Wikipedia:Reference desk/Archives/Language/2022 November 4

= November 4 =

How many Modern Standard Arabic (MSA) speakers are there?
According to Ethnologue, there are 280 million Modern Standard Arabic (MSA) "speakers" (all of them being L2 "speakers"). However, as you need formal education to learn MSA I find this figure too high. (see also: Learning Poverty in the Middle East and North Africa) My back-of-the-envelope calculation done last year led to a number of speakers at least 2x lower.

Do we have better sources on MSA proficiency in the Arab world? MSA being mostly a literary language, how do we define what an MSA speaker is? (reading only? reading and writing? reading and listening? reading, writing, listening, and speaking?) What's the level of proficiency required to qualify as a "speaker"? A455bcd9 (talk) 20:06, 4 November 2022 (UTC)


 * There's an interesting article "Knowing standard Arabic: testing Egyptians' MSA abilities" by Dilworth B. Parkinson in Perspectives on Arabic Linguistics V edited by Mushira Eid and Clive Holes (1993, ISBN 1556195540).  I don't think he gave any estimates of total number of speakers (and if he had, they'd be 30 years out of date). AnonMoos (talk) 22:25, 4 November 2022 (UTC)


 * Many fantasy films (such as Disney films and the Harry Potter films) and children's cartoons are dubbed in standard Arabic. It has been suggested that this is done in order for kids to learn standard Arabic, but I guess another explanation is as a cost-cutting measure. Whatever the reason, a reported effect is that "younger generations understand Standard Arabic rather well, and have a much larger Standard Arabic vocabulary than earlier generations did at the same age". --Lambiam 09:04, 5 November 2022 (UTC)
 * @Lambiam, on cartoons, the World Bank says:
 * "Firstly, before starting school, some children have very little exposure to formal (non-dialect) Arabic; others may hear it through cartoons dubbed in MSA, being read to, or through listening to the Qur’an being recited at home (where applicable). Children’s experience with MSA is very limited before they reach school age because of low rates of parents reading to children at home and a lack of participation in formal early childhood education."
 * They add:
 * "Teachers, including many Arabic language teachers, are themselves the product of ineffective Arabic language education and often are not comfortable using it as a medium of instruction."
 * If teachers aren't proficient in the language, what can we infer about the rest of the population?
 * On children in general, this January 2022 paper says:
 * "formal exposure to MSA takes place only at the onset of formal studies in school (Hudson, 2002). For this reason, coupled with insufficient exposure to MSA at home, MSA is considered a second language. When children begin to learn it in first grade, their shortcomings in phonemic repertoire and phonological awareness affect their phonological processing and, in turn, their acquisition of reading proficiency (Abu-Rabia, 2000). Saiegh-Haddad (2003, 2004, 2007) shows clearly that Arab children in preschool and first grade find it immensely difficult to deconstruct MSA structures into phonemes."
 * "The language of instruction in Arab schools today is SAV. Students learn most subjects—science, geography, math, and English, to name only a few—in this language. A schoolchild has no linguistic role model. Even classes held for the purpose of imparting MSA are taught in the vernacular by most if not all teachers. Consequently, pupils learn in a vernacular linguistic environment and are not exposed to the standard language. In this state of affairs, one cannot expect pupils to master MSA and use it to write and express themselves adequately."
 * I found some studies assessing proficiency, such as the one cited by @AnonMoos, but they're either old, or focused on one country only, or on one demographic group (university students, women, etc.). So it's hard to reach a conclusion for the whole Arab world. (I posted these sources here and there.)
 * Is there another way to better assess this number? A455bcd9 (talk) 10:16, 5 November 2022 (UTC)
 * It is very hard, also because, as you yourself already indicate, there is no standardized measure for language proficiency. This makes it practically impossible to combine the findings of different studies, even when adjusting for variables such as the date of the study and demographic characteristics of the group studied. --Lambiam 10:44, 5 November 2022 (UTC)
 * The Common European Framework of Reference for Languages provides a standardized measure. And we could say that "proficient" = at least C1. But until recently, there wasn't even a standardized test for MSA... A455bcd9 (talk) 10:52, 5 November 2022 (UTC)
 * It is only a fraction of FL-language learners which take such tests for evaluation. And even less speakers are evaluated in a language that is not perceived as a FL in the context of diglossia. –Austronesier (talk) 11:10, 5 November 2022 (UTC)
 * For sure. But we could imagine a study where participants would be evaluated against that framework (or any other framework). Unfortunately, as reported by the World Bank: "[in the Arab World] cases of grade inflation are high, and there is generally an absence of well-functioning national assessment systems to track student performance, make results widely available, or monitor a national literacy strategy". They add:
 * "More than one-half (59 percent) of the children in MENA countries are in learning poverty—they cannot read and understand an age-appropriate text by age 10."
 * "Findings from other national and international assessments confirm that students in the region lack basic literacy and numeracy skills from the earliest grades up to secondary school. For example, the Early Grade Reading Assessments (EGRAs) conducted in several MENA countries suggest that 52 percent of grade 2 children in Iraq, 67 percent in Morocco, and 81 percent in Yemen could not answer a single reading comprehension question correctly."
 * "By age 15, performance in reading, mathematics, and science remains well below the average of Organisation for Economic Co-operation and Development (OECD) countries, as seen in the 2018 Programme for International Reading Assessment (PISA)"
 * A455bcd9 (talk) 11:20, 5 November 2022 (UTC)
 * Found this July 2022 paper:
 * "no reading fluency rates (FR) or standards in Arabic have been established to date."
 * "Besides the results from the Early Grade Reading Assessment (EGRA), no established norms for oral reading rates in Arabic were found and very few studies looked at fluency (Clarke et al., 2015; Dowd & Bartlett, 2019; Dowd et al., 2020; Faour, 2012;)."
 * "This is a key point as most private schools in the Arabian Gulf, which form about 60% of all schools, restrict Arabic language teaching and exposure to 45 min a day, while the rest of the time is spent in another language, mostly English."
 * "Participating teachers asked students to read one leveled, grade appropriate, authentic, connected MSA text for 1 min without stopping. Teachers timed the test from the first word read aloud and ended the test after 1 min. During the 1-min reading, teachers marked on their sheet all the words read incorrectly by the student. The teacher then counted the number of miscues and subtracted that from the total number of correct words read to arrive at the wrcpm result."
 * "The first observation in this study is that ORF explored (Table 7) for Arabic language seem to be in line with ranges published for the Arabic MSA version of EGRA for the West Bank (Palestine), 30–35 wrcpm for Grades 2 and 3, but appear to be much lower than ranges published for Egypt and Jordan, that have reported 50 and 46 wrcpm, respectively, for Grades 2 and 3 (RTI International, 2017)."
 * A455bcd9 (talk) 12:06, 5 November 2022 (UTC)
 * The "levels" of the CEFR are defined separately and differently for five categories: Listening, Reading, Spoken Interaction, Spoken Production and Writing. The descriptions are verbal and their application may vary wildly between assessors. How is one supposed to determine whether someone can deal with most situations likely to arise while travelling in an area where Modern Standard Arabic is spoken (Spoken Interaction B1)? This is hardly what I'd consider a standardized measure, and for wide adaption by researchers we need objective, more conveniently applicable operational definitions. --Lambiam 12:40, 5 November 2022 (UTC)
 * My impression was that Modern Spoken Arabic really wasn't spoken anywhere naturally, except perhaps on international media offices as a lingua franca, or something similar... 惑乱 Wakuran (talk) 13:54, 5 November 2022 (UTC)
 * The point is that the CEFR description applied to MSA requires one to assess someone's competence in dealing with travellers' issues in an area where Modern Spoken Arabic is spoken. This requirement is kinda problematic if no such area exists. --Lambiam 18:07, 5 November 2022 (UTC)
 * Yes, "pure" MSA isn't "spoken" naturally. That's also an issue in defining the number of MSA "speakers": do we only consider people able to read and write it? (instead of assessing speaking and listening skills) A455bcd9 (talk) 14:31, 5 November 2022 (UTC)

