This article aims to use a corpus-based analysis to examine loan translation (LT) as an Arabicization technique in Written Standard Arabic (WSA) in the field of information technology (IT), and to investigate the factors that impact the formation of LTs. The data were collected from four distinct types of corpora: four Arabic online newspapers, the Microsoft Language Portal, two user manuals, and the Arabic Corpus of King Abdul-Aziz City for Science and Technology (KACST). Four LT strategies were identified: morphosyntactic pattern alteration, literal translation, loan blending, and metaphorical extension. Literal translation occurred in three degrees of calquing: full LTs, partial LTs, and non-LTs. Metaphors based on similarity in function play an important role in facilitating the understanding and acceptance of IT terms in WSA. The study also found that paradigmatic variation prevails in LTs, with synonymous variations being more common than nonsynonymous ones, and a total absence of syntagmatic variation. The KACST analysis showed that these variants differ a lot in terms of frequency of occurrence. The creation and variation of LTs are influenced by linguistic factors such as lexical need, semantic transparency, and lexical borrowing, as well as extralinguistic factors such as nationalism and the lack of coordination among Arabic language academies. The study concludes with some implications for overcoming the absence of coordination between these academies and for improving the quality of IT-related LTs between English and WSA.
Research Article
Arabicization via loan translation: a corpus-based analysis of neologisms translated from English into Arabic in the field of information technology
https://doi.org/10.21203/rs.3.rs-3408038/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
loan translation
Arabicization
IT terms
neologisms
Written Standard Arabic
English
1.1 Arabicization
Arabicization (or ʔat-taʕri:b) is an ongoing process that requires continuous improvement at both the institutional and individual levels. In a broader sense, Arabicization refers to transforming foreign elements into Arabic. Arab scholars agree that this term encompasses different methods (Baker, 1987; Ghazala, 2012; Khasara, 1998). Khasara (1998) identifies several concepts associated with Arabicization, including 1) the use of Arabic as the language for instruction and administration, 2) the translation of knowledge from different fields into Arabic, 3) the creation of Arabic scientific terms, and 4) the adaptation of foreign elements into Arabic through lexical borrowing. However, in this study, the focus is on the second and third concepts of Arabicization.
1.2 Loan translation
Loan translation (LT) occurs somewhere between lexical borrowing and translation. The terms LT, calque, semantic loan, and semantic calque are used interchangeably in the literature (Crystal, 2008; Zabawa, 2017). LT is a distinct type of borrowing that distinguishes itself from direct borrowing in that both borrowing and translation are used. In contrast to direct borrowing, where the donor language's form and meaning are adopted by the recipient language, LT only borrows the meaning while translating the form.
LT is a specific type of loan shift, which refers to morphemic substitution without importation (Haugen, 1972b). Backus and Dorleijn (2009, p. 77) define LT as "any usage of morphemes in Language A that is the result of the literal translation of one or more elements in a semantically equivalent expression in Language B." According to González and Knospe (2019), calquing occurs when a loan element from a donor language is replaced with a native term in a receiving language to create a conceptual replica of the model sign. Many researchers, including Capuz (1997), Haspelmath (2009), Haugen (1972b), Manfredi (2020), Thomason (2001), Trask and Millar (2015), and Zabawa (2017), define LT in a similar way.
Forms of LTs representing foreign meanings are always native. Examples of LTs in WSA include saţħ ʔal-maktab < destop, ʔašba:h ʔal-muwaşila:t < semiconductors, muħarrik ʔal-baħθ < research engine, ʔal-ʔamn ʔas-sibra:ni < cyber security, and kart ʔaš-ša:šah < graphics card. Thus, the new meanings of these neologisms are foreign (from English), but the forms are native (from Arabic).
LT is one of the most productive devices of Arabization because the neologisms created by this technique are Arabic and contribute to its purism, unlike other devices of lexical borrowing where the final products remain foreign. Ali (2005, p. 117) notes that LTs "seem to be more readily acceptable in almost all languages [and] they represent an intermediate solution allowing source language meanings to be expressed through the use of TL elements." LT is used to create terms in various fields such as economics and trade, politics and military, science and technology, media and sports, etc. Studies such as Ali (2005), Al-Wahy (2020), and Bader (1994) have investigated LT neologisms in these domains to some extent. However, research on LT in the IT domain is still limited, possibly due to its recent and rapidly growing nature.
This study, therefore, aims to bridge this research gap by investigating LT in IT domain. The study also explores the formal characteristics of LT, and the Arabicization strategies used to render English LTs into WSA.
1.3 The purpose of the study
This study aims to investigate the morpho-semantic structure of the Arabicized terms translated through LT. It seeks to elaborate on the different styles and strategies used in translating these terms. Putting differently, the present study attempts to address the following research questions:
1) What are the different Arabicization strategies used in loan-translating English IT terms into WSA?
2) What kind of lexical variation do LTs have?
3) What are the possible factors that may influence the creation of new IT terms in WSA?
Over the last seven decades, technological innovations, especially in IT, have led to the creation of new terms in many languages through LT. Extensive research was conducted on LT and IT terminology from both translational and linguistic perspectives in various languages. These include Bergh and Ohlander (2017), Bullock, Serigos, and Toribio (2021), Celiešienė and Juzeleniene (2020), Dimova (2007), González and Knospe (2019), Izwaini (2005), Oncins-Martinez (2009), Solano (2012), and Zabawa (2017; 2022). They all emphasized the significant role of English as an international language and a language of science and technology, serving as the source model for LTs.
