An Orthographic Phonological-Based Error Analysis of the Arabic of English-speaking Learners
Dr. Muntasir Al-Hamad, Arabic for Non-Native Speakers Center, Qatar University
Dr. Yehia A. Mohamed, Arabic Program, Georgetown University Qatar
ABSTRACT
The complex interrelation between orthographic and phonological aspects in L2 learning pose different linguistic and pedagogical challenges. However, this paper focuses on the orthographic phonological-based errors resulting of this relationship and tries to explain these errors committed by learners, since Arabic and English have different phonological systems. For this purpose, a study was conducted on a corpus of about 250 pages generated by the written production of sixty-one English-speaking A1 learners of Arabic as a second language (acc. CEFR) at the Manchester Metropolitan University to examine how the phonological competence at the early stages of learning Arabic reflects on their orthographic production. It is hoped that the results of this study will provide answers for linguist researchers and educationists alike.
KEYWORDS: Arabic for non-native speakers, error analysis, orthography, phonology
INTRODUCTION
Since the 1950s, the collection, classification, and analysis of errors in the written and spoken performances of second language learners have played a significant role in language pedagogy. However, in the late 60s and early 70s, the study of errors in non-native language performance, or Error Analysis (EA), came to the forefront in applied linguistics, and in particular within the field of second language acquisition research (Richards, 1980, p.91).
According to Corder (1981, p.10), ‘error analysis is a method used to document the errors that appear in learner language, determine whether those errors are systematic, and (if possible) explain what caused them.’ Such errors can occur in all aspects of a language, including phonology, morphology, syntax and semantics, as well as across the language skills spectrum, so their exploration and analysis assists in the second language acquisition process by developing students’ second language proficiency and at the same time informing pedagogy and materials’ development.
Interlingual transfer is a significant source for language learners. The Dictionary of Language Teaching and Applied Linguistics argues that interlingual errors are the result of language transfer, which is caused by the learner’s first language via various psycholinguistic phenomena, including: fossilization, overgeneralization, oversimplification, underuse, and lack of knowledge of the rules (Jabeen et al, 2015, p.55). Interlingual errors may occur at different levels involving transfer of phonological, morphological, grammatical, and lexical-semantic and other elements of the native language into the target language (Selinker, 1972 and Vacide, 2005, p.265). However, studies of second language acquisition have tended to imply that Interlingual transfer may be most predictive at the level of phonology (Richards, 1971, p.2).
With the beginning of the 1960s, a large body of research has been conducted on Arabic learners’ errors. Khoury (1961) focused on the writing errors of American students at the high school level. Hanna (1964) conducted a study on Arabic learners’ errors at the college level and concluded that these could be attributed to the nature of the Arabic language and interlingual factors. However, the study had a significant weakness, as there were only 12 subjects used for the analysis. Kara (1971) argued that it was specifically the lack of morphological awareness, both of teachers and in textbooks, that was responsible for common errors in writing. Rammuny (1978) conducted a statistical study of errors made by American students when writing Arabic, highlighting four types of errors: i. orthographical and phonological, ii. semantic, iii. structural, and iv. stylistic. Raslan (1984) used contrastive analysis to study phonological and morphological errors made by Malay college students learning Arabic.
While all of these studies have added to the knowledge base in the field, certain methodological shortcomings, such as the limited number of participants and broad scope of the studies, leave some doubt about the generalisability of their findings. To overcome these issues, the present study involves a large pool of participants. It also adopts a narrow scope aiming to explore and analyze the phonological errors in writing made by 60 English speaking learners of Arabic, studying at Manchester Metropolitan University (MMU) in the UK during the time period 2009-2012.
This paper focuses on the complex relationship between phonology and orthography, with a focus on the errors resulting of this relationship. Written Arabic is an alphabetic system based on 28 letters, with 25 consonants and three long vowels. A further three short vowels and other phonological features, such as gemination are not represented as letters, but as diacritical marks (Abu-Rabia and Taha, 2006, p.321). Arabic uses diacritical marks, such as short vowels and gemination, mainly for educational and religious purposes. The Arabic language thus uses an alphabetic orthography, more technically, what is sometimes termed as an ‘impure abjad.’ This is not a true alphabet, as some of the vowels are written with the letters, while others are written optionally as diacritic marks above or below the letters (Wiley and Rapp, 2019, p.975). Just as languages differ from one another, orthographic systems represent phonology, or aspects of phonology, differently (Frost, 1989, p.162).