It's interesting that children's cartoons with a single pan-Arab soundtrack are in MSA now; during much of the 20th century, it would have been "Egyptian movie Arabic". Anyway, if "speaking MSA" means getting the i'rab vowels right in spontaneous conversation, then there would be relatively few MSA speakers by that criterion, while if it means being able to basically functionally communicate with someone who natively speaks an Arabic colloquial which is very different from your own colloquial, then the number would be much larger... AnonMoos (talk) 23:25, 5 November 2022 (UTC)


 * Not all cartoons are in MSA: Disney dubbed its cartoons in Egyptian until 2012 then switched to MSA. The switch was met with widespread backlash. So Disney switched back to Egyptian this year: all cartoons on Disney+ will be available in Egyptian and MSA, and all the movies produced between 2012 and 2022 will be re-dubbed in Egyptian. ( & ) A455bcd9 (talk) 08:46, 6 November 2022 (UTC)
 * Yes, because there are no native MSA speakers it's hard to assess what "speaking MSA" means. So it's mainly prescriptive. "functionally communicate with someone who natively speaks an Arabic colloquial which is very different from your own colloquial" => wouldn't they most likely use English or French? I know my Moroccan friends in London or Paris do this when they speak with Saudi or Iraqi for instance. A455bcd9 (talk) 08:49, 6 November 2022 (UTC)
 * 2017 survey of 91 first-year students at a UAE university: all native Arabic speakers and all UAE nationals:
 * 16%: Unable to write in Arabic ("unable to write a single word in Arabic")
 * 43%: Poor Arabic writing skills ("their Arabic writing was very poor compared to their English writing. The Arabic writing of this group suffered from several problems. Most notable are: spelling mistakes, grammatical errors, wrong word choices, coherence, cohesion, use of code switching (Arabic/English), literal translation from English to Arabic, and the insertion of colloquial Arabic." [...] "rudimentary ability to write in Arabic, and produced extensive errors in their paragraphs.")
 * 41%: Good Arabic writing skills ("no serious errors")
 * If even university students in the second richest Arab country in the world (after Qatar, by GDP per capita PPP) cannot write in MSA, I assume there are not a lot of MSA speakers in the Arab world. A455bcd9 (talk) 12:40, 6 November 2022 (UTC)