ARTIGO LANGUAGE POLICY IN THE VIRTUAL LINGUISTIC LANDSCAPE: THE CASE OF THE KINGDOM OF BAHRAIN E-GOVERNMENT NATIONAL PORTAL

This study examined language policy in the Kingdom of Bahrain virtual linguistic landscape as exemplified in its E-Government National Portal. It aims at extending the view of the linguistic landscape beyond the mainstream research that is typically anchored and defined within the parameters of given geographic boundaries. It sought to answer the question of how the linguistic diversity is experienced in cyberspace. Spolsky and Cooper’s (1991) Preference Model Theory of language choice was used to address this question. The findings authenticate the assumption that the Bahraini virtual linguistic landscape shares not only specific features with physical counterparts, but also displays unique attributes.


INTRODUCTION
The main domain of investigation in the current study is linguistic representation, which includes linguistic reality, perceived reality, as well as desired reality. The strategic potential and political power of language in all its incarnations and underlying intents is the major focus of this study, which sheds lights on intention of governments or societies to ascertain or impose their own take on linguistic items by combining and framing various symbolic or immaterial linguistic and semiotic resources. These linguo-semiotic resources may be employed both in a consistent fashion, through concerted efforts of top-down official policies, and ad hoc, through bottom-up more or less spontaneous practices and grassroots linguistic activism. In this connection, the study concerns with the heterogeneity of linguistic voices represented in the Kingdom of Bahrain Virtual Linguistic Landscape (VLL, henceforth) as exemplified in its E-Government National Portal. It extends the Linguistic Landscape (LL, hereafter) framework to address the question of how the linguistic diversity is experienced in cyberspace-as-a-landscape, by exploring the existence and heterogeneity of the linguistic representation in the public sphere of electronic communication, which is here referred to as the VLL. It examines the nature of language presence in cyberspace-as-the-publicsphere, as exemplified in the Bahraini E-Government National Portal (BE-GNP, hereafter).
The VLL, in this study, is approached as a particular place or environment of language communication in cyberspace by extending the sociolinguistic discussion of the LL research to the domain of new media that embodies "Cultural objects which use digital computer technology for distribution and exhibition" (MANOVISH, 2010, p. 17). These objects include, among others, Internet, electronic portals, websites, computer multimedia, and computer games as well as social networking sites. Accordingly, the VLL, in this study, refers to language presence and linguistic diversity in cyberspace-as-the-public-sphere. That is, the saliency and conspicuousness of different linguistic elements representing ethnolinguistic and power relations in cyberspace.
In past years, government institutional websites were static webpages with constant content, displaying the same information to all visitor. The main goal of these websites, at that time, was neither to interact with the general public nor to provide access to services and facilities. They were employed to provide information about these institutions (e.g., mission, vision, statistics, news, location, and their service centers business hours). Due the lack of technological development, they were typical examples of one-way information provision. In present day technology, services previously provided by a public utility or a customer-oriented service agency, in many countries around the globe, are replaced by multifarious online services. In addition, the official e-government websites of these countries have begun to be a fertile ground for government-to-service consumers' communication. Service consumers are no longer required to be physically present to receive services. Ivković and Lotherington (2009) argue that these official websites are a constituent of the VLL, extending the scope of the LL to include the delineation of languages available in the digital public sphere. The term 'virtual' is not related to imagined or unreal landscapes. It refers to the non-physical linguistic representation in the public sphere of electronic communication and digital technologies.
LL and VLL are two domains of the linguaspace (i.e., language presence and linguistic diversity in physical and digital worlds). Semantically speaking, LL and VLL are hyponyms of the linguaspace. LL is "The language of public road signs, advertising billboards, street names, commercial shop signs, and public signs on government buildings combined to form the linguistic landscape of a given territory, region, or urban agglomeration" (LANDRY; BOURHIS, 1997, p. 25).
With reference to this definition, VLL is a visible representation of linguistic and language diverseness in the digital world. It is a sociolinguistic construct that offers new insights into studying language contact in the digitized world.