Research on LT in Arabic within the IT domain is limited. While some studies examined the creation of technical terms, they did not specifically address LT phenomenon. Hamdan and Al-Salman (2021) investigated loanwords related to social media, Hassan (2017) analyzed selected terms from the Microsoft Terminology Collection, and Solimando (2017) discussed the lack of agreement among Arabic language academies in creating new IT terms. However, the subject of LT in Arabic within the IT domain was not addressed in any of these studies. The studies highlighted the impact of English on Arabic and its users and the traditional methods used in coining new words in Arabic, but more research is needed to study the role of LT in Arabic language in the IT domain.
In contrast, studies by Ali (2005), Al-Wahy (2020; 2022), Bader (1994), and Manfredi (2020) examined LT as a term-creation technique in Arabic. However, none of these studies investigated LTs in IT.
Bader (1994) focused on LTs in Jordanian news media. The study approached them from a lexical point of view and listed several examples related to politics, the military, economics, and science. Similarly, Ali (2005) examined calquing in standard Arabic in technical fields other than IT. The study identified seven types of calques, concluding that Arabic seems to prefer foreign-based neologisms to direct borrowing. It is worth noting that the data used for the study came from translations done by students, which may not reflect the authentic use of the language. Unlike these two studies, Manfredi (2020) investigated calquing in some Arabic dialects used in non-Arab countries such as Nigeria and South Sudan and provided a comprehensive overview of the notion of calquing. Additionally, the study discussed how calquing can be related to Van Coetsem's psycholinguistic principle of language dominance, which distinguishes between source-language agentivity and recipient-language agentivity.
Finally, Al-Wahy (2020; 2022) are highly informative and offer valuable contributions to phraseology and contact linguistics. However, these studies, like others, concentrated solely on LTs that have been calqued from non-IT domains and examined Arabic phraseological units (PUs) that have been calqued from English.
Al-Wahy (2020) proposed an explanatory approach, analyzing a corpus of data collected from various sources. He dealt with patterns of lexical variation in PUs that arise due to phonetic or semantic differences between languages, and those resulting from different cultural contexts or sociolectal influences. The study investigated the types and causes of lexical variation in the calqued PUs. PUs were defined as fixed or semi-fixed expressions with conventionalized meaning and usage, such as compounds and collocations. Al-Wahy argued that PUs are not as rigid as traditionally assumed but rather show a high degree of variability, especially when they are calqued from another language.
The explanatory approach combines linguistic and extra-linguistic factors to account for the variation, and suggested that lexical variation in Arabic calqued PUs is influenced by factors such as semantic transparency, syntactic compatibility, stylistic preference, and cultural adaptation. Two lexical variation types were identified: paradigmatic and syntagmatic. Paradigmatic variation involves using different words or phrases in the same position of the PU while syntagmatic was occurs when a word or phrase is added to the PU. The study found that paradigmatic variation is much more common than syntagmatic variation.
On the other hand, Al-Wahy (2022) offered new insights into the phenomenon of phraseological calquing in Arabic. He investigated the syntactic, semantic, and sociocultural factors that affect the acceptance of Arabic PUs calqued from English. The author proposed a borrowability scale for PUs that ranges from high to low, depending on the degree of transparency, decomposability, and cultural compatibility of the units. He argued that some PUs can be easily calqued from English into Arabic, while others are more resistant to this process, concluding that the degree to which a PU can be borrowed depends on various factors, such as syntactic structure, semantic properties, and cultural relevance. This scale can help identify which PUs are most likely to be borrowed successfully and which ones may require more adaptation.
Both studies make valuable contributions to phraseology and contact linguistics. They provide insights into the nature and causes of lexical variation in Arabic calqued PUs and offer a systematic framework for describing and explaining such variation. In addition, the studies emphasize the significance of corpus-based methods and interdisciplinary perspectives in the study of PUs and their variability.
This study draws on the theoretical framework of Al-Wahy (2020; 2022) to account for lexical variations that may occur in IT terms calqued from English into WSA. By utilizing this framework, the study contributes to the understanding of calquing and the factors that affect the borrowing of phrasal units in a specific linguistic context.
The previous review indicates that LT in the IT domain in Arabic still requires further attention and investigation. To date, there have been no studies that have specifically examined the linguistic and translational aspects of LT in this field. Therefore, this study aims to fill this gap in WSA. By exploring the linguistic and translational issues related to LT in the IT domain, this study seeks to contribute to a deeper understanding of the challenges involved in calquing and provide insights into effective strategies for achieving successful LTs in this context.
The study applied a qualitative method to deal with the LT data. Since the IT domain encompasses numerous sub-disciplines such as computer and the internet the researcher collected a sample from all these domains. The data were gathered from written online sources, which were classified into four distinct types of corpora, as outlined in Table 1.
The first source comprised more than 250 texts collected from four online Arabic newspapers including Asharq Alawsat and Sabq (KSA), Al-Ahram (Egypt) and Albayan (UAE), all of which feature IT sections. These texts were scanned in order to obtain the available LT data. The second source is the Microsoft Language Portal which contains a vast collection of IT terms that enables users to search for terms in different languages. Thus, searches for English-Arabic pairs were performed and saved in TBX file format (Microsoft Term Collection.tbx). The third source refers to two user manuals for mobile phones and laptops, containing the original texts in English along with an Arabic translation. These parallel texts facilitate extracting and collecting LT data.
The last source is the Arabic Copus of King Abdul-Aziz City for Science and Technology (KACST). It is a tremendous electronic searchable Arabic corpus that contains approximately one billion words. It enables users to search for Arabic words in their various contexts. KACST covers different historical periods of Arabic starting from the pre-Islamic period up to the modern age. However, for this study, the search is restricted to materials related to the modern period, specifically newspapers, news agencies, magazines, and the internet. The main purpose of using ACKAC is twofold: to validate and identify the contexts in which Arabic LTs are employed and to search for their frequency distribution and concordance.