The Arabic phonemic inventory consists of 28 phonemes, a quarter of which do not have any plausible approximate correspondent in English, such as the emphatic consonants /tˤ/, /dˤ/, /sˤ/, and /ðˤ/; the pharyngeal consonants /ħ/, and /ʕ/; and the glottal stop /ʔ/. In respect to vowels, Arabic has three short vowels, and three corresponding long vowels. Gemination of consonants is another feature of phonological segments. However, it is important to note that although certain Arabic phonemes such as /r/ and /l/ do not have exact equivalent phonetic realizations in English, this does not appear to create a significant problem in the written form. For example, learners mostly do not get confused in writing the letter <ر> regardless if they recognize it as a flap or a trill sound.
METHODOLOGY
The present study focuses on specific phonological errors in the Arabic writing of English native-speaker learners of Arabic at a university level. This research was conducted by using a mixed methods approach which involved the creation of student-generated written and audio corpuses in addition to long-term class observation notes.
Participants
The researchers recruited 63 learners, aged 18-23, who were studying different majors but had selected Arabic as an elective module under University-wide Languages at the Department of Languages at MMU. In 2016, another paper analyzed the syntactical, morphological and lexical language transfer of the same group’s work (Al-Hamad and Alaoui, 2016).
None of this group had studied Arabic before. They were approximately equally divided in terms of gender (35 females and 28 males), and background (29 of European background, 32 British of Asian background, and two heritage learners). The research eliminated three participants, because two participants were heritage students and one British Asian spoke Urdu as her first language, which could have risked her acquaintance with the Arabic script.
Data collection and procedure
The participants were selected out of a larger population of students who underwent a personal interview to assess and determine their level. They were enrolled on the University-wide Languages - Arabic Beginner’s Level classes, which is equivalent to A1 level according to the Common European Framework of Reference for Languages (CEFR 2018). The students had had 40 contact hours of learning Arabic taught for two hours per week over two terms (September - April), in addition to similar amount of time with assistants in the Language Centre located in the department. They were taught using the Mastering Arabic 1 course book in addition to the teacher’s handouts and resources.
Prior to the corpus stage, the researchers used planned observations of classroom activities and drills to generate their hypothesis regarding orthographic phonological-based errors. Hopkins (1996) describes classroom observation as a ‘pivotal activity’, one which plays a crucial role in classroom research. In respect of our study, used together with the generated corpus, it allowed us to view the topic from multiple perspectives.
The participants were informed about the aim of the study and were assured that their identity would remain confidential. They were also informed that the data collected from them would only be used for research purposes. Furthermore, they were given the option of dropping out at any stage.
RESULTS AND DISCUSSION
Before discussing the different types of errors committed by participants of this study, it is necessary to briefly review the phonemes of English in comparison with those of Modern Standard Arabic (MSA). Tables one and two contrast the phonemes of Arabic and English.
| Labial | Dental | Alveolar | Post-alveolar | Palatal | Velar | Glottal | |
| Nasal | m | n | ŋ | ||||
| Stop | p b | t d | tʃ dʒ | k ɡ | |||
| affricate | f v | ||||||
| Fricative | θ ð | s z | ʃ ʒ | h | |||
| Approximant | l | r | j | w |
Table 1: English phoneme inventory based on Underhill (2005)
| Labial | Dental | Dental-alveolar | Palatal | Velar | Uvular | Pharyngeal | Glottal | |
| Nasal | m | n | ||||||
| Stop
Emphatic |
b | t d
tˤ dˤ |
k (ɡ) | q | ʔ | |||
| Fricative
Emphatic |
f | θ ð
s z
sˤ ðˤ |
ʃ ʒ
|
χ ɣ | ħ ʕ | h | ||
| Approximant | l | r | j | w |
Table 2: Arabic phonemes inventory based on Holes, 2004 and Watson (2002)
The error types which are categorised in the remainder of this paper result from differences in their respective phonological systems.
Glottal consonants
Arabic has two glottal consonants, /h/ and /ʔ/, represented in Arabic orthography by <ه> and <ء> respectively. It is important to note that <ء> is not a fixed letter in Arabic, but rather a diacritic symbol, a phenome, that has complex and specific rules and realizations governing when and where it is placed. This includes zero Ø realization, this is particularly true at the beginning of words, and more significantly, when Arabic words are borrowed into the English language, and are used in an Arabic-based script, as the hamza is typically dropped. This makes it unclear if such errors are phonological or orthographic.