VLL shares some of the features with the LL. However, it creates its own characteristics and dynamics, which are afforded and constrained by the medium and modalities of communication in virtual space. Referring to this, Elleström (2010) argues that VLL is an extended form of the LL in the sense that today's communication is essentially transmedial and intermedial, taking place both across and in between media, with unclear and intertwined boundaries. In a general sense, any representation of linguistic activity in digital environments, in the public sphere of electronic communication or in cyberspace ˗ a mode of mediated and nonphysical interaction between human minds ˗ constitutes the VLL. Following the tradition of LL research, and focusing on communication in multiple linguistic codes in cyberspace, VLL is concerned with the indexicality of digital artefacts, and their transformative nature and potential to evoke and change opinions, beliefs, and attitudes about ethnolinguistic power relations, the latter manifesting in the public space of the digital sphere.  (2006, p. 9) defined LL as "Linguistic objects that characterize the public space." For them, LL data is "Any sign or announcement located outside or inside a public institution or private business in a given geographical location." Many previous studies categorize LL data with regard to the source or originator of the linguistic message and the direction of linguistic items. In general, they made a distinction between public official (public) and non-official (private) signs, depending on weather the item is produced or used by a public institution or by individuals or groups (BACKHAUS, 2007). If the message originator is a language management authority (e.g., a government; a religious and ethnic authority), with the power to impose and execute policies on language choice, the sign is categorized as a topdown item. On the other hand, if the message originator is a private entity, the sign is considered a "bottom-up" item. The "top-down" flow was described by Ben-Rafael (2009, p. 49) as linguistic items that "Start off from foci of public authority to reach 'common citizens," whereas the "bottom-up" flow "Consists of LL items that are produced and presented by countless actors who -as individuals or corporate bodies -generally sprout from the public." In addition, Landry and Bourhis (1997) argue that non-official (private) signs are advertising and commercial signs exhibited in public sphere, including business institutions, storefronts, billboards and message on private transport, whereas official (government) signs are those established and used by different levels of government, and include place and street names as well as various inscriptions on government buildings, hospitals, and public parks. In like manner, Calvet (1993, p. 7) distinguishes between "in vitro" and "in vivo" language management. In "in vitro" management, linguists analyze and describe languages and situations, then advance hypotheses and offer solutions to problems. However, in "in vivo", from the speakers 'linguistic feelings, that these solutions are either accepted or rejected.
It is crystal clear that the primary emphasis of early LL studies was on the analysis of cityscapes in multilingual public spaces. With the advent of new media and the introduction of increasingly multimodal communication, the LL framework calls for an extended perspective, theoretical and analytical, to account for a range of "information design modes, including images, sounds, words in mono, multiple and mixed, hybrids and fusions, reflecting different modalities and their interactions" (SHOHAMY; WAKSMAN, 2009, p. 314). Indicative of the de facto language practices, these digital (audio or written) postings, similar to the bottom-up items from the physical LL referents, reflect linguistic vitality and attitudes, and affect the global linguistic balance through language mixtures and hybrids. Multilingualism and its associated non-linguistic Internet co-referents (i.e., sounds, images, and colors) depicts the linguistic ambience and represent the position of a specific language among the available languages in the public sphere.
The top-down items, such as websites of a government, directly reflect the current language policy of a polity. The top-down linguistic item flow is "An instrument of language policy" (SHOHAMY, 2006, p. 185), confirming the present linguistic ecology, or striving to alter it by manipulating language choices and modalities of multilingual discourses. On the other hand, the bottom-up item flow "Represents de facto language practices and initiatives" (p. 181) created in public spaces, whereby "The public space, whether through signs, forms, instructions or cyberspace, is a very important arena where language battles are negotiated and crafted" (p. 129). Either guided by the concerns of deliver information or motivated by symbolic principles, to reflect, gradually but firmly establish or change opinions, beliefs and perceptions about the linguistic vitality and status of a particular language, ethnolinguistic groups or individuals, as autonomous actors, do alter the linguistic ecology (MUFWENE, 2008), foregrounding certain linguistic choices and influencing language attitudes. In cyberspace, because of the scope of Internet access, the potential impact of messages by far exceeds the influence of a graffito, message board, poster, or advertisement both in multilingual and in monolingual discourse (IVKOVIĆ; LOTHERINGTON, 2009).
