DOI: 10.184132313-8912-2015-1-4-35-46



This article aims to place the Arabic language in its recent historical context and proposes to describe the situation of Arabic in the period preceding and leading to the Arab Spring from the perspective of the degree to which Arabic language change and variation are moving in the direction of more fusion or diffusion. By diffusion I mean a situation in which divergences among the dialects of one language continue to grow and fragment, causing them eventually to develop into separate and largely mutually unintelligible systems. Fusion, on the other hand, is a process where dialectal variations in one language contract and gain wider acceptance, bringing more vitality for the Standard dialect at the level of social use and resulting in higher levels of mutual intelligibility among the dialects. I dedicate Part One of this paper to an overview of nomenclature and the ideological controversies surrounding Arabic language variation and where this variation is heading. In Part Two, I review the various arguments advanced by a substantial number of researchers who are of the view that forces of Arabic diffusion are solid enough to lead Standard Arabic and the dialects in the direction of a growing chasm. Part Three is the antithesis of Part Two and represents the major contribution of this paper. Here, I argue, based on the literature review and on oral and textual observations and analyses, that the forces of fragmentation notwithstanding, the changes that have obtained since the post-independence era at the level of rates of literacy militate for consolidating the role of Standard Arabic, for increased intelligibility among the dialects, and for closing the gap between the dialects and the Standard, albeit slowly, especially with the increased use of Arabic as one medium of expression of the Arabic Spring, on the Web and on the street.

  1. Background

The publication in 1959 of Ferguson’s Diglossia opened the gates on a plethora of Arabic sociolinguistic studies and compilations (Altoma, 1969; Badawi & Hinds, 1986; Blau, 1977; Fishman 1967; Holes, 1987; Maamouri, 1967; Shubashy, 2004, etc.). ‘Diglossia’ marked the start of an era in which Arabic linguistic scholarship enlarged its purview to include not just philological, stylistic and structural aspects of codified Arabic, but also its functional and dialectal dimensions.  Between 1959 and 2011, three major transformations unfolded with crucial impact on the standing of Arabic. The first is the rise in rates of Arabic literacy among Arab populations as of the 1950s and 60s. The second dates back approximately to the mid-1980s and refers to what Ong (1988) calls the ‘technologizing of the word’, the emergence of word processing in Arabic and the subsequent localization of the web in Arabic around the mid-1990s.   These changes saw a transition from massive illiteracy to wider access to digital Arabic literacy mediated by social networks, and manifest in the third transformation, the Arab Spring. 

This is an argumentative paper with something of a historical perspective on where Arabic is heading. While it attempts to develop a thesis on Arabic language variation and change, it goes beyond technical description of form to adduce evidence from a variety of fields.  Throughout, my claim is that Arabic, since the fast-changing post-independence contexts, has moved along a path of coalescence and convergence between its varieties including the Standard, rather than a route of divergence and shift. 

I dedicate Part One of this paper to a review the arguments advanced by a number of researchers claiming that forces of Arabic diffusion are solid enough to lead Standard Arabic and the dialects in the direction of fragmentation. Part Two is the antithesis of Part One and represents the major thrust of this paper. Here, I argue, based on the literature review and on observations and analyses of internet materials, that the forces of fragmentation notwithstanding, evolution of Arabic since after independence suggests a consolidation for the role of Standard Arabic and increased intelligibility among the dialects, assisted by growing use of Arabic as the medium of expression of the Arab Spring.

With this introduction, we proceed to defining two important linguistic processes at work in relation to this nomenclature:

  • In his analysis of the forces impacting the future of English as a world language, Raddaoui (1988) distinguishes two forces pulling Received English and its many dialects in opposite directions, which he calls ‘fusion’ and ‘diffusion’. Such forces include the mass media, educational institutions, language academies, government institutions, the publishing industry, and other socioculturally dominant forces, which operate as locomotives for determining whether dialects of one language come together or move in disparate directions.  Below are working definitions for fusion and diffusion in relation to Arabic:
  • Diffusion is a situation in which divergences among Arabic dialects and MSA continue to grow, causing them eventually to develop into separate and largely mutually unintelligible systems. 
  • Fusion is a process where variation between MSA and the dialects contracts and diminishes, leading to increased mutual intelligibility among the dialects and the consolidation of MSA as a rallying point, indicated by mounting social acceptance and use.   

Contending as I do that that there are strong, rallying forces at work causing Arabic to move slowly but steadily into the direction of coalescence and fusion rather than on a path of shift, is, ideologically speaking, an uncomfortable and contentious position to hold.  In the next sub-section, I survey the causes of this discomfort and expand on the issues attendant upon this controversy. 

  1. Diffusion arguments

The arguments I articulate in this section emphasize the distance among Arabic dialects and between the dialects on the one hand and MSA on the other.   This distance is presented as forebodingly enormous to a point that precludes general communicative ease and educational viability. Let us call this the generalized distance thesis. I start this sub-section with a survey of emotive arguments followed with a review of linguistic arguments corroborating diffusion. I then proceed to querying the literature from a psycholinguistic perspective.  Finally, I move to the wider sociolinguistic plane, where this distance manifests itself at the relatively new channel of communication, the internet. 

First, a commentary of the kind of discourse employed in expounding the distance thesis is in order. Other than the term ‘distance’, which comes with Ferguson’s initial characterization, Salameh (2011) refers to the situation as a “deep chasm” between MSA and the dialects (p. 56). Owens (2001) equates learning MSA by its native Arabic speakers with learning a second language (p. 426).  Borrowing the term from contexts similar to Canada’s bilingual society, Amer, Adaileh and Rakhieh (2011) describe the Arabic linguistic situation as one of “cultural unity within linguistic diversity” (p. 19). Edward Saeed (2004), one of the staunchest defenders of causes Arab, is brought into this mix, and pronounces MSA to be "equivalent of Latin, a dead and forbidding language" (cited in Salameh, 2011). 

These characterizations of the relationships between diverse varieties of Arabic and their communities of speakers are not confined to the linguistic/sociolinguistic register.  As these depictions depart from the diglossic register, they start drawing on a wholly new lexicon mostly akin to mental dysfunction, where, in the words of Shubashy, diglossia is “crippling the Arab mind and stunting its capacities” (cited in Salameh, 2011). Far from being a medium in which social and communicative functions are transacted among Arabs, diglossia represents a situation of “pathology, schizophrenia and incoherence” (p. 53). 