Both sounds /h/ and /ʔ/ occur in English, although the latter occurs not as an individual phoneme but rather as an allophone of /t/ in specific environments. The sound /h/ is described as a voiceless glottal fricative. It occurs in English as a phoneme, although it is more limited than the Arabic /h/ in phonotactics, since /h/ can appear in the syllable head or coda in Arabic while in English it can only appear at the syllable head. Many English dialects exhibit a phenomenon called h-dropping in which /h/ is dropped in some, or all, phonetic environments.
| Number of errors | Number of tests | Percentage of tests | |
| /ħ/ instead of /h/ | 16 | 12 | 20% |
| /Ø/ instead of /h/ | 3 | 3 | 5% |
| /x/ instead of /h/ | 2 | 1 | 1.6% |
| /ʔ/ instead of /h/ | 1 | 1 | 1.6% |
| /ʕ/ instead of /h/ | 1 | 1 | 1.6% |
Table 3: Glottal consonants errors I
The most common sound substituted for /ħ/ is /h/. The errors in which /h/ was deleted altogether can be explained as an application of h-dropping to Arabic.
| Number of errors | Number of tests | Percentage of tests | |
| /h/ instead of /ħ/ | 5 | 5 | 8.3% |
| /h/ instead of /ʕ/ | 2 | 2 | 3.3% |
| /h/ instead of /a/ | 1 | 1 | 1.6% |
| /ħh/ instead of /ħ/ | 1 | 1 | 1.6% |
Table 4: Glottal consonants errors II
The sound /h/ was most commonly used instead of /ħ/ as a closer sound for English speakers. The case of the two errors made by one of the students, where they employed /h/ instead of /ʕ/ was somewhat unexpected, since /h/ and /ʕ/ differ in both place of articulation and voicing. That being said, it is possible for English speakers to confuse /ʕ/ with /ħ/ as it is the unvoiced counterpart of /ʕ/.
Pharyngeal consonants
Arabic has two pharyngeal consonants, a voiced pharyngeal fricative and its unvoiced counterpart, transcribed in the IPA as /ʕ/ and /ħ/ and represented in the Arabic alphabet by <ع> and <ح> respectively. As English lacks pharyngeal phonemes, it is expected for native English speakers to have difficulty distinguishing /ʕ/ and /ħ/ from non-pharyngeal sounds, in both perception and production. Previous studies have indicated that English-speaking learners of Arabic tend to produce non-pharyngeal sounds (such as /a/ or /h/) instead of pharyngeal ones in speech, though very rarely the converse. We expect this to be reflected in learners’ writing, but with a significant rate of hypercorrection that would not be found in speech; i.e. learners could be expected to use <ع> and <ح> in place of non-pharyngeal consonants as they automatically assign these letters different phonetic values than they actually have in Arabic.
The sound /ʕ/ is usually described as a voiced pharyngeal fricative, although some sources classify it as an approximant instead. It is found in most modern varieties of Arabic, in some cases it has merged with the glottal stop, and in others the glottal stop is in free variation (Heselwood, 2007, p.1). There is no corresponding phoneme in English, and while in Arabic it is represented by the letter <ع>, in many scholarly transliterations of Arabic it is represented by a right-facing apostrophe. However, in looser ‘popular’ transliterations in English, it is commonly represented by a plain apostrophe, or by nothing at all. Therefore, it is often confused with vowels and glide sounds, as they share a similar phonetic nature. There were a total of 59 errors distributed across 47 tests where learners used a non-pharyngeal letter instead of <ع>. The following table demonstrates each particular form of error with the number of instances as well as the number and percentage of tests that it appeared in.
| Number of instances | Number of Tests | Percentage of Tests | |
| zero Ø or /a/ instead of /ʕ/ | 18 | 13 | 23.6% |
| /j/ instead of /ʕ/ | 15 | 12 | 20% |
| /ʔ/ instead of /ʕ/ | 10 | 8 | 13.3% |
| /ħ/ instead of /ʕ/ | 7 | 5 | 8.3% |
| /w/ instead of /ʕ/ | 3 | 3 | 5% |
| /h/ instead of /ʕ/ | 2 | 2 | 3.3% |
| /ɣ/ instead of /ʕ/ | 2 | 2 | 3.3% |
| /i/ instead of /ʕ/ | 1 | 1 | 1.6% |
| /n/ instead of /ʕ/ | 1 | 1 | 1.6% |
Table 5: The pharyngeal /ʕ/ errors I
Due to the fact that the data is based on written samples, these variables do not indicate whether learners actually perceived or produced these sounds instead of /ʕ/, but their writing suggests they might have. For example, one learner wrote <اشر > instead of < عشر >, suggesting that he/she perceives zero Ø instead of /ʕ/. As expected there are a number of vowel and glide sounds (/a/, /j/, and /w/) that are used in place of /ʕ/. This is due to their sounds’ similar nature, and because the sound is frequently transliterated in English either with an apostrophe, or with nothing at all. In addition to these two reasons, this sound in MSA is not a stop or affricate, i.e. the vocal tract is not closed off at any point during its articulation with the necessity of tongue retraction; this explains why /a/ is the most common sound used in place of /ʕ/.