All the online written sources stated above were navigated for data collection, resulting in the attestation of hundreds of IT terms. From this large number of data, we identified approximately 3350 IT tokens. However, for this study, we focused only on a sample of 300 ITs, representing the most frequent and newly-coined LTs used in IT domain in Arabic. This sample was selected to provide a representative overview of the lexical and semantic structures of Arabic LTs.
The unit of analysis is the phrasal units calqued from English; the one-word LTs were excluded. Moreover, LT is usually categorized into two types: lexical calquing and grammatical calquing. Lexical calquing involves the transfer of semantic properties of lexical items, whereas grammatical calquing implies the transfer of functional properties of morphemes and syntactic constructions (Manfredi, 2020, p. 625). This study focuses solely on lexical calquing rather than grammatical one, as the latter involves the borrowing of grammatical aspects at the sentence level, whereas the focus of this study is on phrasal units. Therefore, grammatical calquing is beyond the scope of this study.
| No. | Online source/ corpus | language | website |
|---|---|---|---|
| 1. | Sabq newspaper | Arabic | https://sabq.org/technology |
| 2. | Asharq Alawsat newspaper | Arabic-English | https://aawsat.com/home/international/section/information-technology |
| 3. | Al-Ahram newspaper | Arabic | https://cutt.us/Q9sfg |
| 4. | Albayan newspaper | Arabic | https://www.albayan.ae/technology |
| 5. | Microsoft Language Portal | English-Arabic | https://www.microsoft.com/en-us/language/Search |
| 6. | User manual, Lenovo IdeaPad 100 | English-Arabic | https://cutt.us/NOO6x, https://cutt.us/elCAz |
| 7. | User manual, Samsung Galaxy, Note10 Lite | English-Arabic | https://cutt.us/h50XS, https://cutt.us/Vrfjw |
| 8. | The Arabic Corpus of King Abdul-Aziz City for Science and Technology (KACST) | Arabic | https://corpus.kacst.edu.sa |
With the help of the KACST corpus, the data were first examined to identify the main features of LTs. Then, the data were qualitatively analyzed to identify the Arabicization strategies used in calquing English terms into WSA, explore their lexical variations, and understand the linguistic and non-linguistic factors affecting LT creation.
One salient feature is related to the context concordance in which an LT appears. For instance, the LT ʔal-faḍa:ʔ ʔar-raqami ‘the digital space’ is consistently found in the IT context (as shown in Fig. 1) while the LT saţħ ʔal-maktab may appear in more than one context, including the IT domain (as shown in Fig. 2). In the latter case, the LT is based on the English term ‘desktop’. Thus, the Arabic combination saţħ ʔal-maktab is assigned a new sense, that is ‘a computer screen displaying files, programs, etc.’ In other contexts, saţħ and ʔal-maktab usually refer to ‘a surface (saţħʔ) of a desk (maktab) located in a room’. Figure 2 illustrates the concordance of these two types of contexts of saţħ ʔal-maktab. The general contexts are highlighted in bold.
The other important feature pertains to the frequency distribution of LTs across the various domains. Table 2 shows the frequency of eight LTs which were selected randomly from the data. The analysis revealed that out of ten source categories of the KACST corpus, the majority of LTs are frequently encountered in three main sources: newspapers/news agencies, magazines, and the internet. This is because these domains are heavily influenced by the use of foreign languages, particularly English. As a result, many of the technical terms and concepts used in these domains are originally expressed in English. Arabic writers and professionals who work in these fields often find it easier to use LTs, which are direct translations of these terms into Arabic, rather than creating new Arabic terms. Additionally, the fast pace of these industries and the need for quick communication also contribute to the widespread use of LTs, as they provide a convenient and efficient way to convey complex concepts and ideas.
4.1 Strategies of LTs
When calqued into WSA, English IT terms exhibit certain translation strategies, which will be presented below.
4.1.1 Morphosyntactic pattern alteration
Witalisz (2015) classifies phrases calqued from one language into another as “reversing” or “non-reversing” constructions. Non-reversing constructions maintain the sequence of the components in the original construction (e.g. modifier + head is retained) while reversing constructions invert the order (e.g. modifier + head becomes head + modifier). In WSA, almost all LTs are of the reversing type with the modifier + head constructions becoming head + modifier like the examples in (1a-d). For instance, ‘artificial intelligence’ is rendered as ʔað-ðaka:ʔ ʔaş-şina:ʕi, where ʔað-ðaka:ʔ ‘intelligence’ is the head and ʔaş-şina:ʕi ‘artificial’ is the modifier. This is because Arabic phrases have a different syntax than English. Thus, we conclude that neologisms adhering to the fundamental rules of Arabic syntax, which require adjectives/modifiers to be positioned after the noun, can be considered acceptable.
Other morphosyntactic alterations involve changes in word class, affixed words, and acronyms. Concerning word class, three situations emerge: word class of LT constituents may remain the same (1a), have some changes (1b), a preposition can be added (1c), or undergo nominalization (1d). Changes are shown in bold. The second example in (1b) is interesting; the syntactic structure of the English term ‘about us’ (related institutional websites) is transformed from a phrase into a question, man naħnu, meaning literally ‘who are we?’. In (1c), a preposition is typically inserted before the modifier. For instance, ‘video conference’ is calqued as muʔtamar ʕabr ʔal-fi:diyu, where the preposition ʕabr means ‘through’ or ‘via’, resulting in ‘conference through video’. Finally, the verbs ‘insert’, and ‘arrange’ in (1d) are nominalized, emerging in ʔidra:j ‘inserting’, and tarti:b ‘arranging’, respectively. The imperative form is also used (e.g. ʔadrij laqţat ša:šah, etc.) but it is less frequent.
Many English IT terms are affixed words, and when translated into WSA, their structure is also altered. The affixes, whether prefixes or suffixes, become free morphemes because the English and Arabic affixation systems differ as in (2).
