1 Introduction

 Rafat Y. Alwazna, Zoe Hurley, Tariq Elyas

Legal translation between Arabic and other languages, whether or not used in the Middle East and North Africa (MENA), poses real challenges due to de facto differences in legal systems and legal cultures, which give rise to terminological asymmetry and conceptual incongruency. Legal Arabic, as any legal language, is deemed a technical language, which is viewed as an adjunct of typical/ordinary language [1, 2]. It is claimed that legal language is employed in legal contexts that ascribe different meanings to legal terms from those they usually relay in non-legal contexts [3]. This is lent credence by Hart [4], who argues over the merit that legal terms acquire their legal meanings within a particular legal context and through specific legal provisions. From the syntactic perspective, legal language exhibits a formal and impersonal style, comprising convoluted, lengthy and complex sentences [1, 5]. Such complexity of sentences in legal texts is demanded in statutes, for instance, due to the intricate nature of the legal issues contained therein and that of the legislative law [1]. This applies to the majority of legal languages. Legal language is also characterised by an extensive use of conditions, qualifications, and exceptions, which are adopted to describe complicated contingencies [5]. Such linguistic features, as Bhatia [6] points out, create problems of understanding to non-specialists, including translators. Therefore, legal translators should take into account the specific problems caused by the aforementioned features [7].

As previously mentioned, legal Arabic is considered a legal language, which goes back to Babylon with introducing Hammurabi code [8]. During ʿabd Almalik Ibn Marwān’s rule (685–705), in the Umayyad era, Arabic was adopted as the official Empire language, thus replacing Greek in Damascus, Coptic in Egypt and Pahlavi in Iraq [1]. The translation of multifarious official documents into legal Arabic played a substantial role in the Arabising movement at that time [9]. Certain terms in Arabic are indeed borrowed from Persian and Greek [10]. Arabic is the language of the two primary sources of Islamic Law, namely the Qurʾān and the Sunna (the Prophetic traditions of the Prophet Mohammad (peace be upon him)) [11]. Arabic is also the language of both Islamic jurisprudence as well as Islamic jurisdiction. In Islamic jurisprudence, legal Arabic contains different opinions of Muslim scholars on a specific legal issue [12]. This is in order not to restrict judges to a particular legal source and give them a leeway to choose the legal ruling that is compatible with the legal case under investigation. In such Fiqh texts, Qurʾānic verses and Prophetic traditions are provided along with explanations and examples to uncover and elucidate the practical application of the legal provision in question [1, 12]. Legal Arabic, which is used in Islamic jurisprudence texts, often contains religious, legal, system-based and culture-specific terms/statements [13, 14]. It further encompasses archaic terms and gender-biased terms [1]. The formality of legal Arabic is usually expressed through forms of address or honorary titles [1, 8, 13]. Syntactically, legal Arabic is marked by lengthy and complex sentences [1, 13], with multiple clauses that separate the subject and the main verb [1, 12]. Furthermore, modal expressions, active participles as well as passive participles are deemed syntactic features of legal Arabic [1]. It is noteworthy that legal Arabic is also used in secular laws in Arab states with more inclination towards Modern Standard Arabic (MSA). Having considered the foregoing, it goes without saying that translating laws from and/or into Arabic poses major challenges and causes acute problems at the legal, linguistic as well as cultural levels.

However, due to the largely increasing demand for legal translation in trade, International Law and diplomacy and owing to the importance and the pivotal role played by legal Arabic across the globe, technology has become an indispensable tool in translating laws all over the world, particularly from and into Arabic. The present paper addresses the role of integrating semiotic technologies in Arabic legal translation, including texts, images and videos, highlighting the benefits, drawbacks and future directions. It introduces at the outset the use of computer-assisted translation (CAT) tools in Arabic legal translation, underscoring the merits and challenges involved in their use and highlighting the importance of the role of human translators in the whole process. It then addresses the use of machine translation (MT) systems in Arabic legal translation, casting light on how they have come into existence along with discussing their main types and how they are used. Within the context of presenting the benefits of the use of MT systems in Arabic legal translation, the paper confirms the significant role of the human post-editing in rectifying potential errors at the linguistic, cultural and/or legal levels. A complete section is devoted to addressing the use of artificial intelligence (AI) in Arabic legal translation, highlighting its benefits, the challenges involved in its implementation and the necessity of the human post-editing to exercise after each dependence on AI-generated translation, particularly in Arabic legal translation domain. The paper next offers a summary of each paper published in the current special issue, highlighting the main argument made by each of them. The paper concludes with a special emphasis placed on the dynamic nature of the MENA’s semiotic landscapes, arguing over the merit of the necessity of considering the legal semiotic context along with possessing the appropriate linguistic, cultural and legal knowledge to achieve acceptable Arabic legal translation. Thus, the human element remains essential to achieve this arduous legal process appropriately while integrating technology, such as CAT tools, MT systems and AI tools, is regarded as a significant supplementary tool that not only reduces human effort, but also accelerates the entire process of legal translation.