The second most common error is using /j/ in place of /ʕ/. This happens only in environments where the /ʕ/ is perceived as breaking up a hiatus between /i/ and a vowel of a different quality, for example جامية instead of جامعة. English typically uses /j/ to break up a hiatus of this sort, and thus the Arabic letter <ع> was given the same function amongst the learners. A similar explanation can be given for the three instances as /w/ was used instead of /ʕ/ where this consonant was found between /u/ and some other vowel.
The third most common type of error is using /ʔ/ instead of /ʕ/. This can be explained by the fact that the place of articulation of glottal consonants is relatively close to that of pharyngeal consonants. A further confusing factor may be that both /ʔ/ and /ʕ/ are commonly transliterated in English -if transliterated at all- by a simple apostrophe.
The fourth most common type of error noticed in this group is the use of /ħ/ instead of /ʕ/. This may have two possible explanations, one is phonological and the other is orthographic. Phonologically, the main phonetic difference between /ħ/ and /ʕ/ is that the former is voiceless while the latter is voiced. The orthographic explanation on the other hand is not as clear; this cannot be explained as a lack of motor skills or simply confusing the shapes of the letters <ح> and <ع>, because we attested in four out of five test a clear demonstration of writing the closed form of <ـع> and <ـغـ>. Another assumption is that <ح> and <ع> might appear as mirror images of one another for a beginner user who usually uses the open form of <ع>. Thus, there is a bit of uncertainty as to the validity of these errors having a purely phonological basis.
There was a noticeable number of errors that can be described as hypercorrections, wherein the letter <ع> was used instead of other letters. The number of hypercorrections was still lower than the number of errors of other types, such as with <ع> and <غ>, but was still significant. It is important to note that this type of error is very rare amongst learners’ speech, as /ʕ/ tends to be eliminated altogether.
| Number of instances | Number of Tests | Percentage of Tests | |
| /ʕ/ instead of zero Ø or /a/ | 10 | 8 | 13.3% |
| /ʕ/ instead of /ʔ/ | 6 | 4 | 6.6% |
| /ʕ/ instead of /j/ | 1 | 1 | 1.6% |
| /ʕ/ instead of /h/ | 1 | 1 | 1.6% |
| /ʕ/ instead of /i/ | 1 | 1 | 1.6% |
Table 6: The pharyngeal /ʕ/ errors II
As reflected in the previous group of errors, the most commonly confused sounds with /ʕ/ were zero Ø and /a/. In this group there were 10 instances across eight tests (13.3% of the corpus) of a learner using <ع> to represent zero Ø or /a/. We do not believe that these learners are mentally replacing zero Ø or /a/ with /ʕ/ as their writing suggests, but rather that they have automatically assigned <ع> to represent the phonetic value zero Ø or /a/. The same applies to learners who used <ع> in places where /ʔ/ is needed.
The sound /ħ/ is typically described as a voiceless pharyngeal fricative. As with /ʕ/ it may actually be epiglottal rather than pharyngeal, at least in some Arabic dialects (Khattab, Al-Tamimi and Alsiraih, 2018, p.311). There is no corresponding phoneme in English, and scholarly transliterations often use <ḥ>, while more ‘popular’ transliteration schemes use a simple <h>, thus not differentiating between <ح> and <ه>. It is therefore expected that /h/ and /ħ/ would be frequently confused during learners’ tests.
| Number of instances | Number of tests | Percentage of tests | |
| /Ø/ instead of /ħ/ | 6 | 5 | 8.3% |
| /h/ instead of /ħ/ | 5 | 5 | 8.3% |
| /χ/ instead of /ħ/ | 3 | 2 | 3.3% |
| /k/ instead of /ħ/ | 1 | 1 | 1.6% |
| /ħh/ instead of /ħ/ | 1 | 1 | 1.6% |
Table 7: The pharyngeal /ħ/ errors I
As evident in this table, /h/ is used instead of /ħ/ five times in five different tests, accounting for 8.3% of the corpus. What is interesting, however, is that there are six errors across five tests where learners replaced /ħ/ with nothing at all. This is most likely a result of h-dropping which occurs in many English dialects, especially in England itself. Learners assign <ح> to represent the /h/ phoneme, and some of them, especially, it is to be suspected, those whose accents exhibit h-dropping, have applied this rule to their Arabic writing and omitted the letter <ح> altogether. The use of /χ/ instead of /ħ/ may have a phonological explanation—both sounds are voiceless fricatives pronounced near the back of the vocal tract, albeit at two different points of articulation—although it may also be explained orthographically as <خ> and <ح> are only differentiated by the fact that the former has a single dot above the main letter form.