The distinction between official (public) and non-official (private) signage is not necessarily clear. Spolsky (2009, p. 28) argues that "Both private and government signs can be governmentregulated, whereas the government signs can undergo local control to a greater or lesser extent." By the same token, Coupland (2010) questions the private versus government distinction, as some bottom-up, commercial, "Initiatives may have their own 'from above' qualities." He reconsiders the from above/from below division, proposing "To see all linguistic landscaping as generated 'from above,' if this means it [LL] is conditioned by language-ideological forces and strategies that find value in putting linguistic text into the visible environment for some particular purpose" (COUPLAND, 2010, p. 97), including department store notices, magazine covers, and bumper stickers. On the other end, according to Coupland ((2010), are vernacular text-types, such as bodily tattoos, hand-written small-ad notes in comer shop windows or the cries of street vendors, to which he refers as grassroot landscape initiatives. He goes on to note that ultimately even these initiatives are instrumental in fulfilling the agendas from below into the public sphere. Coupland's (2010) viewpoint emphasizes the interconnectedness of the official and non-official linguistic (and literacy) practices. However, one could also argue that the flow is bidirectional and that the vernacular-text types and grassroot landscape practices, and "the norms of the periphery" (BLOMMAERT, 2010, p. 80) may also impact and shape the top-down practices and initiatives. Similarly, Calvet and Brown (2006) argue that for understanding language policy in vitro, one needs to understand in vivo practices and trends, since "Any attempt to affect language and linguistic situations is going to be put into practice" (p. 13).

SIGNIFICANCE OF THE STUDY
The Worldwide Web, which was introduced in 1991 by Tim Bemers-Lee, undergone a radical change in human communication. At the present time, the Internet is a basic human right, and everyone has a virtual presence. In most countries, virtual inhabitants exist next to regular citizens.
With the advent of the Internet and online social computing, a distinct LL, in computer-generated and mediated environment, is emerging. This LL, which was referred to earlier as VLL, is shaped by bilingual/multilingual communication. It shapes unique LL in cyberspace. That is, a cyberspace extension of the the physical LL. It can be considered as a sociolinguistic construct that offers new insights into the studies of language contact in the digitized world. In this connection, the current study was carried out with the aim of extending the view of the LL beyond the mainstream research that is typically anchored and defined within the parameters of given geographic boundaries (e.g. towns, cities). It sheds light on bilingual/multilingual options afforded by cyberspace and its vast possibilities as a new public sphere. Besides, it investigates communication in the public sphere of human activity beyond the tangible materiality of the biological world, using the LL framework.