2.1 Emotive arguments for diffusion

Emotive arguments refer to thoughts, perceptions, and feelings of an impalpable, unobservable, and qualitative nature.  The feelings reportedly associated with MSA/CA are generally negative and do not index the proximity and identity values normally associated with what a person or a community considers their own language. A 2003 United Nations report writes the following about MSA: MSA is “not the language of cordial, spontaneous expression, emotions, daily encounters, and ordinary communication (…) It is not a vehicle for discovering one’s inner self or outer surroundings” (cited in Salameh, 2011, p. 54). Chouairi (2009) illustrates this situation by commenting on the dubbing by Arab television stations of Japanese cartoons in MSA, and writes that these cartoons are “not naturally appealing”, that they border on the absurd, and that they are not capable of generating laughter among children who get bored very quickly while watching them (p. 41).  In contrast, she notes that comedies presented in the dialect by such renowned actors as Dureid Laham of Syria, make both “children and adults giggle and laugh” (p. 41). Chouairi does not provide a direct answer for her question on how long H will survive in the media. While she declares her love for classical Arabic and its literature (p. 1), she is of the view that CA is far from being a viable tool for emotive identification. Let me and rationalize this lack of identification with CA by bringing to bear a number of additional arguments, this time drawing upon the linguistic domain.

2.2 Linguistic distance

Linguistic distance between MSA/CA on the one hand and Arabic dialects on the other has been the focal point for research in the Arabic diglossic tradition. This distance covers aspects pertaining to language forms, notably the phonological, morphological and syntactic levels. It is not the purpose of this paper to provide a comprehensive account of these differences, but for the sake of illustration, I briefly discuss two aspects: phonology and lexis. 

 Contrasting phonological systems, Chouairi (2009) makes the obvious note that no inventory of sounds in any Arabic dialect is in full correspondence with that of MSA, and this applies both to the consonant and vowel systems. Lexis is another area where MSA-Dialect distance is reportedly high.  Gumperz (1964) argues that comparative study of the differences between MSA and dialects reveals crucial divergences in how new lexis is added to the language: While MSA tends to draw into its past in order to add words to its lexicon as a response to the requirements of modern life, the dialects continue to integrate loanwords (p. 423). As the lexical base of each expands, communication and equivalency between the two suffers, indicating they are embarking on a path of linguistic divergence. This linguistic distance is not without its ramifications on other levels, including that of language learning. 

2.3 Psycho-pedagogic arguments for diffusion

Psycholinguistic distance between the dialects and MSA/CA impacts learnability and teachability of Arabic as a native and foreign language.  This argument is mainly advanced by Ayari (1996), Maamouri (1998), Salameh (2011) and Ibrahim (2009). Its gist is that the linguistic gap between the spoken varieties and MSA is responsible for functional illiteracy in the Arab world, and for the difficulty encountered by children when learning to read. Maamouri (1998) argues that instead of reading to learn, Arab children spend their time learning to read. 

It is possible that the difficulty encountered by native Arabic speakers in learning MSA has the contrastive analysis hypothesis (CAH) of the 1950s as its reference point.  CAH deems learning ease to be a function of large similarities and small differences between the learner’s L1 and the target language. Though CAH was mainly used to apprehend the learning processes for second/foreign language learners in terms of transfer and interference between L1 and the target language, the terms of this hypothesis are used to describe the relationship between Arabic dialects and MSA, making these two dialects of Arabic analogous to two unrelated languages. Dakwar (2005) quotes a seven-year-old girl as saying: "Ammiya and Fusha do not differ much, that's why it is easy. I think in Ammiya before I write, I later transfer to Fusha. Sometimes, while reading I feel I am going back to Ammiya” (emphasis mine) (p. 92).  Diem (1974) goes so far as to treat that MSA chunks infiltrating dialectal speech as ‘interference’ (cited in Owens 2001, p. 426) thus equating MSA to a foreign language. 

Another tenet of CAH is that L1 forms are habits to be unlearned and hurdles to be overcome. Dakwar (2005), reports that when Jewish children who learn Palestinian Arabic at elementary school later start learning MSA, their teachers urge them to let go of their dialectal acquisitions so as to minimize interference, thereby stipulating that MSA and the Palestinian dialect are two unrelated languages. Shubashy (2006) points out that foreigners with high proficiency levels in MSA fail to see the connection between their formal acquisitions and Colloquial Egyptian Arabic: “they don’t understand a single word I say in that language” (emphasis added) (cited in Salameh, 2011, p. 56). Shubashy (2004) thinks failure to implement MSA reform in the direction of simplification is subjecting the Arab child learning MSA to “suffering untold” (p. 45). This is visible in the “increased reluctance among the youth to learn the complicated rules of the language and the outdated words and phrases that are no longer fit for the modern person to express themselves (my translation) (p. 52). 

Psycholinguistic research adds credence to these pedagogic hunches about how native Arabic speakers cognitively engage the task of learning MSA. Ibrahim’s empirical study (2009) seeks to determine the linguistic distance between Palestinian Arabic and MSA by finding out whether Palestinian Arabs learning both MSA and Hebrew combine the lexical forms of Palestinian Arabic and the forms of MSA in a single lexicon in the brain or whether they access them as separate lexicons as they do for Arabic and Hebrew. His conclusion is that “the status of LA [Literary Arabic] is similar to that of Hebrew and is consistent with the typical organization of MSA in a separate lexicon. Thus, learning MSA appears to be, in some respects, more like learning a second language (p. 96). 

Elgibali (1996) emits an interesting proposition on the nature of Classical Arabic as a language type.  Noting the difficulties Arabic dialect speakers encounter while acquiring the Standard, Elgibali points to “the inherent qualities of the Classical variety itself”. He explains that “if Classical Arabic is essentially an eclectic composite, then it is not a natural language” (p. 12). Elgibali does not expand on the notion of “eclectic composite”, nor does he clearly qualify Arabic as an “unnatural”, or “partly artificial” language. Chouairi (2009) weights in on this issue and writes: 

…classical Arabic (H variety) has a lexicon built from all the languages that bordered Arabia: Syriac, Egyptian, Bedouin Arabic, Greek and Persian while its grammar is a constructed, conscious grammar that does not lead itself to natural speech since it was formulated by linguists and writers (p. 42).


