeuwen, 2006) so that the represented images are part of the viewers’ world. On the other hand, one-third are taken from oblique angles (Figure 3), conveying a message of exclusion and that the depicted image is not part of a shared world (Kress & van Leeuwen, 2006). Although this implies a detached stance (Kress & van Leeuwen, 2006), these photos are often co-deployed with close or social distance shots. They seem to beckon the viewers with a large field of vision to create seductive experiences. As pointed out by Suen (2009), the use of this photographic technique aims at providing a broader view or more details of the hotel instead of creating detachment from the viewers. Affording a broader view or greater detail may strategically invite the visitors to explore the hotel websites in greater depth.

With regard to the vertical angle shots, both homepages project an equal relationship and form friendly bonds with the viewers through the sustained use of the camera at eye level. Images are seldom presented at a vertical angle from below eye level to exert some symbolic power on the viewers. When viewers are placed in such an inferior position to the focal items, these images are often encoded as sublime and imposing (Hallet & Kaplan- Weinger, 2010; Francesconi, 2014). Occasionally, the viewers are positioned at the elevated level to overlook the images featured in the homepages, as if the featured scene is at their command (Figure 6).

Overall, the high frequency of offer images along with the use of intimate and social shots, and with involvement and equality angles, unveil that there is a frequent use of direct interpersonal appeals in the websites of these luxury hotels. This visual semiotic design induces potential customers to visualize what “being there” would be like. Viewers are encouraged to acknowledge the superior quality of services and goods, although some photos are taken in long shots and at oblique angles or above eye level. The use of various fields of vision invites viewers to “enter the scene, enjoy that panorama, from the particular point” (Francesconi, 2014: 77).

6. Discussion

The present study is the first attempt to connect the discourse analysis approach to the brand construction of hospitality companies, and thus it extends the published literature on hotel websites by conducting a systematic multimodal analysis. The findings reveal that common traits associated with resort hotels can be observed in their homepages’ focus on verbal descriptions of serene and enchanting geographical locations, and visual representations of relaxation activities/services and the grandeur architecture/design of hotel buildings, along with strong interpersonal appeals. This suggests that the affluent consumers who are likely to stay at resorts hotel are searching for relaxation or rejuvenation in an esthetically pleasing environment with arresting ambience. What distinguishes these two luxury hotel chains is their respective positionings of their luxury images. Banyan Tree resorts projects an identity of sensual luxury, positioning each resort as a romantic, welcoming and pleasurable setting. In contrast, Aman resorts insists on understatement and pristine luxury, promoting simplicity, tranquility, and plainness. This reflects that both hotel firms aim to obtain a clear brand position in the global hospitality industry, working to evoke a distinct sense of upscale exclusivity in order to zero in on a well-defined segment of the market.

These results lend support to the importance of a strong, unique image in constructing a corporate brand identity for survival within a globally competitive marketplace (Echtner & Ritchie, 2003; Qu et al., 2011). Creating a differentiated image is the essence of hotel positioning, since it can give firms the power to differentiate themselves from competitors, and thus capture consumers’ minds in the target market (Qu et al., 2011). The findings have some practical implications for hospitality management professionals with regard to the significant role of developing a distinctive image as part of their brand construction process, with this emerging as their endeavor to offer an enriching and rewarding experience for prospective guests.

Despite the distinct luxury attributes seen in both hotel websites, both firms intend to promote the “esthetic experience” of staying in these locations, according to the categories of experiential marketing within tourism and hospitality (Pine & Gilmore, 1999: 30). Such experiences entail guests’ enjoyment and passive appreciation of an enriched, unique physical setting, such as admiring a spectacular gorge landscape or breathtaking mountain views. Interestingly, this result is similar to the verbal and visual representations used by Japanese Prince Hotels, but in contrast to those of American Sheraton Hotels, where more emphasis is placed on escapist experiences, as reported in Cheng and Hsu (2013). Escapist activities involve consumers’ active participation in intense activities and immersion in the environment, such as golfing, biking or snorkeling (Pine & Gilmore, 1999). it is intriguing that the three Asian hotel brands, despite their increasingly global nature, gave more prominence to the esthetic elements on the hotel homepages. This can be partially attributed to the different cultural attitudes toward activity. American culture stresses activity and action, characterizing a “doing” orientation (Samovar, Porter, & McDaniel, 2010). This value orientation contrasts with that fostered in the Asian tradition, where the individual is not encouraged to be an active agent and thus quiescence and inactiveness are often promoted (Samovar et al., 2010). To what extent such cultural orientations affect the promotional features activated in the websites deserves further attention by studying a greater number of international brands with a wider variety of origins.