2 The Use of CAT Tools in Arabic Legal Translation

CAT tools, such as MemoQ, Trados, Wordfast, and so on, are largely utilised in the translation of legal texts to maintain consistency and enhance accuracy as well as productivity. Such tools are chiefly contingent upon terminology databases and translation memories, which are deemed particularly effective in the translation of legal texts on account of the repetitive nature of such texts along with legal terms [15, 16]. In Arabic legal translation, CAT tools play a substantial role in ensuring terminological consistency, which is of paramount importance considering the lexical features that characterise legal Arabic, such as archaic terms, gender-biased terms, formal terms as well as religious, culture-specific and system-based terms and expressions [8, 17]. Maintaining consistency in the use of these terms is highly demanded in Arabic legal translation not only to ensure accurate rendering of the intended legal meaning, but also to guarantee full conveyance of the intended legal effect for the sake of application and implementation [14, 18]. CAT tools can also provide Arabic legal translation with creation of translation memories, thus storing previously used terms and reducing redundancy. They help maintain quality assurance through enhancing consistency, performing checks for terminology and formatting, while reducing the rate of possible errors.

Conversely, certain challenges may arise from the use of CAT tools in Arabic legal translation on account of the complex Arabic morphology and the diglossic nature of Arabic (the use of Modern Standard Arabic (MSA) versus different Arabic dialects), which can impact the compatibility and retrieval of translation units [19]. What is more, the creation of accurate, precise and comprehensive terminology databases specific to legal Arabic stands as a major challenge. Nonetheless, Arab and other legal translators who translate texts from and into legal Arabic, do employ CAT tools for they achieve coherence in lengthy legal documents of long-winded sentences, while reducing human exerted efforts. It is worth noting that CAT tools are effective supplementary aids that assist legal translators, but cannot replace human translator’s expertise. The complexity of legal Arabic syntax, including nominalisation, participles, modals as well as lengthy and complex sentence structures [1] with no doubt demands human intervention.

3 The Use of Machine Translation Systems in Arabic Legal Translation

MT is viewed as one of the first non-numerical applications of the digital computers, which came into existence after the Second World War [20, 21]. MT is classified under computational linguistics, which investigates the use of computer software in the translation of texts across different languages [22]. The intricacies of natural languages make MT a formidable task. These intricacies manifest themselves in the polysemic nature of certain terms, the sentences that are open to different interpretations as well as the de facto grammatical relations in a specific language, but not in the others [21, 22]. Based on the foregoing, MT systems tend to make errors at different levels, including lexical, grammatical, collocational, syntactical as well as stylistical levels along with omissions and additions [23].

MT systems can broadly be divided into two types: rule-based machine translation (RBMT) and corpus-based machine translation (CBMT) [21]. According to the former, expert linguists lay down certain rules to elucidate the process of translation. This demands a large amount of input on the part of the linguists [24]. This necessitates offering the programme all the words particular to the source and receptor languages in addition to the rules required for sentence structure in the languages in question. This is followed by a specification of the method through which the words of both languages are linked, providing the programme with guidelines of how to use such information to generate translated texts [20]. The latter, however, entails carrying out the extracted knowledge through analysing translation examples drawn from a compiled parallel corpus. After designing the relevant techniques for a specific language pair, MT systems are established to comprise more language pairs, utilising the restored data [22]. Such paradigm hinges upon the premise that the programme would acquire and learn the data stored therein rather than being offered the information it needs [21].

Two important approaches are derived from the CP paradigm, namely statistical machine translation (SMT) and neural machine translation (NMT). There are two statistical models in SMT grounded in the training data known as the translation model and the language model [21]. The translation model is a bilingual model containing the source text words and their translations in the target language in a table, with each source-target language pair is provided with a probability score. Such translation examples in SMT are used to train the translation model [22]. By contrast, the language model is deemed monolingual or an amalgamation of models of the target language [20]. NMT, on the other hand, makes use of examples contained in parallel corpora. These MT systems are chiefly contingent upon bilingual corpora that include parallel texts and treat them as their crucial data foundation. NMT is viewed as a translation by analogy and is considered a practical approach of machine learning built on a specific case [22]. The witnessed advancement in translation made by NMT, compared to SMT, may spring from the type of models it is trained on and the type of representation it produces [23]. It is worth pointing out that NMT employes models that are human-inspired. It exploits artificial neural networks in which multifarious artificial neurons are linked to other multiple artificial neurons [20, 21].

Although MT systems, particularly NMT systems, have made an undeniable improvement on translation across various text types, Arabic legal translation still remains problematic due to the syntactic complexity and culture-specific terms/statements that characterise Arabic legal texts [1, 14, 21] alongside the lack of comprehensive Arabic legal corpora that can serve as a training data for such MT systems. The main problem resides primarily in the fact that Arabic legal concepts designated by culture-specific terms/statements contained in Arabic legal texts comprised in Arabic legal language derived from Arabic legal system (Islamic Law) and Arabic legal culture (Islamic culture) do not exist in other legal systems as the law has no universal reference [18, 25]. Therefore, MT from and/or into legal Arabic cannot be relied on individually, however, human post-editing is highly demanded, especially if the legal translation is achieved for application and implementation purposes. Nonetheless, MT still remains a good supplementary tool, even within the context of Arabic legal translation, as it undergoes ongoing development and improvement. Translators usually exploit MT systems within Arabic legal translation to obtain the first draft, which, with no doubt, requires human post-editing to rectify possible terminological errors, modify syntactic structures and replace certain legal terms for the sake of accuracy and precision. Moreover, human post-editing is necessary to ensure full adherence to the legal norms, style and formatting.