2.4 Sociolinguistic arguments for diffusion

The gist of the sociolinguistic argument in support of the distance thesis, probably a consequence of MSA’s reported failure to flow naturally, harks back to the nonexistence of a speech community for MSA, and its non-viability as a means of communication.  This argument sits on Malinowsky’s (1923) dictum that the proper study of language should be conducted “against the background of human activities” (cited in Chouairi, 2009, p. 40). Chouairi agrees that the notion of an Arabic speech community is appraised against a kind of “non-situated theorizing”, by which she means that “Classical Arabic H is absent from current human activities other than reading and writing” (p. 40). Salameh (2011) agrees with a United Nations report that [Classical] Arabic “has in effect ceased to be a spoken language” and is a “largely a learned, cultic, ceremonial, and literary language” (p. 54). Since it is limited to the school, and is not spoken naturally, the argument goes, it has little social relevance. Salameh (2011) quotes the doyen of Arabic Letters, Taha Hussein, as saying, “Nobody speaks it [the Standard] at home, [in] school, [on] the streets, or in clubs; it is not even used in [the] Al-Azhar [Islamic University] itself” (p. 51).  From this vantage point, CA has no actual speakership even among its most loyal guardians, the renowned Al-Azhar Islamic University. 

This reported irrelevance of CA in social life spills into its modern offshoot, MSA, as having little if any existence as ‘speech’, ‘parole’, or ‘performance’. Thus presented, CA and MSA are thus mere “competence” in the heads of linguists, and have no social basis. This reasoning leads to an interesting conclusion that the relationship between the dialects of Arabic and the H variety cannot be described in terms of the former being dialects of the latter: dialects “are not dialects of H because H is spoken nowhere” (Chouairi, 2009, p. 38). In other words, Arabic dialects are not, sociolinguistically speaking, related to CA/MSA.

 Salameh likens the distance between MSA/CA and the dialects to the distance between French and other Romance languages on the one hand and Latin on the other (2011, p. 48).  He uses mutual unintelligibility as a carbon test of the distance. Abdelali (2004) and Chouairi (2009) distinguish Moroccan (or North African) Arabic, Cairo Arabic, North Syrian (Levantine) Arabic and Gulf Arabic as “separate languages”. Salameh (2011) states that “Egypt has an Egyptian language; Lebanon has a Lebanese language; the Hijaz has a Hijazi language” (p. 51), and goes so far as to talk of relative unintelligibility between Bagdadi and Damascene Arabic. 

Finally, the traditional domains of language are the family, the school, the workplace and society at large. Now, we have to supplement these spaces with the new setting of the internet as an emerging medium.  Here, the extent of the gap has yet to be gauged between users of written dialectal Arabic, Written Standard Arabic, and Latin-scripted Arabic. Which social variables determine which form of Arabic, including Arabic-French code switching, and how much felicitous communication is taking place among users of these varieties is another area of inquiry awaiting investigation.

Thus, according to the generalized linguistic distance thesis, Arabic dialects and MSA invite different typological, pedagogic, emotional, linguistic, and socio-linguistic descriptions and conclusions. Consequently, fragmentation and diffusion are suggested as a projection of the distance thesis.  

3.0 Arguments for fusion

In the final section of this paper, I discuss, from a number of perspectives, the factors I consider to have caused Arabic varieties to embark on a convergence path. Here, I do not at all seek to take one-by-one the arguments put forward in the previous section and empty them of substance, but rather to draw attention to new trends and emerging data with significant impact on the direction of Arabic.

Back in 2001, Owens asked the question about why either of the two levels of speech, MSA or the dialects, has not yet become the unique medium of communication in the Arab world.  He attributes maintenance of both not to language structure, but to the levels of social and political motivations: 

Should these become impelling enough SA would doubtlessly become the spoken norm throughout the Arabic world. Lacking such motivation, however, and at present there are probably as many reasons for maintaining NA [Native Arabic] as for adopting SA [Standard Arabic], diglossia will continue to prevail (p. 449). 


Regardless of such motivations though, a host of actors have been at work, yielding a certain degree of linguistic change in the direction of proximity rather than distance. It is however not in the nature of change to be readily observable.  This is all the more true as the ‘diglossic frame’ has remained the most important paradigm for examining Arabic language for over half a century. I propose that this frame, through its grip, appeal, and the more than critical mass of research it has spawned, has deflected attention away from change. The question I wish to address now is this: what are these new topographies that the diglossia prism has failed to register and react to? In answering this question, I should like to begin by suggesting that we reconceptualize the notion of diglossia itself.

3.1 Reconceptualing diglossia

Albirini (2011) is one of the first to call for a drastic reformulation of the construct of diglossia through reexamining the initially reported division of labor between MSA and the dialects.  Analyzing naturally-occurring spoken data from religious speakers, political debaters and soccer commentators, he finds that “speakers create a functional division between the two varieties by designating issues of importance, complexity and seriousness to SA [Standard Arabic], the High code, and accessible topics with DA [dialect Arabic], the Low code” (p. 537).  He concludes that it is not context that determines use of either one or the other, but function: the use of H is not determined by where a person is speaking, say in the parliament, in the court or at the mosque, but selection of formal/informal register depends on the effect a particular chunk of speech is meant to achieve; speakers typically use dialectal Arabic if they wish to “downplay a particular segment of the discourse” but will shift to Standard Arabic to highlight the importance of a segment of discourse (p. 547) even when they are in the same event and in the same context. Thus analyzed, use of the two varieties in the same text and the same context represents a weakening of the diglossia frame.

Additionally, instead of apprehending the co-presence of H and L in terms of code-switching or interference, we could argue that what is taking place is an emerging hybrid mode of address encompassing the dialect and the Standard. This hybrid mode, applies not just to macro-level analyses of extended oral transactions, but also to the micro-level where smaller chunks indicate that the Standard and the dialect co-occur within the confines of one word or a phrase. Examples abound, but for the sake of illustration, let us cite one example discussed by Owens (2001): 

  1. /reet/:  Cairene dialect form: gloss:  I have seen
  2. /ra’aytu/:  MSA form: I have seen
  3. /ra’eet/:  crossover between Cairene dialect and MSA (p. 432)

While forms 1 and 2 present us with instances of the Dialect and the Standard respectively, variation 3 is problematic because it is a composite of features: from the Standard, we have the stem /ra’/ and from the dialect, the suffix /eet/. The diglossic framework is incapable of handling this and similar phenomena. Owens (2001) suggests that form 3 is an emerging variation resulting from increased contact between two varieties of the same language producing a typical Educated Spoken Arabic hybrid. Space does not allow for citing more such examples, but based on the evidence, we can conclude that diglossia defined as a ‘stable’ situation where two varieties exist side by side, is not applicable in a strict sense as it assumes the formal variety and the dialect are moving on parallel tracks, without intersecting and affecting each other. The fact of the matter is that H and L are interacting and producing a new breed combining structural features from both codes.      