LTs modeled on English words that are affixed with prefixes are more numerous than those with suffixes. Out of thirty-four cases, only two LTs are found with suffixes, which are the last two examples in (2). The English affixes are either converted into nouns or adjectives in the calqued expressions. For example, the prefix ‘semi-’ in ‘semiconductors’ is transferred as the noun ʔašba:h, meaning ‘something like’ in ʔašba:h ʔal-muwaşila:t, and the prefix ‘mal-’ in ‘malware’ is rendered as the adjective xabi:θah, meaning ‘malicious’ in bara:mij xabi:θah.
The last point in this section concerns the calquing of English acronyms. In Arabic morphology, the occurrence of abbreviations and acronyms is infrequent and lacks consistency (Al-Qinai, 2007). Therefore, foreign acronyms in WSA are either rendered in their full forms by using the full words of a given acronym (3a) or by adding a gloss before or after the full form of the acronym, especially with longer acronyms consisting of more than two initials (3b). In the first situation, English acronyms are also used as direct loanwords such si: di: ‘CD’ and bi: si: ‘PC’, but only in Arabic dialects, not in the standard variety like WSA. Thus, the hypothesized Arabic abbreviations like (*ق.م.) for CD or (*ك.ش.) for PC are never used neither in the standard or colloquial varieties of Arabic. In the second situation, the glossed part represents the acronym itself and is enclosed in brackets either in Latin characters as in the first and third examples of (3b) or in Arabic transliteration as in the second example of (3b), usually after the acronym’s full form/explication (cf. Arrabai, 2022).
4.1.2 Literal translation
This strategy pertains to the degree to which the source expressions are calqued. As previously defined, an LT is a word-for-word translation of a phrasal unit from the source language into the recipient language. However, literal translation is not always straightforward. It rather occurs in various degrees. According to Al-Wahy (2020), English phraseological units can be transferred into Arabic in three main ways: full calque (each word in the phrase is translated literally), partial calque (some words are translated literally and others are translated more freely), and non-calque (the whole phrase is translated freely, using Arabic words that express the same idea rather than the same wording). Full calques are also known as “exact calques” while partial calques as “inexact calques” or “loan renditions” (zabawa, 2022, pp. 609–610). All three methods have been attested in the data of IT terms in WSA, with full calquing being the most frequent, followed by partial calquing as shown in (4a-c).
The examples in (4a) demonstrate literal translation, where the English models are translated element by element: ʔism for ‘name’ and ʔal-mustaxdim for ‘user’, resulting in ʔism ʔal-mustaxdim ‘user name’; muškila:t for ‘problems’, lawħat for ‘panel’ and ʔal-ʕarḍ for ‘display’, resulting in muškila:t lawħat ʔal-ʕarḍ ‘display panel problems’, etc.
In (4b), the components in boldface in each calque are not translated literally but instead are rendered into another native word that conveys the intended meaning of the term, resulting in a partial calque. For example, instead of translating the word ‘hard’ in the first example as şulb(ah), it is instead translated as maţbu:ʕah, meaning ‘printed’. Moreover, literal translation in such instances may lead to a lack of semantic transparency. If the English model ‘hard copy’ is calqued as *nusxah şulbah for ‘hard copy’, this would make it opaque and unacceptable.
The third type is non-LTs or non-calques, which is very interesting but rather controversial. All the constituents of each calque are translated freely in such a way that they convey the underlying meaning of the calqued units. For instance, the literal translation of the English term ‘wallpaper’ should be something like ʔawra:q ʔal-ħa:ʔiţ, but it is transferred as xalfiyyat ʔaš-ša:šah ‘the background of the computer screen’, conveying the intended meaning of the source form ‘wallpaper’. The second example in (4c) is also a non-calque because it is translated freely into ʔal-ħa:sib ʔal-maħmu:l ‘portable computer’ to stand for English ‘laptop’ while the element-by-element translation of the compound will be nonsensical.
While it can be argued that non-LTs, by definition, do not involve literal translation and may not be categorized as LTs, the components of the calqued phrasal units are still native (Arabic) and ultimately express the semantics of the original terms. Therefore, they are subsumed as one type of LT. Furthermore, non-calques may exhibit lexical variation patterns that could be absent in the source units (see section 4.2 below).
4.1.3 Loan blends
In Haugen’s (1972b) terms, loan blending involves both importation and substitution, a process whereby one part of the phrasal unit is borrowed directly as a loanword, while the other part is substituted by a native word. Therefore, a loan blend is a mixture of calquing and lexical borrowing and sometimes is referred to as semi-calque (Zabawa, 2022). For instance, in ‘laser printer’, the component ‘printer’ is translated literally into the native ţa:biʕah while layzar remains foreign and based on the English term ‘laser’. Similarly, the boldfaced foreign elements in (5) such as banura:ma ‘panorama’, burutuku:l ‘protocol’, and sibra:ni ‘cyber’, are all directly borrowed from English and integrated into the native language using the loan blending strategy.
This borrowing strategy is often used when the borrowed element is difficult to translate, not easily substituted by words in the target language, or has already been in use in the target language for a significant amount of time.
4.1.4 Metaphorical extension
Metaphor is a figurative form that draws comparisons between a literal object or concept and a new, unrelated object or concept based on similarities in form, function, relationship, appearance, or other characteristics (Sihler, 2000, p. 108). Metaphor is a fundamental and widespread linguistic device that is commonly used not only in creative or artistic expressions, but also in scientific, technical, and social communication (Zabawa, 2017, p. 88). In WSA, metaphorical associations is often based either on similarity in shape or similarity in function.