| Number of instances | Number of tests | Percentage of Tests | |
| /ħ/ instead of /h/ | 16 | 12 | 20% |
| /ħ/ instead of /ʕ/ | 7 | 5 | 8.3% |
| /ħ/ instead of /k/ | 1 | 1 | 1.6% |
Table 8: The pharyngeal /ħ/ errors II
As with the errors related to /ʕ/, it is expected that learners would exhibit hyper-corrective errors related to /ħ/, particularly using <ح> in places where the Arabic has /h/. In fact, this is exactly what was found. What is interesting is that the number of hypercorrections is greater than the number of errors where other letters were used instead of <ح>, in contrast with those errors related to /ʕ/ where the hypercorrections were fewer. There may be several reasons for this, both phonological and orthographic. For example, there may be some hesitance in using <ه> for /h/ due to the fact that it is frequently confused with <ة> and may thus be perceived to have vocalic qualities.
Emphatic consonants
Arabic has a series of so-called emphatic consonants. The emphatic or pharyngealized class, including /sˁ dˁ tˁ ðˁ/ < ظ ط ض ص>, stands in phonemic contrast to the plain class /s d t ð/ <س د ت ذ>. Examples of the plain/emphatic contrast include the following minimal pairs: /nasaba/ ‘imputed’ vs. /nasˁaba/ ‘erected’; /tin/ ‘fig’ vs. /tˁin/ ‘clay’; and /darb/ ‘path’ vs. /dˁarb/ ‘hitting’ (Ryan, Maojing, and Hermes 2018, p. 48). The following table demonstrates the number of errors in which a learner replaced an emphatic letter with a non-emphatic letter. In the vast majority of cases emphatic letters were replaced by their non-emphatic equivalents, e.g. <ط> being replaced by <ت>, although there were a handful of examples of other types, e.g.<ط> replaced by <د>. The replacement of <ط> by < د > may be due to the fact that the two letters are both unaspirated, while <ت> is aspirated.
Removing emphasis
| Number of errors | Number of tests | Percentage | |
| /t/ instead of /tˤ/ | 20 | 17 | 28.3% |
| /s/ instead of /sˤ/ | 19 | 14 | 23.3% |
| /d/ instead of /dˤ/ | 6 | 6 | 10% |
| /d/ instead of /tˤ/ | 1 | 1 | 1.6% |
| /s/ instead of /dˤ/ | 1 | 1 | 1.6% |
Table 9: Emphatic consonants errors I
The fact that /tˤ/ and /sˤ/ are the most commonly confused emphatic letters may be simply due to the fact that they appear more frequently in Arabic than /dˤ/ or /ðˤ/. Notice that /ðˤ/, one of the rarest phonemes in Arabic, has no examples of errors at all.
| Number of errors | Number of tests | Percentage of tests | |
| /sˤ/ instead of /s/ | 7 | 6 | 10% |
| /dˤ/ instead of /d/ | 3 | 3 | 5% |
| /sˤ/ instead of /z/ | 1 | 1 | 1.6% |
| /tˤ/ instead of /θ/ | 1 | 1 | 1.6% |
| /ðˤ/ instead of /t/ | 1 | 1 | 1.6% |
Table 10: Emphatic consonants errors II
One immediately notices that there are fewer instances in which emphatic consonants were used in place of non-emphatic ones than the other way around. This may be because of their unfamiliarity due to the fact that all the emphatic consonants are placed, in most Arabic textbooks, in the order of the alphabet after their non-emphatic counterparts. This means that the non-emphatic letters are usually introduced to learners first.
We can also notice some patterns in errors related to the emphatic (i.e. pharyngealized, etc.) consonants.
| Emphaticizing | De-Emphaticizing | |
| Total | 10 errors | 45 errors |
| /s/ ~ /sˤ/ | 7 errors | 19 errors |
| /d/ ~ /dˤ/ | 0 errors | 6 errors |
| /t/ ~ /tˤ/ | 3 errors | 20 errors |
| /ð/ - /ðˤ/ | 0 errors | 0 errors |
Table 11: Emphatic consonants error III
Notice that the tendency is to write non-emphatic letters in place of emphatic ones, rather than the other way around. This is undoubtedly because English lacks pharyngealized consonants altogether, and thus they are largely unfamiliar to English-speaking learners of Arabic.