There are additional grounds for putting into question the reportedly sizable lexical and structural gap between the dialects and MSA. Owens (2001) considers the lexical overlap between native Arabic and MSA and concludes, overall, that these are 80% similar (p. 449). In his study of the grammar of MSA and Cairene Arabic, McKay (1972), adopts Transformational Grammar as framework, and finds that there are no significant differences between the surface and deep structures of the following constructions: simple equational sentences, sentences embedded by relativization, sentences with verb-initial or verb second order, complementation using /?anna/, and direct and indirect questions (Pp. 29-30).

The above are some lines of argumentation that should be usefully researched in the future to gauge the ability of the diglossic prism to handle at least a spectrum of the data. In the following sub-section I reappraise the perceived psychological distance between Arabic speakers and MSA.

3.2 Psychological attachment to Arabic

Against the sociolinguistic theorizations of its irrelevance to the lives of Arabic speakers today, we need to consider the all-important emotive value attached to CA. Freeman (1996) writes that Arabic script is spoken and read beyond Arab countries. It is also the language of religion for 1.6 billion people, approximately 23.4% of the world’s population (Pew Research Center, 2011). CA, conceivably the highest form of Arabic composition on a universal scale, is learned, read, chanted, copied, translated, quoted and heard on multiple daily occasions throughout Muslims’ and Arabs’ lives. Because of this, it represents part and parcel of the language bath which impregnates life, regardless of educational attainment.  In this connection, an interesting project would be to describe “a day in the life of an Arab”, with specific reference to language to find out how much space/time each variety furnishes, and what psychological coefficient is attached to each. 

In addition to its long history and its knowledge heritage value, the CA/MSA pair is often seen as one of the building blocks of Arab unity (Barakat, 1993; Owens, 2001). If Islam is the defining culture for a large majority of Arabs, Arabic is the language in which this culture is articulated, and which welds together and defines communities living on four time zones. Freeman (1996) asks why individual national strands of Arabic have not each gone their way to become their own language, and ponders that this is unlikely to happen in the foreseeable future because Arabs hold CA is high esteem. In his Cairo fieldwork, Haeri (2010) asks Egyptians about the value of translating the Qur’an into Egyptian Arabic, and reports their utter surprise at the question as for them, the form and meaning of Quranic language represents an untranslatable unit. They explain that the “language of the Qur’an (…), is after all the word of God and one must read His word and not some translation of it”. They add they are Arabs, that they already speak Arabic and do not have a need for translation (p. 75). 

Another interesting index of attachment to CA/MSA is that Arabs often rush to declare their dialect closer to MSA and more in keeping with its rules. While this should not be an understood as a statement on the purity of language, since any language naturally borrows from and injects into others, this perception of proximity indexes the high value they attach to the Standard, and that the Standard carries a high referential coefficient, if not in their day-to-day conversational and written conducts, then at least in their internal representations of reality. 

There is a final argument for why MSA as a “superposed” variety does not engender negative emotive feelings. Inglehart & Woodward (1972) write that tension between speakers of different languages or dialects occurs when “a dominant language group obtains the social, political and economic power within the society and blocks the social mobility of the minority language groups” (in Borjian 2005, p. 65). This situation does not apply to MSA, since, technically speaking, it has no native speakers; maintaining it as a medium of instruction and formal address neither marginalizes nor privileges any social class or group. Quite the contrary, MSA serves a “neutral language”, a term I borrow from Kachru (1986) for whom privileging of a specific Indian language as official in India’s multilingual society can be objected to by speakers of other languages. Because of this, Kachru recommends that English be one of India’s official languages, since it is no one’s native language, and can thus serve as a neutral equalizer. Quite apart from the truth or falsity of this statement in the Indian context, we can argue that MSA is an equalizer in the Arab context, since Arabs, regardless of class, tribe, region or other variables, have similar levels of access to MSA outside the school context Arabic. The literacy they gain in MSA happens largely in formal contexts, which bring MSA and the dialects closer to each other.

3.3 The literacy factor as another index of fusion

This section concerns itself with two types of literacy, traditional and digital, and how they both militate for convergence between Arabic dialects and MSA. Defining literacy is of course a contentious issue that we will not address here, but we can report the following figures to denote the drastic change in the scene since diglossia became the byword among Arabic sociolinguistic circles: in 1950, adult literacy rates were 12% in North Africa (Easterlin, 2000, p. 20). From 1950 to the period between 2000 and 2004, this number jumped up to 62.7% for the whole Arab region (Burnett, 2005). Youth literacy rates (15-24), jumped from 42.7% in 1970 to 66.6% in 1990.  More recent figures put the combined male-female youth literacy rate on a pan-Arab scale at 94.5% (Sika, 2007, p. 30).  Owens (2001) writes:

Since World War II education has expanded enormously in Arabic countries.  Because the Arabic used in instruction, is, in theory, Standard Arabic, this variety has become accessible to a larger segment of the population in a way it has never been before in the history of the language. Its use in education is reinforced by its use in many public spheres… (p. 430). 

Lubliner (2002) proposes an interesting model to explain the historic importance of such increases in literacy rates. He projects that “diglossia (…) remains stable in a society as long as most of its children undergo only minimal schooling”. However, when increasing segments of young people attend school well into their adolescence, a parastandard develops based on the school language, which becomes the medium of peer conversation, and tends, within a few generations, to replace the dialect.

Though the Arabic linguistic setting is not one in which the dialects have been displaced, it seems there is a movement in this direction, albeit slow and hard to perceive. Salameh (2011), who favors adoption of regional dialects as standards, recognizes that “the numbers of users of MSA [are] swelling and hovering in the vicinity of 50 percent” (p. 56). Compared to UNESCO’s figure of 60% (Burnett, 2005), Salameh’s figure appears to be conservative. However, even at this rate, we may begin to understand the changes occurring in the patterns of educated native Arabic speech. Owens (2001) puts it as follows:

Observation of the spoken language quickly revealed that in practice native speakers of Arabic who had access to both the standard language and the dialect in any given stretch of speech rarely used purely one the other variant (p. 425). 