The use of metaphors in IT is less frequent in comparison to other technical domains like mechanical engineering (cf. Al-Athwary, 2023). The present data show only thirteen instances of metaphorical extensions, and some examples are given in (6).
WSA calquing sometimes leads to the adoption of English metaphors in Arabic constructions. For instance, the calques faʔrat ʔal-ħa:su:b, and ʔal-lawħah ʔal-ʔum are metaphors that have been modeled on the English metaphor ‘computer mouse’ and ‘motherboard’, respectively. These metaphorical extensions are associated with the terms (ʔal-)faʔrah/mouse, and ʔal-ʔum/mother.
However, some other metaphors are created only upon their calquing into the language and do not exist in English. Examples of this type include ʔaš-šabakah ʔal-ʕankabu:tiyyah for ‘World Wide Web’ and ħašara:t ʔat-tajassus for ‘spybots’. In these cases, the metaphors lie in the words ʔal-ʕankabu:tiyyah and ħašara:t, respectively. The first word is an adjective derived from the Arabic noun "ʕankabu:t", meaning ‘spider’, and the second word is a noun and literally means ‘insects’.
The last example in (6) demonstrates that a metaphorical expression in English may lose its metaphorical meaning when calqued into WSA. Here, ‘home’ in ‘home page’ carries a metaphorical character while in WSA it is transferred as ʔaş-şafħah ʔar-raʔi:siyyah which literally means ‘the main page’. Thus, the word ‘home’ is rendered as ʔar-raʔi:siyyah ‘the main’, and the figurative form bayt or manzil ‘home/house’ is not used.
All metaphors in the data are either based on similarity in shape or similarity in function with the second being the most common. The former can be exemplified by faʔrat ʔal-ħa:su:b ‘computer mouse’ and wisa:dat ʔal-faʔrah ‘mouse pad’ where the likeness in shape is obvious: the mouse device is like the animal mouse and the mouse bad is like ‘a billow’. All other examples of LTs in (6) are instances of the latter, i.e. similarity in function. The metaphorical element ʔal-ʔum/mother in both the source and calqued constructions indicates the essential and significant function of this ‘type of board’ in the computer device which resembles the mother’s function at home. Also, the behavior of ʔal-fayru:sa:t ‘computer viruses’, which involves multiplying and infecting other systems, can be regarded as similar to that of biological viruses.
4.2 Lexical variation of LTs
In the LT context, lexical variation is defined as “the presence of different lexemes in different calques of the same phraseological unit which are all used to express the same meaning or perform the same function” (Al-Wahy, 2020, p. 281). For instance, English ‘password’ is calqued in WSA in three ways: kalimat ʔal-muru:r ‘password’, kalimat ʔas-sir ‘secret word’ and ʔar-raqam ʔas-sirri ‘the secret number’. All of these variants carry the same meaning of the referent ‘password’.
Calqued expressions often have many differences from source expressions, and therefore they may show lexical variation patterns that are not typical of the source language. These differences are related to their origin (emerging in a different culture and reflecting a different way of thinking), degree of conventionalization (usually less fixed than original constructions), and context of use (more common in formal situations and media discourse) (Al-Wahy, 2020, p. 276). This variation is the result of the interaction between a number of linguistic and nonlinguistic factors (see the discussion section below).
Lexical variation does exist in Arabic LTs calqued from English in the IT domain. In Arabic, the main reason that may lead to the rise of this phenomenon is the absence of coordination among Arabic language Academies in the Arab World.
Based on the taxonomy provided by Al-Wahy (2020), the lexical variability of LTs is divided into two types: paradigmatic variation and syntagmatic variation.
4.2.1 Paradigmatic variation
Paradigmatic variation refers to the use of different words in the same position within the calqued expression. It is more frequently observed than syntagmatic variation. This type of variation is also referred to as “synonymic variation” by González and Knospe (2019) because the alternative forms used in a given LT are often synonymous.
Paradigmatic variation in WSA can be divided into two types based on the synonymity of the variant forms in a given LT: synonymous (7) and (8) and non-synonymous (9).
Examples in (7) show that the synonyms are native and derivationally unrelated. For instance, the term ‘data’ in ‘database’ is calqued as ʔal-baya:na:t ‘data’ and ʔal-maʕlu:ma:t ‘information/data’, and thus are synonymous. The first constituent of the English phrase ‘access provider’ has four counterparts in WSA: ʔad-duxu:l, ʔal-wuşu:l, ʔal-wulu:j, and ʔan-nafa:ð, which all mean ‘access’. The same idea applies to the other examples. However, in the context of the last example, both components that make up the LT can have variant forms, which results in the formation of three potential calques for the English 'compact disc': qurş maḍġu:ţ, qurş mudmaj, and ʔisţiwa:nah mudmajah where qurş and ʔisţiwa:nah refer to ‘disc’ and maḍġu:ţ and mudmaj(ah) refer to ‘compact’. The English words that have Arabic variants in the input form are highlighted in bold in the examples.
Examples in (8) are LTs having variant forms that are also synonymous, but one of them is a borrowing. In the three examples, the loanwords ʔal-kumbiyu:tar ‘computer’, ʔas-sibraniyyah ‘cyber’, and ka:rt ‘card’ are used together with their Arabic literal equivalents ʔal-ħa:sib, ʔal-maʕlu:ma:tiyyah, and biţa:qah, respectively. In the third example, the borrowed adjective ʔal-ʔiliktru:niyyah ‘electronic’ can be used instead of ʔas-sibraniyyah.