Of special interest are the similar interactive features of visual elements displayed across both hotel homepages, which contain a proportionally higher number of close-up shots, and offer images, along with involvement and equality angles. Surprisingly, this common trait differs from that seen on the webpages of commercial 5-star hotels located in Hong Kong, as documented in Suen (2009). In these hotel homepages Suen found the frequencies of close, medium and long shots were roughly even. This marked difference in the use of long shots can be possibly attributed to the hotel type: commercial vs resorts hotels. As noted previously, the images of long shots generally depict panoramic views as out of reach and for contemplation, creating a sense of detachment. This may engender an effect of an uplifting experience through a momentary detachment from business professionals’ competitive worlds and an immersion in the picturesque scenery. Images with long shots can thus enhance the appeal of commercial hotels to potential customers, thus accounting for their higher frequencies in relation to these hotels. However, given the paucity of the literature on the images represented on hospitality websites, more research is necessary to substantiate this explanation.

Another noteworthy finding is the prominence of promotional evaluative lexis on the Banyan Tree homepages. It would be interesting to investigate the relationship between the types of evaluative language and brand identity construction by applying Martin and White’s (2005) appraisal framework. This framework has categorized three subtypes of attitudinal positionings, and can provide useful insights into examining in what ways the evaluative description differs among hospitality websites across resort hotels, star categories, or well-established international brands.

It is important to recognize this study’s limitations, which, in themselves, indicate fruitful avenues for future work. First of all, Kress and van Leeuwen’s theory of visual grammar is a general framework, without targeting any specific industry or discipline. Given the absence of theory on the classification of hotel images, the current analytical scheme of representational meanings that is based on former research and the data collected in this work may be subjective, despite being proven as an effective way of categorizing the various image types. There is thus a need for additional research to refine and apply the current framework to other luxury hotel chains or star categories to better understand the possible image items manifested in hospitality websites. This can lead to the development of a standard image model within the hospitality field, which can not only reduce idiosyncratic image categories varying from study to study, but also build up a universal image item list to conduct a structured comparison of image representations across specific studies (cf. Pan & Li, 2011, on tourism destination image).

Second, the findings of this research are, to some degree at least, limited to the homepages of two Asian luxury hotel chains. Though verbal and visual elements on the homepages of hotel websites play a dominant role in projecting a strong brand image, it cannot be denied that the semiotic resources offered via the website hyperlinks will also contribute to identity construction. Future research should compile a larger data base including all the multimodal resources offered from the hotel-owned website to give a more comprehensive understanding of the brand identity induced via this platform.

The final limitation is that the visual analysis was mainly based on the researcher’s own perspective; consumers from different walks of life may generate different perceptions of the same image. How a hotel image is cultivated in the consumers’ minds can be explored either through a survey research about their views of the hotel-owned websites, or through investigation into their opinions presented on different web platforms, such as, social media, web chats and forums, and online reviews. As noted in the recent tourism literature (i.e. Llodrà-riera, Martínez-Ruiz, Jiménez-Zarco & izquierdo-Yusta, 2015), diverse web platforms, reflecting either supplier- or user-generated information, are often combined to exert influence on consumers’ brand associations, due to the growth of these platforms as major information resources. Such knowledge can offer practical implications for hotel marketing organizations to define and communicate preferred corporate images on different web platforms, since the identity projected by the corporate supplier could be distinct from the perceived identity in the potential consumer’s mind (Cai, 2002).

One caveat worthy of attention for future research is that both the hospitality companies examined in this work modified and revamped the appearance and infrastructure of their websites toward the end of 2015. A diachronic comparison of the old and new versions of the hotel homepages is thus strongly recommended, given that the changing nature of these verbal and visual representations suggests the brand image and marketing values they communicate also evolve over time.

7. Conclusion

Overall, this research sought to expand the knowledge base regarding brand identity construction and promotion in the hospitality industry. This study has offered insights into the growing number of research streams within entrepreneurship that endeavor to identify the communication strategies of corporate websites to promote international brands on a global scale (e.g. Degano, 2009; Ventola & Guijarro, 2009). Analyzing the deployment of semiotic resources can elicit a holistic understanding of brand identity representation, and act as the initial step for further in-depth understanding of the effects of such message delivery on consumers’ purchase intentions.

References

Bowcher, W.L. (2012). Multimodal Texts from around the World: Cultural and Linguistic Insights. NY: Palgrave Macmillan.

Cai, A. (2002). “Cooperative branding for rural destinations”. Annals of Tourism Research 29,3: 720-742.

Catenaccio, P. (2009). “Promoting medical tourism in India: Building identity and reputation in health tourism websites” in G. Garzone & P. Catenaccio (eds.), Identities across Media and Modes: Discursive Perspectives, 59-86. Bern: Peter Lang.

Cheng, F.W. & A.C. Hsu (2013). “A cross-cultural analysis of Japanese and American hotel websites”. Paper presented at the 6th International Conference on the Discourse, Communication and the Enterprise, Beijing, China.