One extrapolation of these figures is that the degree of interaction between the dialects and MSA is much higher than when mastery of MSA was confined to the 10-15% who had any claim on it in the 1950s and 1960s. 

Chouairi (2009) provides further evidence for this crossbreed when she writes that personal letters often start with expressions and idioms from the H variety though they tend to switch back to L after a few lines (p. 37). This very switch would not have been possible on any significant scale in the decades preceding the spread of Arabic literacy. Daniëls (2004) examines the interplay between Fuṣḫa and the Egyptian dialect and concludes that:

Many local non-fushā characteristics have made their way to formal speech levels. In news broadcasts, for instance, the alveolar fricative /ğ/ (fushā) is systematically realized as a velar plosive /g/ (Cairene/Egyptian), so that /g/ has become part of fushā in Egypt (with the exception of recitations of the Koran (p. 82). 

In this case, it is the Standard that is playing the role of matrix language, wherein elements of the dialect get grafted. Owens (2001) points to the opposite trend where elements of the Standard get grafted onto a dialect base: 

“…the degree of influence of SA [Standard Arabic] on spoken Arabic in modern Arabic countries can hardly be understated….The linguistic mechanism by which this is accomplished is via the introduction of SA lexical structures into the NA base (p. 450). 

As stated earlier, this changing linguistic landscape is an invitation for further research, especially when conducted in the context of the paradigmatic shifts brought about the information and communications technologies (ICT).

3.4 ICT as another tributary of fusion

A detailed sociology of ICTs and their impact on Arabic is outside the scope of this paper, however, a sketch of the ways ICT has brought Arabic, both formal and colloquial to the radar of the Arabic speaking communities, is important to draw. First, here is a definition of ICT:

ICT consist of the whole range of technologies designed to access, process and transmit information: hardware, software, networks, and media for collection, storage, processing, transmission, and presentation of information in the form of voice, sound, data, text and images (Detschew, 2007, p. 28). 

While traditional media, such as newspapers, landlines, radio and terrestrial television channels, offered limited opportunities for Arabs of different nationalities to meet, test their “common language”, and gauge the coverage of and interaction between their dialects, ICT has brought Arabs together like no other platform. Use of Arabic word processing software programs such as Arab Word, Arabic Word Perfect, Arabic-enabled Microsoft Word, and generally Arabic desktop publishing solutions among schools, administrations, commercial institutions, homes, and individuals, has increased manifold. Arab populations, who had historically conducted their affairs in mostly oral fashions, have thus moved from the short-lived stage of traditional literacy in the sixties and seventies, to digital literacy from the late eighties onwards.

With Satellite television, exemplified by the Rotana Group, established in 1987, Middle East Broadcasting Center (MBC) 1991, Arab Radio and Television (ART), 1993, Children’s Cartoon TV, 1996, Al-Jazeera, 1996, and according to Battah (2011) as many as 250 satellite television channels (in Khater, 2011 p. 363) unbounded, new, forums “established a virtual on-air community, and a sense of Pan-Arab unity” (p. 363). Al-Mayadeen Satellite TV is the latest addition to this growing network.  Added to the benefits of literacy in the Standard, the rise of “on air-communities” brings together, in Arab living rooms, speakers of dialects as widely distant as a Mauritania and Jordan. Analysis of the dynamics of interaction and the adjustments made by one speaker or another will quickly reveal that opinions, issues, and disagreements are not necessarily handled in MSA, but in a makeshift dialect comprehensible to the guests, the host, and the Arab-wide audience to varying degrees.  Presumably, this medium is constantly negotiated, with the caveat of mutual intelligibility and the presence of MSA and the speaker’s dialect/idiolect as modulators of and circumscribers of variation. Thus, the average Arab’s repertoire has been enriched with increased understanding of far-flung dialects and augmented exposure to formal Arabic, not just as reader and writer, but as listener and more often than before as interlocutor in an unfolding pan-Arab discussion theater, almost in synchrony with the unfolding Arab Spring.

In a volume titled ‘The Real Arab World: Is reality TV democratizing the Middle East?’, Armbrust ((2005) recalls speaking to an Egyptian Film director who, in search for an” Arab perspective”, switched off CNN and BBC and opted for Al-Jazeera (Pp. 1-2). Because of her strategic decision to air an Arab perspective, Al-Jazeera no doubt elected to use Fuṣḫa rather than target a narrower viewership through privileging a dialect. Al-Shamrani (2012) conducted a comparative study of seven of Al-Jazeera programs and found the overall use of Standard Arabic by program presenters to be 83%, while that of guests was at 76%, producing an average of 80% use of Standard Arabic (p. 60). Al-Shamrani also reported on the rate of MSA usage in a live call-in program where the callers are children. While the broadcaster’s reported use of Arabic as 96%, children’s handling of Standard Arabic is described as “competent”  despite their young” age. Al-Shamrani attributes children’s ease with MSA to language planners, policy makers and families, who prefer children to master Fuṣḫa (Pp. 62-64). 

Eventually, even satellite televisions stations known for favoring the dialect or a diluted form of the Standard, such as Lebanese Broadcasting Corporation (LBC), are enriching the average Arab’s dialectal repertoire, not so much as a productive user, but at least at the level of reception, to an extent not possible in the past. To this picture of convergence, among the dialects and between the dialects and the Standard, is added another tributary encompassing the internet and the Arab Spring.

3.5 The Internet, Arabic and the Arab Spring

As an event taking shape mostly in Arabic, the Arab Spring is as much lived on the Arab streets as it is channeled on the internet. The internet is arguably not only a channel, but also something of a dynamo, contributing its own momentum. Because of the relative unity of the language, MSA, dialects and intermediate varieties, slogans transfer from one locale to another in real time.  Some slogans have reverberated verbatim; /a-ʃaʕb juri:d ʕisqa: ṭa –niðˤa:m/, ‘The people want regime downfall’ in MSA, was the byword for objectors in Tunisia, Egypt, Libya, Syria, and Yemen. In some cases, syntax is preserved, but lexical units change to suit the situation, as in ‘The people want regime reform’, or ‘The people want the cleansing of the judiciary”. Standard Arabic slogans, such as ‘Leave’ /ʔirḫal/ or /jasquṭ/ ‘Down with…’ cut across social and geographical boundaries. These slogans have become part of popular and shared memory of Arabs regardless of any social variables, in a manner that is possibly only true of the Quran, some of whose Surahs most people know to varying degrees. Now, Arabs have in common, not just their heritage language, CA, not just their sacred Book, the Quran, not just MSA, but also a largely shared and active popular memory of slogans, events, stories and a stronger sense of common destiny.