The second type involves variant forms that are not necessarily synonymous. The lack of synonymity between variant forms is often due to the use of full calque and partial calque translation methods (9a-b), or the use of full/partial calque and non-calque translation methods (9c-d). This can result in differences in meaning or nuance between the original and loan-translated forms. For instance, example (9a) includes two variants for the translation of ‘user manual’: ʔal-mustaxdim ‘user’ is a full calque of the English term, and ʔat-tašġi:l ‘operation’ is a partial calque in which the word ‘user’ in the English compound is translated more freely rather than literally. Likewise, in (9b), the affix 'bio-' has two translation variants: ʔal-ħayawwiyyah ‘biological’ is a full calque using literal translation, and ʔal-ʔamniyyah ‘security’ is a partial calque using a free translation instead.
The examples in (9c-d) represent a slight departure from the previous cases. In (9c), the first variant, ʕaşa ʔal-ʔalʕa:b ‘game stick’, is a partial calque of English ‘joystick’, while the second variant, miqwad ʔal-ʔalʕa:b ‘game steering wheel’, is the result of a free translation of the two components, creating a non-calque. Similarly, in (9d), the first variant is a full calque (the two are translated literally), while in the second variant, the first component of the English term ‘password’ is translated freely, resulting in the non-calque kalimat ʔas-sir ‘the secret word’.
The variant forms are used with different frequencies, as obtained from the KACST corpus. To ensure accurate frequencies from KACST and to prevent occurrences from other domains, the search tools on the corpus have been restricted to those domains that potentially contain IT terms.
Table 3 indicates that the frequencies of variant forms of each LT are not similar. It also shows that variants with loanwords (e.g. ʔal-kumbiyu:tar ʔaš-šaxşi (554 times) and kart ʔaš-ša:šah (12 times)) are preferred to variants with native words (e.g. ʔal-ħa:sib ʔaš-šaxşi (154 times) and biţa:qat rusu:ma:t (5 times)). This preference for loanwords can reflect the influence of English as a dominant language in the IT domain and its penetration into different cultures.
| Original term | Variants of LTs in WSA | Frequency |
|---|---|---|
| database | 1) qa:ʕidat ʔal-baya:na:t | 1426 |
| 2) qa:ʕidat ʔal-maʕlu:ma:t | 324 | |
| inbox | 1) şundu:q ʔal-wa:rid | 27 |
| 2) ʕulbat ʔal-wa:rid | 2 | |
| compact disc | 1) qurş maḍġu:ţ | 80 |
| 2) qurş mudmaj | 372 | |
| 3) ʔisţiwa:nah mudmajah | 88 | |
| personal computer | 1) ʔal-ħa:sib ʔaš-šaxşi | 154 |
| 2) ʔal-kumbiyu:tar ʔaš-šaxşi | 554 | |
| graphics card | 1) biţa:qat rusu:ma:t | 5 |
| 2) kart ʔaš-ša:šah | 12 | |
| password | 1) kalimat ʔal-muru:r | 383 |
| 2) kalimat ʔas-sir | 876 |
Syntagmatic variation, which involves adding a word or phrase to an existing calqued expression, was not found in the data for IT domain, although it is common in other semantic domains such as politics and literature (cf. Al-Wahy, 2020).
The study showed that some of the LTs identified in the KACST corpus may appear in contexts other than the IT domain, indicating a potential for these terms to be used in a broader range of domains. This finding suggests that LT is not only a strategy for Arabicizing IT terms but could potentially contribute to the enrichment of the Arabic language in general.
It is necessary to underline that what is borrowed in LT is the meaning while the form is always native. Moreover, the semantics of LTs never changes. By contrast, their morphosyntactic structure may be adjusted to fit Arabic grammar. All English “modifier-head” constructions are reversed into “head-modifier” ones, including “complex calques” that consist of three or more components such as the last two examples in (4a) (cf. Zabawa 2022).
The findings indicated that LTs are only associated with one type of lexical variation which is paradigmatic variation, but not syntagmatic variation. This result is not in line with Al-Wahy (2020) who found that both types do exist in Arabic phraseological units calqued from English in other fields of knowledge such as politics and literature.
Moreover, the analysis revealed that the Arabicization of IT terms through LT is not a uniform or straightforward process. Instead, it is highly influenced by linguistic and non-linguistic factors which means that Arabicization is not just a linguistic issue, but also a social and cultural one that involves the interaction between language, technology, and society.
One of the linguistic factors that may govern the creation of LTs in WSA is lexical need. The results of this study disagree with Al-Wahy (2022) who concluded that the majority of Arabic phraseological units in fields like politics are mostly borrowed for reasons of prestige rather than necessity. In this study, and unlike other domains of knowledge, LTs, in the IT domain, are adopted from English due to the lexical need for IT terms in WSA.
Semantic transparency is another linguistic factor that leads to LTs formation (González & Knospe, 2019). Transparency is a concept that varies in degree and pertains to the capacity to infer the meaning of a phrasal unit based on the meanings of its constituent words or morphemes, which is determined by whether those components are used in a literal or metaphorical sense. Most calques from English are most probably classified according to their degree of transparency rather than decomposability because they are non-idiomatic phraseological units and consequently used literally not figuratively as is the case in expressions from non-technical fields (cf. Al-Wahy, 2022). As analyzed in section (4.1.4) above, figurative calquing may occur in the IT domain, but the LTs created are still transparent. Metaphors based on similarity in function is a common feature of technical terminology in many languages, and it allows for the transfer of complex ideas and concepts between different languages. These metaphors contribute to the development of a unique technical vocabulary and style of expression in Arabic.
The Arabic language academies’ efforts are fruitful, but they can be sometimes problematic. They are fruitful in that they promote a sense of linguistic nationalism which strengthens support for native terms, including LTs. Nationalism is one of the extralinguistic factors that enhances the use of LTs in WSA (cf. González & Knospe 2019). Another non-linguistic factor refers to the lack of coordination among these academies, which results in various solutions being presented in dictionaries and specialized literature, leading to synonymous variation. This causes confusion in scientific and cultural communication among WSA users. Therefore, IT terms must be unified and documented in reliable and accessible references.