Post-Velar consonants
Also known as uvular consonants, Arabic has three post-velar consonants: /q/, /χ/, and /ɣ/, represented in the Arabic alphabet by <ق>, <خ> and <غ>. The latter two may be velar in some Arabic dialects (Dickins, 2007, pp. 38-43). For this reason, we have chosen to use the broader term post-velar rather than uvular.
/q/ is a voiceless post-velar or uvular stop, and has no equivalent in English. Since /k/ is the closest English consonant, it is expected that confusion between /q/ and /k/ will be common.
| Number of errors | Number of tests | Percentage of tests | |
| /k/ instead of /q/ | 13 | 11 | 18.3% |
| /ɣ/ instead of /q/ | 2 | 2 | 3.3% |
| /χ/ instead of /q/ | 2 | 1 | 1.6% |
Table 12: Post-velar consonants errors I
As expected, /k/ is the most frequently used sound instead of /q/. The same is reflected in hyper-corrective errors.
| Number of errors | Number of tests | Percentage of tests | |
| /q/ instead of /k/ | 9 | 6 | 10% |
| /q/ instead of /χ/ | 3 | 1 | 1.6% |
Table 13: Post-velar consonants errors II
/χ/ is a voiceless post-velar or uvular fricative, and in most English dialects there is no equivalent. However, the sound is somewhat more familiar to English speakers due to its existence in Scots English, e.g. ‘loch’, and other European languages, e.g. German and Castilian Spanish.
| Number of errors | Number of tests | Percentage of tests | |
| /k/ instead of /χ/ | 9 | 7 | 11.6% |
| /kħ/ instead of /χ/ | 2 | 2 | 3.3% |
| /q/ instead of /χ/ | 3 | 1 | 1.6% |
| /xħ/ instead of /χ/ | 2 | 1 | 1.6% |
| /kh/ instead of /χ/ | 1 | 1 | 1.6% |
Table 14: Post-velar consonants errors III
As expected, the most common sound used instead of /χ/ is /k/. Interestingly there are a few errors that appear to be based on orthography. Several students used letter combinations such as <كح> or <خح> to represent /χ/. This appears to be based on the fact that the most common transliteration for <خ> in English is the digraph <kh>. In such cases it is unclear if there is also an underlying phonological reason for such errors.
| Number of errors | Number of tests | Percentage of tests | |
| /χ/ instead of /k/ | 5 | 5 | 8.3% |
| /χ/ instead of /ħ/ | 3 | 2 | 3.3% |
| /χ/ instead of /h/ | 2 | 2 | 3.3% |
| /χ/ instead of /q/ | 2 | 1 | 1.6% |
| /χ/ instead of /ɣ/ | 1 | 1 | 1.6% |
Table 15: Post-velar consonants errors IV
/ɣ/ is a voiced post-velar or uvular fricative, the voiced equivalent of /χ/. There is virtually no dialect of English in which it has an equivalent, although it may be familiar to those with knowledge of languages, such French or German, where it occurs as the rhotic.
| Number of errors | Number of tests | Percentage of tests | |
| /k/ instead of /ɣ/ | 1 | 1 | 1.6% |
| /χ/ instead of /ɣ/ | 1 | 1 | 1.6% |
| /ɣ/ instead of /q/ | 2 | 2 | 3.3% |
Table 16: Post-velar consonants errors V
The scarcity of clear examples makes it difficult to analyze the errors related to /ɣ/. It is clear, however, that all of the sounds it is confused with are velar or post-velar. It is important to keep into mind that the number of errors where the learner uses /χ/ instead of /ɣ/ or vice versa may be higher than our tables suggest, due to our interpretation of some errors being orthographic in nature rather than phonological.