The Arab Spring and the internet have squeezed geographical space and linguistic distance. All of a sudden, Arabs, previously confined to living within the borders of their ‘modern’ nation states and to crossing border posts with extreme difficulty, have found themselves communicating their experiences, stories, struggles, and dialects to other Arab countries with the speed of light.  They read, listen, share, disseminate and comment upon materials in Standard Arabic and their various dialects. Hofheinz (2005) comments that the internet whose use was limited to middle-aged professionals in the 1990s has rapidly become a factor in the socialization of the Arab Spring generations (p. 83). In 2003, internet penetration in Arab homes was 4%. An estimate of the number of internet-connected homes in 2006 was 11% (Hofheinz, 2005, p. 82).  One year before the Arab Spring the rate stood at 17.5% (Internet World Stats). In 2011, while the revolution was underway, International Telecommunications Union reports the figure at 29%. This increased penetration goes hand in hand the number of household-equipped computers, which, in 2012 stood at around 31%. This translated itself in a substantial increase in Arabic materials on the internet.  Within years, from 2004 to 2012, Arabic materials on the internet increased around twofold, from 1.7% (Abdulla, 2007, p. 146) to 3% (Internet World Stats). Salama, Director of the Cairo Microsoft Innovation Center comments, “The amount of Arabic I use on the Internet has tripled since the revolution… On Facebook, for example, we communicate much more in Arabic now than we did previously” (cited in Gantenbein, 2011).

The claim that Arabic dialects including MSA have come closer to each other can be substantiated through examination of Facebook pages, Twitter posts, YouTube videos, instant messaging services, synchronous and asynchronous discussion forums, weblogs, etc. In the interest of space, I sample internet Arabic in two ways in the remainder of this section. First, I preview a series of reader comments in MSA on an article carried by Al-Jazeera.  I then analyze a popular Yemeni revolution song where MSA and the dialect coalesce in new and revealing ways.

3.5.1 Analyzing reader comments

I refer to a news briefing produced by Associated Press and published on Al-Jazeera.net, February 8, 2012, around 3 months after the then President Moncef Marzougi of Tunisia took office and just over a year after the start of the Arab Spring in Tunisia.  The excerpt, 333 words long, is titled: “Marzouqi in Morocco at the Start of a regional tour”. Two days after its publication, 148 readers commented on the article, from 8 different Arab countries, mainly North African but also from Yemen, Syria and Palestine.  Al-Jazeera’s policy is that comments be “written obligatorily in Arabic”, without specifying level of formality. Clearly, however, all commentators responded only in MSA, and wrote a total of 8,335 words, and an average of 58 words per comment. Why commentators responded only in MSA can be attributed to Al-Jazeera user comment policy, but commentators also know they are addressing an Arab-wide readership, in writing mode, which can be interpreted as an invitation to avoid colloquialisms. What this says is that in Al-Jazeera’s theater of open and free transaction, commentators use MSA, which they handle correctly.  The Arab Spring is transacted in Arab city squares via oral slogans, banners, and other artistic expressions, and also in written, audio, video and other internet materials. Again, the internet and the Arab Spring are helping move Arab citizens from traditional literacy forms to new forms of cyber literacy.     

3.5.2 Sampling the Arab Spring on the street

Let us now sample the language of the Arab Spring as lived on the street, and as reported by social media, such as Facebook, YouTube, forums, blogs and other Web 2.0 tools. To this end, I introduce a popular song retrieved from Amr Khaled’s Forum (2011) by Yemeni singer Mohamed Al-Adhru’i, in San’aa’s Taghyeer (Transformation) Square. In this video, the singer parodies the efforts deployed by a representative of the then President Ali Abdullah to pacify Yemen. To the representative’s overtures, a large audience responds with a refrain indicating determination to effect regime change:

Last stanza of the song in the dialect

rendition of the last stanza in MSA (my translation)

Transcription of song in the dialect

Transcription of song in MSA

Gloss (my translation)

إذن أنا عرفت قصدكم

الكذب زاد

وطال صبركم

وهم ببنجخوا

من حقكم






وانتم مساكين

ايش بيدكم.

إذن أنا عرفت قصدكم

الكذب زاد

وطال صبركم

وهم يتمعشون

من مالكم






وانتو مساكين

أي شيء بيدكم

iðan ʔana: ʕarifti ɡaṣdakum

ʔal-kaðbi za:d

w-a:l ṣabra-kum

wa hum bi-banɡixu

min ḫagga-kum





w-antu: masa:ki:n

e:ʃ  bi-yadd:a-kum




iðan ʔana ʕariftu ɡaṣdakum

ʔal-kaðibu za:da

Wa (qad) a:l ṣabru-kum

wa hum yatmaʕʕaʃu:na  min ma:li-kum




wa-ʔantum masa:ki:nun ʔayyu ʃayʔin bi-yadi-kum


So, I now know what you mean

Lying is on the increase

and  your patience wearing thin

They are living off your possessions


While your hands are tied

Refrain in MSA

الشعب يريد إسقاط النظام


إذا الشعب يوما أراد الحياة،

فلا بد أن يستجيب القدر



بلادي بلادي بلادي اليمن

أحييك يا موطني مدى الزمن








Refrain  MSA

(Same as in column 1)





The people want the regime to fall.


If, one day, the people will life, destiny will have to comply.


My homeland, my homeland, My homeland Yemen, I salute you my nation for ever.





Table 1:  Sampling the Arab Spring: a popular Yemeni revolution song

I limit my analysis of this excerpt to points corroborating the thesis that the Arab Spring is bringing the dialect and the Standard, at least in this song, closer to each other. Looking at the dialect stanza in relation to its equivalent in MSA, we note the following similarities and differences:


Yemeni Dialect Feature

MSA features

Additional comments

Number of word tokens



The one word difference is due to phonological coalescence ʔayyu-shayʔinè?e:sh

Vocabulary differences



/tamʕaʃa/; /ma:l/


/hag/ is an MSA word meaning ‘right’, ‘what is your lawfully yours’

Other spelling differences



ʔi: ʃ



ʔayyu ʃaj-in

Deletion of Standard Arab plural suffix/m/ in the dialect

Structural differences





Interrogative sentence

Verb and noun cases


Usually articulated

Formal spoken Arabic tends to drop certain case markings. 