Examples in (8) above indicate that lexical borrowing is an additional linguistic factor that may lead to paradigmatic variation. It is worth noting that although LTs with loanword variants are usually less formal than those with native variants, they are much more frequent than their native counterparts, as shown in Table 3 regarding the examples ʔal-kumbiyu:tar ʔaš-šaxşi ‘personal computer’ and kart ʔaš-ša:šah ‘graphics card’. This is because loanwords are often associated with colloquial language use, whereas native equivalents are considered more formal and more prestigious.
The purpose of this study was to investigate Arabicization through LT in the field of IT. Using a corpus-based approach, the study analyzed a sample of neologisms calqued from English into WSA, with a focus on identifying the most common LT strategies and their frequency of occurrence. In addition, Al-Wahy’s (2020) explanatory approach was adopted to examine the lexical variations that may exist in Arabic LTs.
The majority of LTs in WSA are frequently encountered in newspapers, magazines, and the internet as shown by the KACST corpus analysis. The study identified four main strategies used in Arabicizing IT terms, including morphosyntactic pattern alteration, literal translation, loan blending, and metaphorical extension. The most common strategy used was morphosyntactic pattern alteration, which involved reversing the order of phrase constituents, changing the word class, and rendering acronyms into full forms. Literal translation occurred in three degrees of calquing: full LTs (element-by-element translation), partial LTs (some components are translated freely), and non-LTs (all the constituents of each calque are translated freely). Loan blending involved borrowing one part of the phrasal unit directly as a loanword, while substituting the other part with a native word. Finally, metaphorical extension involved using metaphors in the data based on similarity in function or shape. It was also found that paradigmatic variation is prevalent in LTs, with synonymous variations being more common than nonsynonymous ones, with no instances of syntagmatic variation in the data. The KACST analysis showed that these variants differ a lot in terms of frequency of occurrence. Moreover, the study identified the linguistic and extralinguistic factors that influence the creation and variation of LTs. Linguistic factors included the lexical gap in vocabulary, semantic transparency, and lexical borrowing. Extralinguistic factors included linguistic nationalism and the lack of coordination among Arabic language academies.
The study's findings have some important implications. One of these implications is that The study's results can be used to improve the quality of IT-related translations between English and WSA. By identifying the common lexical patterns for LTs in this domain, translators can ensure that their translations are more accurate and natural-sounding. They can also help improve machine translation systems. By incorporating the identified lexical variations into machine translation algorithms, the accuracy and naturalness of the translations produced by these systems can be improved. Furthermore, The IT data of LT can constitute a basis for an IT database in standard Arabic and consequently a basis for compiling a specialized bilingual dictionary.
Another notable aspect of this study is the role of metaphorical extension in the Arabicization of IT terms through LT. The study found that most of the metaphors identified in the data were based on similarity in function rather than shape. This finding suggests that metaphors play a vital role in facilitating the understanding and acceptance of IT terms in WSA and can contribute to the development of specialized IT terminology in Arabic.
The prevalence of paradigmatic variation over syntagmatic variation implies that Arabic speakers tend to use different but equivalent terms to express the same concept, which could contribute to the richness and flexibility of the Arabic language.
Paradigmatic variations reflect the diversity of the Arabic language and its interaction with other languages and cultures. It can affect the understanding, acceptance, and standardization of IT terms in Arabic-speaking communities. One implication is that lexical variations can lead to confusion and misunderstanding among Arabic-speaking communities, especially in the context of cross-border communication and collaboration. For instance, a particular IT term that is commonly used in one Arabic-speaking country may not be understood or recognized in another country, leading to communication breakdowns and delays. Another implication is that lexical variations can hinder the standardization of IT terminology in Arabic. The lack of standardization can lead to inconsistencies and inaccuracies in the translation of IT terms, which could affect the quality and reliability of IT-related content and applications in Arabic.
Therefore, the Arabic language academies should collaborate and coordinate with each other in order to standardize LTs calqued from English in the IT field. This can be achieved by 1) establishing a standardization committee of experts in linguistic and technical fields, 2) developing a standardization process, 3) creating a specialized IT dictionary, 4) providing training and resources to dictionary compilers, and translators, and 5) collaborating with IT industry to cope with the latest technological advances and the corresponding IT terms in English.
The current study enhances some further research. The more specialized technical terms (together with the common ones) might be examined in future research which may lead to the compilation of a database and/or bilingual dictionary in the field of IT terminology. Moreover, it is important to note that syntagmatic variation may still occur in specific contexts within the IT domain. Future efforts could investigate the occurrence and patterns of syntagmatic variation in different domains and contexts in WSA.
Funding: This research work was funded by the Grants Program in the field of translation, affiliated with the Arabic Observatory for Translation, Ministry of Culture, Saudi Arabia under grant number (150) for the year 2023.
Conflict of interest: The authors declare that they have no conflict of interest to declare.
Author Contributions Statement: A.A. worked on the major part of the manuscript including the methodology, analysis, results, and discussion. H.K. did the literature review and the conclusion. Both authors prepared figures and tables. Both authors proofread and edited the manuscript.
- Al-Athwary, A. A. (2023). A comparative lexical analysis of vehicle spare parts terms in Arabic. Saudi Journal of Language Studies, 3(1), 19-31. https://doi.org/10.1108/SJLS-08-2022-0065
- Ali, A. S. M. (2005). Calquing: A means of terminological enrichment. Turjuman: revue de traduction et d'interprétation= journal of translation studies, 14(1), 113-135.
- Al-Qinai, J. (2007). Abbreviation and acronymy in English Arabic translation. Meta, 52(2), 368-375.