Specifically, it aims to describe and analyze the phenomenon of bilingual/multilingual communication in cyberspace, as exemplified in the BE-GNP. Gorter (2013, p 197) argues that " [LL] items are "Mechanisms of language policy that can perpetuate ideologies and the status of certain languages and not others." That is, the discernibility and visibility of various languages on the official websites of different countries is closely related to the recognition of these languages and their relative ranking in these countries. The current study attempts to investigate the interrelationship between official language regulation and language practices. It endeavors to prove that VLL can reflect social ordering as well as power relations in different countries. This is, by all means, is based on the assumption that linguistic representation, strategic potential and political power of language or linguistic code, in country-hegemonize VLL "Transmits symbolic messages as to the legitimacy, relevance, priority and standards of languages and the people and groups they represent" (SHOHAMY, 2006, p. 110). It should be noted here that the present study is not a comprehensive account of what the VLL, but it is a step in the direction of a comprehensive view of communication and language contact in cyberspace. It contributes to the research in the LLs and computer mediated communication in linguistics, offering insights into modalities of language contact and linguistic ecology online, including the role that English plays in cyberspace. It also may serve in raising metalinguistic awareness and in confirmation of linguistic identities, including tolerance for linguistic and cultural differences. Accordingly, this study may be of interest to linguists and to all those intrigued by the question of how diverse communication in cyberspace is. The official language in Bahrain is Arabic. It plays a major role in Bahrain politics. For example, anyone who wants to become a Member of Parliament must know Arabic. English is common and most-well-spoken in Bahrain. It is a mandatory second language in educational institutions. A considerate percentage of the population converse in Urdu or Persian. Nepali is used by the Gurkha Soldiers and Nepalese workers. Hindi, Bangla, Tamil, and Malayalam are also used among numerous Indian communities. All road signs and commercial institutions in Bahrain display both the Arabic language and English (GOMAA, 2017). English is quite widespread in Bahrain. It is used in business and is a compulsory second language in schools. Although state radio and television are broadcast primarily in Arabic, newspapers and magazines in English are available (e.g., Al-Ayam, in Arabic, and the Bahrain Tribune and the Gulf Daily News are English).

INFORMATION AND COMMUNICATION TECHNOLOGY IN BAHRAIN
According to the Internet World Stats in 2018, Kingdom of Bahrain is one of the world's most technically advanced countries for providing internet access and using digital services, with a high rate of Internet penetration. It has achieved advanced positions in the United Nations report, preserving the top ranks for the fourth consecutive time since 2010. Besides, it was positioned first in the Arab region within the e-Government readiness by presenting numerous online services to its citizens and residents. In current years, the Bahraini government managed to actually have an ideal vision of how the government should be run. It decided that information technology is the one thing that helps the country organize itself. Therefore, it has put nearly all its government services online.
99% of its services are easily accessible online (e.g., visas, licenses, permits, payments). Virtually, everything can be done digitally online. The Bahraini government institutions have expeditiously been moving a substantial part of their contact with the general public to the Internet via the BE-GNP, which is a gateway to its information and services. Making life easier is the main goal of this egovernment portal that represents a virtual public sector in which technologically mediated communication between the government and the general public. Allocating space and visibleness to certain languages on this portal reflects a symbolic structure of this virtual public sector.

THEORETICAL FRAMEWORK
In this study, the VLL is viewed as a sister sub-field of the LL within a common field of sociolinguistic inquiry into language presence and language use, linguistic diversity, bilingual/multilingual communication and linguistic code preference in cyberspace. To this end, the current study seeks to answer the question of how the linguistic diversity is experienced in cyberspace-as-a-landscape as exemplified in the Bahraini E-Government National Portal (BE-GNP, hereafter). The Preference Model Theory of language choice (PMT, hereafter) devised by Spolsky and Cooper (1991) was used to address this question. The PMT is geared towards unravelling the underlying processes that motivate the choice of languages, their relative salience and the processes that govern production and reception of multilingual messages, including unveiling their hidden meanings. The framework of this theory developed originally by Jackendoff (1983) for semantics, postulates that certain rules within a system need not constitute necessary conditions but may admit exceptions. These non-committal rules can be defined in terms of likelihood of choice and as a continuum of typicality. As the choices are not absolute, but are based on gradient conditions of probabilities, such system may easily contain two contradictory conditions, while varying ideologies or situations will often account for the weighting given to each, and so explain the probable outcome in a specific case. On this view, public signage may weave together two or more sign types and address multiple purposes, such as providing information and advertising the product, whereby one or the other function is dominant. To account for the language selection on public signage, Spolsky and Cooper's (1991, p. 81-84) propose the following three rules of signs (which will be exemplified and explained in the analysis of the BE-GNP: Rule 1: 'sign-writer's skill' condition, or write signs in a language you know; Rule 2: presumed reader' condition, or prefer to write signs in the language or languages that intended readers are assumed to read; Rule 3: symbolic value' condition, or prefer to write signs in your own language or in a language with which you wish to be identified. The 'sign writer's skill' rule posits a necessary condition, contingent on one's knowledge of languages. However, this rule has a relative value in the case of online writing, which allows for machine translation, such as Google Translator, with "Translation assistance easily accessible, if not very accurate" (IVKOVIĆ; LOTHERINGTON, 2009, p. 19). The outcome is an indication of the sign-writer's-rather than that of the sign-owner -linguistic repertoire. The other two rules are potentially in conflict since they are based on typicality conditions. Specifically, the 'presumed reader' condition is concerned with information delivery and addresses the expectation and the linguistic repertoire of the desired or presumed reader, whereby the principal motivation is economic and pragmatic. In contrast, the 'symbolic value' condition, as a rule, is less influenced by pragmatic concerns.