Other phonological differences







Table 2:  Summary of the differences between the Yemeni dialect and MSA

Discounting vowel diacritics, which are usually not added to Arabic text, the dialect and the Standard orthographically differ from each other in minor ways. Lexical differences are minimal, with 14 out of the 16 words in the dialect preserving the same word tokens in the Standard. Use of dialectal phoneme /g/ instead of MSA phoneme /q/, coupled case dropping may not impede understanding; the /g/-/q/ distinction is not made in Modern Standard Yemeni Arabic. There are no differences in sentence structure. I am of course not attributing these minor differences to the Arab Spring, but what this passage does is to invite a reconsideration of the long-held belief that the dialect and the Standard are significantly different from each other, to start with.

Other interesting phenomena in this passage beg a comment.  Firstly, the dialect and the Standard are lumped together, in one unified, cohesive text. This goes against the traditional definition of diglossia, where the dialect and the Standard specialize as a function context. Here, in the same setting, are two varieties of Arabic appearing as one unified artistic construction.  Further, the singer, playing the spokesperson for the President, uses the dialect to convey governmental plans for a more democratic Yemen. Government business is conventionally couched in the Standard, but here, the linguistic tables are turned: it is the people who chant anti-regime slogans in Modern Standard Arabic. Thus, the Standard is appropriated by the people and no longer the privilege of the ruling or educated class who traditionally use it to index status, as opposed to the ‘vernacular’, usually associated with people. Nor is the use of Standard Arabic on the part of the people made in a formal context. In this downtown square, it is the people who rule, while the leaders are repudiated.  In this exercise of people power, people take possession of the Standard dialect that was previously used to subjugate them. 

Finally, it important to note that the refrain itself is extended and contains three Standard Arabic slogans, which are first chanted consecutively and then simultaneously as an ensemble capable of overpowering authority.  The first slogan is from the national Yemeni anthem, indicating a state of revolution. Under normal circumstances, anthems are often reserved for government rituals.  In the Arab Spring, however, the anthem belongs to the people. The second part of the refrain is the one slogan that has been branded by Arab people, across the vast swathe of the Arab world, ‘The people want the regime to collapse’. Adding this slogan to the song is an indication that sources and potential audiences of the words and meanings are from both within Yemen and across the Arab world. The final part of the refrain is a quote, again, in MSA, with a very special place in popular Arab memory, from Tunisian poet, Abu Al-Qassim Al-Shabbi, whose poem is titled ‘The Songs of Life’. These are famous lines from the Tunisian national anthem, now chanted in the Arabian Peninsula.

  1.  Conclusions
    1. Summary

The purpose of this paper has been to show that since the rise of interest in Arabic diglossia in the late 1950s until the current events of the Arab Spring, Arabic variation has been on a contraction and convergence path, rather than one of fragmentation and shift. The main problem seems to be that the diglossic prism, descriptively productive as it may be, has not taken account of three transformations that have galvanized the Arabic scene which are bringing Arabs and Arabic varieties closer together, namely, rising literacy rates in MSA, the ICT revolution, including the internet, and the latest events of the Arab Spring still unfolding.  

4.2 Provisional conclusions

The arguments of this paper suggest the following conclusions: (i) the initial assumption of relative stability in Ferguson’s definition of diglossia may no longer be applicable, given the  nature and speed of change in the Arab world, (ii) the presumed division of labor between formal Arabic and popular dialects depends more on function than context; (iii) Formal Arabic and the dialects may not be as structurally divergent as the diglossia frame indicates; (iv) in today’s more advanced literacy settings, formal and colloquial Arabic often co-exist, borrow from each other, and are not compartmentalized, and (v) the Arab Spring and the dialects in which it is transacted, are breaking the boundaries between national dialects and their communities; new inter-Arab and cross-dialectal forums are developing, where dialects are adjusting toward each other and toward the Standard.

  1. Areas for further investigation

Being the medium of such a disruptive event as the Arab Spring, Arabic/inter-Arab communication should offer an exciting and fertile area to describe, monitor and theorize. Five questions in particular appear to me to be worthy of further investigation: (i) Is it possible that diglossia, as a western instrument of description, is more attentive to the many details of difference between Standard Arabic and the dialects than to the perceptions of similarity and coalescence seen from within?  (ii) What are the dynamics and forms of inter-Arab virtual communication? (iii) Given four audio/video excerpts in Arabic dialects in the four regional Arabic dialects, and four Arabic speakers from each of these regions, what would be the rate of inter-dialectal intelligibility? (iv) Considering a full day in the life of an Arab, what is the depth and breadth of their encounters with formal Arabic, Arabic dialects, intermediate varieties and other languages? (v)  What would be the shape of a research tool to gauge Arabs’ attitudes toward different varieties on the Arabic dialectal continuum?