Dann, G.M.S. (1996). The Language of Tourism: A Socio-linguistic Perspective. Oxon UK: CAB International.

Degano, C. (2009). “Marketing identities on Nestle’s websites” in G. Garzone & P. Catenaccio (eds.), Identities across Media and Modes: Discursive Perspectives, 189-214. Bern: Peter Lang.

Echtner, C.M. & J.R.B. Ritchie (2003). “The meaning and measurement of destination image”. Journal of Tourism Studies 14,1: 37-48.

Francesconi, S. (2014). Reading Tourism Texts: A Multimodal Analysis. Bristol: Channel View Publications.

Garzone, G. (2009). “Identity in tourist communication on the internet: Italy’s and Spain’s websites contrasted” in G. Garzone & P. Catenaccio (eds.), Identities across Media and Modes: Discursive Perspectives, 29-58. Bern: Peter Lang.

Gnoth, J. (2007). “The structure of destination trends: Leveraging values”. Tourism Analysis 12: 345-358.

Hallett, R.W. & J. Kaplan-Weinger (2010). Official Tourism Websites: A Discourse Analysis Perspective. Bristol, UK: Channel View Publications.

Jack, G. & A. Phipps (2005). Tourism and Intercultural Exchange: Why Tourism Matters. Clevedon: Channel View Publications.

Költringer, C. & K. Dickinger (2015). “Analyzing destination branding and image from online sources: A web content mining approach”. Journal of Business Research 68,9: 1836-1846.

Kress, G. & T. Van Leeuwen (1996). Reading Images: The Grammar of Visual Design, 1st ed. London: Routledge.

Kress, G. & T. Van Leeuwen (2006). Reading Images: The Grammar of Visual Design, 2nd ed. London: Routledge.

Krippendorff, K. (2004). Content Analysis: An Introduction to its Methodology, 2nd ed. Thousands Oaks, CA: Sage Publications.

Landis, J.R. & G. Koch (1977). “The measurement of observer agreement for categorical data”. Biometrics 33: 159-174.

Law, R. & C.H.C. Hsu (2006). “Importance of hotel website dimensions and attributes: Perceptions of online browsers and online purchasers”. Journal of Hospitality & Tourism Research 30,3: 295-312.

Llodrà-Riera, I., M. Martínez-Ruiz, A.I. Jiménez- Zarco & A. Izquierdo-Yusta (2015). “A multidimensional analysis of the information sources construct and its relevance for destination image formation”. Tourism Management 48: 319-328.

Lyon, S. & E.C. Wells (2012). Global Tourism: Cultural Heritage and Economic Encounters. Plymouth, UK: AltaMira Press.

Machin, D. & J. Thornborrow (2003). “Branding and discourse: The case of Cosmopolitan”. Discourse & Society 14,4: 453-471.

Martin, J.R. & P.R.R. White (2005). The Language of Evaluation: Appraisal in English. NY: Palgrave Macmillan.

Pan, B. & X. Li (2011). “The long tail of destination image and online marketing”. Annals of Tourism Research 38,1: 132-152.

Pine, P.J. & J.H. Gilmore (1999). The Experience Economy. Boston, MA: Harvard Business School Press.

Qu, H., L. H. Kim & H.H. Im (2011). “A model of destination branding: Integrating the concepts of the branding and destination image”. Tourism Management 32: 465-476.

Santulli, F. (2011). “Tourism brochures as a means to convey corporate image: The analysis of a case study in a rhetorical perspective” in G. Garzone & M. Gotti (eds.), Discourse, Communication and the Enterprise: Genre and Trends, 371-390. Bern: Peter Lang.

Samovar, L.A., R.E. Porter & E.R. McDaniel (2010). Communication between Cultures. Boston, MA: Wadsworth.

Scott, M. (2010). WordSmith Tools (version 5). Oxford: Oxford University Press.

Suen, O.Y.A. (2009). “Self-representation of five star hotels: A digital genre analysis of hotel homepages” in V.K. Bhatia, W. Cheng, B. Du- Babcock & J. Lung (eds.), Language for Professional Communication: Research, Practice & Training, 111-130. Hong Kong: The Hong Kong Polytechnic University, Asia-Pacific LSP and Professional communication.

Thurlow, C. & A. Jaworski (2010). Tourism Discourse: Language and Global Mobility. New York: Palgrave Macmillan.

UNWTO. (2015). UNWTO Tourism Highlights, 2015 ed. URL: http://mkt.unwto.org/publication/unwto-tourism-highlights-2015-edition// [10/01/2016].

Ventola, E. & A.J.M. Guijarro (2009). The World Told and the World Shown: Multisemiotic Issues. NY: Palgrave Macmillan.

Wirtz, J. (2012). “Banyan Tree: Branding the intangible” in J. Wirtz, P. Chew & C. Lovelock (eds.), Essentials of Services Marketing, 519-527. Singapore: Pearson Education.