METHODOLOGY
The focus of this study lies on the BE-GNP which represents the actual online practices. It is the main source of data. It was chosen because it is central to the interaction between the Bahraini government and the public. It is a means of communication between the Bahraini authorities and most of the Bahraini adult population in the kingdom of Bahrain (e.g., the employed, the unemployed who receive social benefits, applicants for a residence permit, etc.). In this study, the BE-GNP is seen as a "major discourse node" and its public language use is supposed to "Contain traces of ideologies and also to have an impact on their reproduction" (AJSIC; McGROARTY, 2015, p. 185). Accordingly, LL analysis is used in to investigate language policy in cyberspace by examining the visibleness and roles of different languages displayed in the BE-GNP. This reflects, illustrates and sheds light on the local language legislation and policy in the kingdom of Bahrain.

ANALYSIS AND DISCUSSION
As indicated earlier in section 5, Spolsky and Cooper's (1991) PMT was applied to examine the language selection displayed on the BE-GNP. The analysis was focused on two levels of the BE-GNP building design: the start page/home, in all available languages, and the other pages referred to through hyperlinks on the start page. For the purpose of the analysis, screenshots were treated as signs, or sign vehicles, comparable to plates in LL research (e.g., SHOHAMY, 2006). While in physical space, it is common to call the written texts on street plates and billboards displaying message signs, in virtual space it is unconventional to refer to a text displayed on a computer screen or hand-held digital device as a sign. Therefore, in this study, the term 'snagit' is used as a digital equivalent of a notice that is publicly displayed to give information or instruction in a written form.
Accordingly, a screenshot constitutes a snagit that is bounded digital representation. Depending on the mode of this representation, a snagit can be linguistic, pictorial, or multimodal (i.e., linguistic and pictorial). Figure 2 shows a monolingual (Arabic), and monoscriptal (Arabic alphabet) snagit, taken as a screenshot from the home page of the BE-GNP, whereas Figure 3 shows a monolingual (English), and monoscriptal (Latin alphabet) snagit, taken as a screenshot from the home page of the BE-GNP.

Figure 3 -A Monolingual (English) Snagit of the BE-GNP
The analysis process began with applying the rule of 'presumed reader,' which concerns with information delivery (i.e., the pragmatic function of the message), then 'the symbolic value' rule to the language selection, and finally the analysis of English as a marked language choice, testing it against all three rules, including the necessary, 'sign-writer's skill,' condition. The analysis demonstrates that Modern Standard Arabic (MSA, hereafter), as the primary Bahraini national language according to Article 2 of the Bahraini constitution, is used in the BE-GNP front page. In the top left corner of the page, as figure 2 indicates, there is a hyperlink to an English language version.
The choices of the versions in MSA and English meet the rule of the 'presumed reader,' as well as the gradient condition of the rule, in the order of presumed importance of the languages ˗ that is, MSA followed by English ˗ displayed on BE-GNP. As the main consumers of the information originating from the Bahraini government are the domestic readers, the choice of MSA satisfies the 'presumed reader' rule since it is the Bahrainis who are most likely to be interested in the BE-GNP content. In addition, a primarily pragmatic feature of the MSA use ˗ the presumed reader condition ˗ is that it includes both the in-group and out-group readership of the Arabic language in other Arab countries.