Reference lists

  1. Abdelali, A.  (2004, Jan 1). Localization in Modern Standard Arabic. Journal of American    Society of Information Science and Technology, 23-28. 
  2. Abdulla, R.A.  (2007). The Internet in the Arab World:  Egypt and Beyond. Digital Formations,         43, New York: Peter Lang Publishing. 
  3. Albirini, A. (2011, Nov.).  The Sociolinguistic Functions of Codeswitching between Standard         Arabic and Dialectal Arabic.  Language and Society, 40, 537-560.
  4. Aljazeera.net. (2012, Feb. 12). Al-Marzougi bi-maghrib bi-mustahalli jawlah iqlimiyah (Source:         Associated Press). Retrieved from: http://aljazeera.net/news/pages/bb06654b-b68e-416d- 9aee-7056fb5e3b70.
  5. Al-Shamrani, H.  2012.  Diglossia in Arabic TV Stations.  Journal of King     SaudUniversity–Languages and Translation.  24, 57-69. 
  6. Amr Khaled Forum. (2011). Mohamed Al-Adhru’i: bilaadi sabarti sana. Retrieved from http://forum.amrkhaled.net/showthread.php?392754-%D4%D1-%C7%E1%C8%E1%ED%C9-%E3%C7-%ED%D6%CD%DF-%E4%DF%CA-%CF%E6%CA-%E4%DF%CF/page5.
  7. Altoma, S.J. (1969). The Problem of Diglossia in Arabic: A Comparative Study of Classical      Arabic and Iraqi Arabic.  Cambridge: Harvard University Press.
  8. Amer, F.H., Adaileh, B.A. and Rakhieh, B.A.  (2011). Arabic Diglossia: A Phonological Study.      Argumentum7, 19-36.
  9. Armbrust, W. (2005). Aljazeera is Not a Medium. TBS, 1(2), 1-4.
  10. Ayari, S. (1996). Diglossia and illiteracy in the Arab World. Language, Culture and                      Curriculum 9, 243-252.
  11. Barakat, H. (1993).  The Arab world: Society, Culture, and State. Berkeley and Los Angeles: University of California Press.
  12. Borjian, M. (2005, spring). Bilingualism in Mazandaran: Peaceful Coexistence with Persian. In Languages, Communities, and Education. Teachers’ College, Columbia, Society for     International Education.
  13. Burnett, N. (Director). (2005). Education for All: Literacy for life. UNESCO Publishing. Paris,          France.
  14. Chouairi, R. (2009, summer). Arabic to Where? Some Problems that Ail the Arabic Teaching Industry.  The NECTFL Review64, 1-14.
  15. Dakwar, R.K. (2005, spring). Children's Attitudes towards the Diglossic Situation in Arabic and its Impact on Learning. Languages, Communities, and Education.  Teachers’ College, Columbia, Society for International Education.
  16. Daniëls. H. (2004).  Linguistic Conservatism as the Basis for Political Revolution? The Fuṣhā- ‘Āmmīyah Debate in Nineteenth-Century Ottoman-Arab Middle Eastern Society, 79-92.                    In Brisard, F., D’Hondt, S and Mortelmans, T. (Eds.) Language and Revolution –                  Language and Time.  Antwerp Papers in Linguistics, 106. 
  17. Detschew S. (2007).  Impact of ICT in the Developing Countries on the Economic Growth.      Implications Derived from Theory and Empiricism.  A diploma thesis.  Norderstedt,   Germany: GRIN Verlag.
  18. Easterlin, R.A. (2000). The Worldwide Standard of Living since 1800. Journal of Economic   Perspectives, 14(1), 7-26.
  19. Elgibali, A. (1996).  Beginning to understand Arabic, in El-Said M. Badawi, Elgibali, A. (Eds.).  Understanding Arabic: essays in contemporary Arabic linguistics in honor of El-Said Badawi, 1-14. Cairo: American University in Cairo Press.
  20. Ferguson, C.A, 1959. Diglossia. Word 15, 325–340.
  21. Fishman, J. (1967). Bilingualism With and Without Diglossia; Diglossia With and Without      Bilingualism. Journal of Social Issues 23(2), 29–38.
  22. Freeman, A. (1996, Dec. 9).  Andrew Freeman’s Perspectives on Arabic Diglossia. Retrieved    from: http://www.innerbrat.org/Andyf/Articles/Diglossia/digl_96.htm.
  23. Gantenbein, D.  (June 9, 2011) Making Arabic More Accessible.  In Microsoft Research.    Retrieved from: http://research.microsoft.com/en-us/news/features/afkar-060811.aspx.
  24. Gumperz, J.J. (1964). Speech Variation and the Study of Indian Civilization. In Hymes, D. (Ed.). Language in Culture and Society, 416-428. New York: Harper & Row.
  25. Haeri, N. (2000, Oct.). Form and Ideology:  Arabic Sociolinguistics and Beyond. Annual Review                   of Anthropology, 29(1), 61-87.
  26. Hofheinz, A. (2005). The Internet in the Arab World: Playground for Political Liberalization.      IPG, 3, 78-96.
  27. Ibrahim, R. (2009). The Cognitive Basis of Diglossia in Arabic: Evidence from a Repetition            Priming Study within and between Languages.  Psychology Research and Behavior              Management, 2, 93–105
  28. International Telecommunications Union (ITC).  (2012). ICT Adoption and Prospect in the Arab Region. Retrieved from: http://www.itu.int/dms_pub/itu-d/opb/ind/D-IND-AR-2012-PDF-E.pdf.
  29. Kachru, B.J. (1986). The Alchemy of English: the Spread, Functions and Models of Non-Native           Englishes. Oxford: Pergamon Press.
  30. Khater, A.F. (2011). Sources in the History of the Modern Middle East (2nd Ed.) Boston, USA,               Wadsworth: Cengage Learning. 
  31. Lubliner, J. (2002). Reflections of Diglossia. Retrieved from   http://www.ce.berkeley.edu/~coby/essays/refdigl.htm
  32. Maamouri, M. (1967). The Phonology of Tunisia Arabic. Ph.D. Dissertation. Cornell University,       Ithaca. University Microfilms International, (1985).
  33. -------------------. (1998). Language Education and Human Development: Arabic Diglossia and its Impact on the Quality of Education.  ERIC papers, 1-85. 
  34. McKay, J.M. (1972). Syntactic Structures in Arabic Diglossia. Unpublished Ph.D. Thesis.        University of Texas in Austin. 
  35. Ong, W.J. (1988). Orality and Literacy: The Technologizing of the Word.New Accents. Ed.        Terence Hawkes. New York: Methuen.
  36. Owens, J. (2001). Arabic sociolinguistics. Arabica, 48(4), 419-469. 
  37. Pew Research Center. (2011, Jan. 27). The Future of the Global Muslim Population. Retrieved from: http://www.pewforum.org/The-Future-of-the-Global-Muslim-Population.aspx.
  38. Raddaoui, A.H. (1988). Planning Language: Competing Models of Education in Multilingual         Countries with English as a Second Language. Unpublished Ph.D. thesis, Indiana       University of Bloomington. 
  39. Salameh, F. (2011, fall) Does Anyone Speak Arabic? Middle East Quarterly, 18(4), 47-60.
  40. Shubashy, S. (2004). Taḫya al-lughah al-3arabiyah – Yasqut Sibawayh.  Cairo: Al-haiyah al-        misriyah al-3ammah li-l-kitaab. 
  41. Sika, N. (2007). The Millennium Development Goals: Prospects for Gender Equality in the Arab     World. Journal of International Women’s Studies, 12(3), 27-42. 
  42. UNESCO.  (2006). Education for All: Mapping the Global Literacy Challenge, Chap. 7, Global   Monitoring Report, 161-187. Retrieved \from  http://www.unesco.org/education/GMR2006/full/chapt7_eng.pdf.