Vowels length
Errors related to vowel length are much more frequent with the vowel /a/, rather than the vowels /i/ and /u/. Consider the following table:
| Lengthening Vowel | Shortening Vowel | |
| /a/ ~ /aː/ | 75 errors | 105 errors |
| /i/ ~ /iː/ | 9 errors | 31 errors |
| /u/ ~ /uː/ | 14 errors | 10 errors |
Table 17: Vowels errors I
This may be explained by the fact that /a/ is the most common vowel in Arabic. Nonetheless, we believe there is a phonological explanation for this—English has tense and lax vowels that approximate the quality of long and short vowels respectively for /i/ and /u/, but it lacks a tense/lax distinction for /a/ (Duncan, 2016), therefore making it more challenging for a native English speaker to consistently distinguish /a/ from /aː/.
| /aː/ | /iː/ | /uː/ | /a/ | /i/ | /u/ | /ai/ | /Ø/ | |
| /aː/ | X | 3,3 | ||||||
| /iː/ | 1,1 | X | 2,2 | |||||
| /uː/ | 4,4 | X | ||||||
| /a/ | 2,2 | 2,2 | X | |||||
| /i/ | 7,7 | X | ||||||
| /u/ | 2,2 | 1,1 | X | |||||
| /ai/ | 2,2 | x | ||||||
| /Ø/ | 7,6 | 2,2 | x |
Table 18: Vowels errors II
CONCLUSION
This study was conducted with the purpose of determining the most significant phonological-based errors made by English-native learners of Arabic in writing Arabic. Based on the findings and the above discussion, it can be concluded that typical learner errors emanate from the fact that certain Arabic phonemes do not have any equivalent in English. This results in learners having a tendency to revert to the closest sounds to those in their mother tongue, which is subsequently reflected in their writing performance.
Address for correspondence: yam3@georgetown.edu and m.hamad@qu.edu.qa
Acknowledgment
The researchers are cordially thankful to the Arabic for Non-Native Speakers Centre’s students Mr. Bernard Scott O’Connor and Mr Gideon Moorhead who classified and analyzed the data. Their contribution was part of a Student Research Grant from the College of Arts and Sciences at Qatar University.
REFERENCES
Abboud, P. ed. 1968. Introduction to Modern Standard Arabic Pronunciation and Writing. Michigan: UMP.
Abu-Rabia, S. 2019. The Role of Short Vowels in Reading Arabic: A Critical Literature Review. Journal of Psycholinguistic Research. 48(4), pp. 785-795
Anon. 1971. A non-contrastive approach to error analysis. English Language Teaching Journal. 25(3), pp. 204-219.
Brown, G. 1977. Listening to spoken English. London: Longman.
Brustad, K., Al-Batal, M. and Al-Tonsi, A. 2011. Al-Kitaab fii Ta‘allum Al-‘Arabiyya: A Textbook for Beginning Arabic, Part One. Washington: Georgetown University Press.
Capuz, G. 1997. Towards a typological classification of linguistic borrowing. Revista Alicantina de Estudios Ingleses. (10), pp. 81-94.
Council of Europe. 2018. Common European Framework of Reference for Languages: Learning, Teaching, Assessment, Companion Volume with New Descriptors. 2018. Strasbourg Cedex: Council of Europe.
Cook, V.J., Long, J. and McDonough S. 1979. First and second language learning. In: Perren, G.E. ed. The Mother Tongue and Other Languages in Education, CILTR.
Corder, S.P. 1981. Error analysis and interlanguage. Oxford: Oxford University Press.
Cross, S. 1999. A Statistical Study of the Written Errors committed by Native English Speakers Learning Arabic as a Foreign Language. M.A. thesis, Ohio State University.
Dickins, J. 2007. Sudanese Arabic: Phonematics and Syllable Structure. Wiesbaden: Harrassowitz Verlag.
Duncan, D. 2016. “Tense” /æ/ is still lax: A phonotactics study. Proceedings of Annual Meeting on Phonology 2015, pp.1-12.
Ellis, R. 2000. Second Language Acquisition (Oxford Introduction to Language Study Series). Oxford: OUP.
Erdoğan, V. 2005. Contribution of Error Analysis to Foreign Language Teaching. Mersin Üniversitesi Eğitim Fakültesi Dergisi.
Fischer, W. 2001. A Grammar of Classical Arabic. Yale: YUP.
Frost, R. 1989. Orthographic depth and the interaction of visual and auditory processing in word recognition. Memory & Cognition. 17(3), pp.302-310.
Gimson, A. C. 1962, An Introduction to the Pronunciation of English, London: Edward Arnold.
Gosall N.K., Gosall G.S. 2012. Doctor's Guide to Critical Appraisal. Knusford: PasTest.
Greenbaum, S. 1996. The Oxford English Grammar. Oxford: OUP.
Al-Hamad, M. and Alaoui, H. 2016. ‘ghlāṭ naql al-qudrah ‘ind muta‘allimī al-‘arabiyya lughatan thāniyah. Linguistica Communicatio. 18(1-2), pp. 103-136.