By a large majority, they speak the language and can read and write in Arabic. With regard to the "presumed reader" gradient rule (Rule 2), the MSA version on the BE-GNP is considered dominant since it is the language of choice in most Arabic Internet discourse. The choice of MSA is also motivated by the potential out-group readers (i.e. citizens of the Gulf Cooperation Countries) who, although not residents of the Kingdom of Bahrain, use MSA.
Similarly, the choice of an English version for the BE-GNP is guided by the role of English as common and commercial tongue among peoples of diverse speech, addressing the native and non-native Arabic speaking readership. Besides, the availability of an English version in the Bahraini government VLL, via the BE-GNP, can be explained by the reason that English is regarded as a lingua franca. It is the language of the first monolingual cyberspace, in which it has a unique status. Referring to this, Young (2016) argues that English was the first language used on the Internet and by the mid-1990s, it was estimated that it made up 80% of the content. It is the only hyper-central language in De Swaan's (2013) global language system that places languages in a four-level hierarchical structure: peripheral, central, supercentral and hypercentral.
The English version of the BE-GNP accords with the principles stated in Rule 2, since information in English pragmatically caters to non-Arabic speakers, who are interested in the information originating from the Bahraini government. Therefore, the information displayed in English draws on a considerably bigger readership pool. It should be noted here that Gomaa's (2017) argues that the Bahraini LL displays two verities of Arabic (i.e., MSA and Bahraini Arabic) and English together with seven different languages, namely Bengali, Chinese, Hindi, Kannada, Malayalam, Tagalog and Tamil. Nevertheless, the BE-GNP does not offer versions in the later seven different languages. In the absence of language versions in these seven languages, the the BE-GNP English version can meet 'the presumed reader' rule. It is important to note that in her investigation of linguistic diversity on global organizations websites, Kelly-Holmes (2013, p. 135) concluded that English is widely used to "fulfill the impossible task of addressing everyone in their own language." In like manner, the English version of the BE-GNP is used to address various linguistic groups in the country. This implies that the linguistic rights of different linguistic groups in the Kingdom of Bahrain are protected despite the fact that their languages are not included in the portal, which has become a virtual service center and an important medium of communication between the Bahraini government and the public.
The third rule is concerned with the symbolic and indicative value of language choice, whereby the display of a specific language can send a certain message. Spolsky and Cooper (1991, p. 84) maintains that: The motivation for the third rule is primarily political or socio-political. It derives its value from a desire to assert power (by controlling the language of the sign, I declare power over the space designated) or to claim solidarity or identity (my statement of socio-cultural membership is the language I have chosen).
Accordingly, I believe, the availability of an English version on the BE-GNP is not surprising because it is the second or foreign language of approximately half the individuals who access the portal. In addition, the choice of English indicates that Kingdom of Bahrain supports the English version as a form of high-level institutional activity to promote the English language with a goal to enforce and implement the Sustainable Development Goals (SDGs, hereafter) that call for overcoming linguistic barriers. Besides, the presence of the English version sends direct message with regard to the centrality of English in Kingdom of Bahrain. It indicates that non-Arabic speakers who live in the Kingdom of Bahrain are not overlooked and ignored. This, definitely, conveys a positive symbolic message that reflects the efforts of the Bahraini government to implement the SDGs.
It is worthwhile to mention that language choice plays a vital role in shaping the VLL by providing information related to which has power and influence and which has not. That is, the existence or non-existence of a particular language in a certain way transmits straightforward or implied messages with respect to the centralization versus the marginalization of this language. The visibility of language reflects its legitimacy, importance, primacy and the status of the individuals and communities it represents. In addition, ordering of languages on a web page can be compared to a display of languages on signs on government buildings and official documents, where commonly care is taken to secure the status of national languages. These linguistic choices influence perceptions about official language status and linguistic vitality of those who pass or enter government buildings in physical space or visit its portal in cyberspace. Therefore, texts in multiple codes, including different languages and scripts, requires making decisions among available choices. Cenoz and Gorter (2006) claim that expressing code preference and reading/writing hierarchy may include the the order left to right, top to bottom reading, and from the center to the periphery, size of text and font type.