- Al-Wahy, A. (2020). Lexical variation in Arabic phraseological units calqued from English: an explanatory approach. Textual Turnings: An International Peer-Reviewed Journal in English Studies, 1(1), 275-300.
- Al-Wahy, A. (2022). Towards a Borrowability Scale for Phraseological Units: The Case of Arabic Calques from English. CDELT Occasional Papers in the Development of English Education, 78(1), 77-103.
- Arrabai, A. M. (2022). A Corpus-Based Study on the Use of Glossing in Saudi Technology Blogs. Arab World English Journal, 13(4), 562-573. https://doi.org/https://dx.doi.org/10.24093/awej/vol13no4.36
- Backus, A., & Dorleijn, M. (2009). Loan translations versus code-switching. In B. E. Bullock & A. J. Toribio (Eds.), The Cambridge Handbook of Linguistic Code-switching (pp. 75- 93). Cambridge: Cambridge University Press.
- Bader, Y. (1994). Loan translations in written Jordanian news media. In R. de Beaugrande, A. Shunnaq, & M. H. Heliel (Eds.), Language, discourse and translation in the West and Middle East (pp. 91-101). Amsterdam: John Benjamins.
- Baker, M. (1987). Review of Methods Used for Coining New Terms in Arabic. Meta: Journal des traducteurs, 32(2), 186-188. https://doi.org/10.7202/001950ar
- Bergh, G., & Ohlander, S. (2017). Loan translations versus direct loans: The impact of English on European football lexis. Nordic Journal of Linguistics, 40(1), 5-35. https://doi.org/10.1017/S0332586517000014
- Bullock, B., Serigos, J., & Toribio, A. (2021). Exploring a loan translation and its consequences in an oral bilingual corpus. Journal of Language Contact, 13(3), 612-635.
- Capuz, J. G. (1997). Towards a typological classification of linguistic borrowing (Illustrated with Anglicisms in Romance Languages). Revista Alicantina de Estudios Ingleses, 10, 81–94.
- Celiešienė, V., & Juzeleniene, S. (2020). Semantic motivation of the terminologized lexis in the field of digital technologies. Journal of Teaching English for Specific and Academic Purposes, 031-042. https://doi.org/10.22190/JTESAP2001031C
- Crystal, D. (2008). A Dictionary of Linguistics and Phonetics. Oxford: Blackwell Publishing.
- Dimova, S. (2007). English, the Internet and computer terminology in Macedonia. World Englishes, 26(3), 373-387. https://doi.org/10.1111/j.1467-971X.2007.00515.x
- Ghazala, H. S. (2012). Arabization from A to Z: A textbook. Jeddah: Konooz Al-Marifa.
- González, F. R., & Knospe, S. (2019). The variation of calques in European languages, with particular reference to Spanish and German: Main patterns and trends. Folia Linguistica, 53(1), 233-276. https://doi.org/10.1515/flin2019-2009
- Hamdan, H., & Al-Salman, S. (2021). The use of Arabic neologisms in social media applications. International Journal of Arabic-English Studies (IJAES), 21(1), 45-60.
- Haspelmath, M. (2009). Lexical borrowing: Concepts and issues. In M. Haspelmath & U. Tadmor (Eds.), Loanwords in the world’s languages (pp. 35–54). Mouton de Gruyter.
- Hassan, S. (2017). Translating technical terms into Arabic: Microsoft terminology collection (English-Arabic) as an example. Translation & Interpreting, The, 9(2), 67-86.
- Haugen, E. (1972b). The analysis of linguistic borrowing. In E.S. Firchow, K. Grimstad, N. Hasselmo, & W. O’Nell (Eds.), Studies by Einar Haugen (pp. 161–185). The Hague: De Gruyter Mouton.
- Izwaini, S. (2005). A corpus-based study of the translation of IT terms. In Edizioni B.A. Graphis (ed,), ESP Across Cultures, 2, (pp. 76-94). Foggia (Italy): Università degli Studi di Foggia.
- Khasara, M. (1998). naħw manhajiyyah lit-taʕri:bi l-lafḍiyy [Towards a methodology of verbal Arabicization]. The Journal of Arabic Language Academy in Damascus, 73: 747-796.
- Manfredi, S. (2020). Contact and calquing. In C. Lucas & S. Manfredi (Eds.), Arabic and contact-induced change (pp. 625–641). Language Science Press.
- Oncins-Martínez, J. (2009). Towards a corpus-based analysis of anglicisms in Spanish: A case study. International Journal of English Studies, 9(3), 115-132.
- Sihler, A. (2000). Language History: An Introduction. Amsterdam–Philadelphia: John Benjamins Publishing Company.
- Solano, R. (2012). Multi-word loan translations and semantic borrowings from English in French journalistic discourse. In C. Furiassi, V. Pulcini, & F. González (Eds.), The anglicization of European lexis, 10, 199-215.
- Solimando, C. (2017). Arabic neologisms in IT terminology: the role of the Academies. Bulletin d’études Orientales (LXV), 123-133.
- Thomason, S. G. (2001). Language contact: An Introduction. Edinburgh University Press.
- Trask, L., & Millar, R. M. (2015). Trask’s historical linguistics (3rd ed.). Routledge.
- Witalisz, A. (2015). English loan translations in Polish: word-formation patterns, lexicalization, idiomaticity and institutionalization. Frankfurt am Main: Peter Lang.
- Zabawa, M. (2017). English semantic loans, loan translations, and loan renditions in informal Polish of computer users. Katowice: Wydawnictwo Uniwersytetu Śląskiego.
- Zabawa, M. (2022). English loan translations in Polish in the area of computers: Syntactic aspects. Poznan Studies in Contemporary Linguistics, 58(3), 607-647. https://doi.org/10.1515/psicl-2022-0026
No competing interests reported.