Hanna, S. 1964. Problems of American College Students in Learning Arabic, A Diagnostic of Reading Errors, Remedial Instruction and A Proposed Method of Teaching, Ph. D. thesis, University of Utah.
Heselwood, B. 2007. The ‘Tight Approximant’ Variant of the Arabic ‘ayn. Journal of the International Phonetic Association, 37(1), pp. 1-32
Holes, C. 2004. Modern Arabic: Structure, Foundations, and Varieties. Georgetown University Press.
Jabeen, A. Kazemian, B. and Mustafai, M. 2015. The Role of Error Analysis in Teaching and Learning of Second and Foreign Language. Education and Linguistics Research, 1(2), pp. 52-61.
Kästner, H. 1981. Phonetik und Phonologie des modernen Hocharabisch. Leipzig: Verlag Enzyklopädie.
Khattab, Gh. Al-Tamimi, J. and Alsiraih, W. 2018. Nasalisation in the Production of Iraqi Arabic Pharyngeals. Phonetica, 75, pp. 310-348.
Khoury, J. 1961. Arabic teaching manual with an analysis of the major problems American high school students face in learning Arabic. Ph.D. thesis. University of Utah.
Kortlandt, F. 1993. General linguistics and Indo-European reconstruction. [Online]. [Accessed 15 October 2018]. Available from: http://www.kortlandt.nl/publications/art130e.pdf
Lott, D. 1983. Analysing and counteracting interference errors. ELT Journal 37(3), pp. 256-261.
McCarthy, J.J. 1994. The phonetics and phonology of Semitic pharyngeals. Keating, P.A. ed. Phonological Structure and Phonetic Form Papers in Laboratory Phonology III. Cambridge University Press, pp. 190- 233.
Nemser, W., 1991. Language contact and foreign language acquisition. In: Ivir, V. and Kalogjera, D. eds. Languages in Context and Contrast: Essays in Contact Linguistics. Berlin: Mouton de Gruyter, pp.345-364.
Peck, R. Olsen, C. and Devore, J.L. 2011. Introduction to Statistics and Data Analysis. Boston: Brooks/Cole.
Rammuny, R. 1987. Statistical Study of Errors Made by American Students in Written Arabic, University of Michigan, Learning Arabic, Ph. D. thesis, University of Utah.
Raslan, M. 1985. A program to teach Arabic to Malay students, Ph.D. thesis, Ain Shams University.
Richards, J. 1980. Second Language Acquisition: Error Analysis. Annual Review of Applied Linguistics. 1, pp.91-107.
Romaine, S. 1995. Bilingualism. Oxford: Blackwell Publishers Ltd.
Selinker, L. 1972. Interlanguage. International Review of Applied Linguistics in Language Teaching. 10(3), pp.209-231.
Schulz, E. 2010. A Student Grammar of Modern Standard Arabic. Cambridge: CUP.
Shosted, R.K., Fu, M., and, Hermes, Z. 2018. “Arabic Pharyngeal and Empathic Consonants”, In: E. Benmamoun, and R. Bassiouney. eds. The Routledge Handbook of Arabic Linguistics. London: Taylor and Francis, pp. 48-61.
Skiba, R. 1997. Code Switching as a Countenance of Language Interference. The Internet TESL Journal. 3(10), no pagination.
Thelwall, R. and Sa’Adeddin, M.A. 1990. Arabic. Journal of the International Phonetic Association. Cambridge University Press. 20 (2), pp. 37-39 [Accessed 15 October 2018]. Available from: http://journals.cambridge.org/action/displayAbstract?fromPage=online&aid=1799152
Underhill, A. 2007. Sound Foundations: learning and teaching pronunciation. UK: Macmillan.
Wagner, E. 2014. Using IBM® SPSS® Statistics for Research Methods and Social Science Statistics. Fifth edition. Los Angeles: SAGE Publications.
Watson, J.C. 2002. The Phonology and Morphology of Arabic. New York: Oxford University Press.
Wells, J.C. 1997. Whatever happened to received pronunciation? [Online]. [Accessed 19 November 2018]. Available from: http://www.phon.ucl.ac.uk/home/wells/rphappened.html
Wells, J.C. 1982. Accents of English 2: the British Isles. Cambridge: Cambridge University Press.
Wells, J.C. 1970. Local Accents in England and Wales. Journal of Linguistics. 6(2), pp. 231- 252.
Wiley, R.W., Rapp, B. 2019. From complexity to distinctiveness: The effect of expertise on letter perception. Psychon Bull. 26, pp. 974–984.