Accordingly, the Arabic and English versions of the BE-GNP reflect the current Bahraini language policy requiring equal representation of both official languages in both public and cyber spaces. The equally available amount of information in both versions serves both informative and symbolic functions. In relation to language policy, these functions may include, among others, maintain and using language, affecting language status, and determining which languages are cared for and protected. In other words, language policy and planning decisions have a significant influence on language vibrancy and, eventually, on the individual rights.
To further enhance communication and be a more effective virtual communicator, fostering the implementation of the SDGs and the concept of "no one is left behind", the BE-GNP can offer information in 9 languages (i.e., Arabic, Bengali, Chinese, English, Hindi, Kannada, Malayalam, Tagalog and Tamil). These languages can be indexed with the images of flags coupled with a caption either in the country's respective language in the following order (e.g., ‫العربية‬ for Arabic, English for English, বাঙালি for Bengali, മലയാളം for Malayalam, தமிழ் for Tamil, ಕನ್ನ ಡ for Kannada, ह िं दी for Hindi, Pilipino for Tagalog, 中文 for Chinese,), or in English (e.g., Arabic, Bengali, Malayalam, Tamil, Kannada, Hindi, Tagalog, and Chinese). The suggested order is based on the results of Gomaa's (2017) study which concluded that MSA is the language with the highest visibility in the Bahraini LL, followed, in order of importance, by English, Bengali, Malayalam, Tamil, Kannada, Hindi, Tagalog, and Chinese. In like manner, the flag of the Kingdom of Bahrain can be used to index the Arabic version of the BE-GNP, followed by either the UK or USA flag indexing the English language version, flowed by the flags of Bangladesh, India, the Philippines and Chinese.
In addition, a customized visitor counter can be placed on the different language versions of the BE-GNP to monitor its traffic and make sure that it is doing well and having a visibly solid reputation.
This also would build credibility for the BE-GNP as a reliable source of information and service.
Besides, the material in all language versions should not be static. That is, all versions must present updated content targeting domestic and international visitors. Regular or even daily update should be performed. By providing information in nine different languages, the BE-GNP meets the real intercommunication needs of its users.

CONCLUSION
The study extended work on LL in new direction, offering insights into the character of linguistic involvement in digitally-mediated communication. It extended the view of LL beyond the mainstream research, which is bound to be defined by geographical spaces (e.g., cities and states), focusing on language diversity and the choice of cyberspace as a new public nature. The concept of public sphere has been examined in the context of communication in virtual, that is, electronic space.
The main objective of this study was to offer a view on the VLL as a phenomenon in its own right. It investigated the digital sphere of public communication framed as a unique prototype of the LL. To this end, the study borrowed the LL sociolinguistic concept to examine linguistic diversity in the digital world. It has been argued that the LL framework provides a solid basis to throw light on language presence and linguistic diversity in cyberspace-as-the-public-sphere. The results substantiate the assumption that VLL, defined as presence or absence of different languages in cyberspace, shares not only specific features with physical counterparts, but also displays unique attributes.
In addition, the study examined the language policy on the BE-GNP. It answered the question of how the Bahraini national policy is implemented in cyberspace. The results indicate that the Bahraini public sector VLL has become bilingual. This is obvious on the BE-GNP, where Arabic and English have an increasing linguistic presence. As for the status of English, which is regulated by official legislation in online government communication, the study reveals it is an important constituent of the entire linguistic verities in the Bahraini community. This, most certainly, reflects policy-related concern and sends a sociopolitical message that it is the default foreign means of online government communication with diverse language groups who do not have sufficient Arabic communication skills. Besides, the availability of two linguistic options (i.e., Arabic and English versions) on the BE-GNP send out a clear message to Bahraini citizens and residents regarding the social position and significance these two languages in community.