4 The Use of Artificial Intelligence in Arabic Legal Translation

Technology within legal context has unquestionably made an evident progression in the form of AI, machine learning, deep learning as well as natural language processing all over the globe [21]. Specifically, the field of legal translation has witnessed an unprecedented technological evolution through employing AI tools in both legal translation practice and legal translation pedagogy. Indeed, AI refers generally to the ability of the machine in fulfilling tasks that often require human-like intelligence, such as pattern recognition, problem-solving and natural language processing [26]. It has appeared as a quick method of achieving the information required and as a current independent subject studied at a university level. It clearly impacts human existence in different aspects [27]. AI is deemed the science of creating intelligent technology, which is based on computer algorithms development. Computational linguistics as well as natural language processing can be employed to establish an interdisciplinary science consisting of AI and standard linguistics [28]. Indeed, AI technology has taken its way of being applied to translation practice, translation pedagogy and scoring of writing [29]. This is lent credence by Yang [30], who points out that with the largely developed MT systems, AI-generated translations have emerged with new creative functions and scenarios constructed in new MT apps to provide better quality translations to human users. Such evolution in translation technology has been driven by development in machine learning algorithms, computational power as well as the availability of big datasets [31].

The development of AI in relation to translation practice has impacted translation jobs [32]. Professional translators have begun utilising AI tools in their translation work. The reliance on AI-generated translations has been so common amongst translators on account of their suitability and effectiveness. AI may also play a significant role in the reduction of translation cost and fostering intercultural communication [26]. With regard to Arabic legal translation in particular, AI, including machine learning and natural language processing, provides different opportunities in fostering Arabic legal translation. For instance, natural language processing techniques can be adopted to identify key legal terms, analyse features of legal texts and explore appropriate information, supporting legal translation research and terminology management [33]. Machine learning algorithms can possibly be trained for the sake of pattern recognition and translation error detection, thus enhancing legal translation quality. This is crystallised by Koehn [34], who claims that AI-generated translations outperform traditional MT through encapsulating contextual nuances and offering high-quality translations.

Even though there has been an undeniable development in AI-generated translations, particularly with regard to Arabic legal translation, certain challenges and limitations still exist, which need to be surmounted by both researchers and technicians alike [35]. Such challenges may chiefly lie in the translation of legal metaphor [36], culture-specific terms/statements [14], and so on. In addition, the development of rigorous AI tools for Arabic legal translation demands big datasets, which seem to be limited at current times. What is more, ethical considerations, including algorithmic bias, data privacy, fairness and job displacement should receive careful treatment [37]. Furthermore, the overdependence on AI-generated translations may affect students’ translation abilities and creative thinking skills [38,39,40]. Finally, Human post-editing remains substantial after achieving AI-generated translations due to the complexity and intricacy of Arabic legal language. This is seconded by Awadh [41], who asserts that AI-generated translations still have insurmountable problems, particularly with regard to Arabic legal translation domain.

5 A Brief Account of the Research Papers Published in the Current Special Issue

This interesting special issue entitled: ‘Semiotic Assemblage, Translanguaging and Translating Laws of the Middle East and North African Region’, co-guest edited by Rafat Y. Alwazna, Zoe Hurley and Tariq Elyas, consists of eleven important research papers, the first of which is entitled: ‘Translating Metaphor in Legal and Non-Legal Quranic Verses into English: A Corpus-Based Study’ by Alwazna and Alfaify. The paper addresses the translation of metaphor as a problematic issue within legal religious texts. It particularly discusses the rendition of Quranic explicit and implicit metaphors in legal as well as non-legal verses into English. It aims to explore and analyse the most appropriate procedures that should be adopted in rendering Quranic explicit and implicit metaphors, which not only relay the source text image itself to the target language, but also provide an explanation thereof. The paper makes use of the Quranic Arabic Corpus to draw a comparison of seven translations of Quranic explicit and implicit metaphors for the sake of exploring the commonalities and distinctions between them and the translation procedures employed by each of the seven translations in light of Newmark’s [42] types of procedures. The paper concludes that amongst the most relevant procedure for rendering Quranic explicit and implicit metaphors is a combination of literal translation and sense-for-sense translation. This can also be used in translating the part of the ellipted vehicle stated in the verse if the implicit metaphor is grounded in personification. This combination guarantees the preservation of the source text image in the target text, while elucidating its sense. This ensures comprehensibility of the Quranic metaphor translation to both the specialist and nonspecialist alike.

The title of the second paper is: ‘A Corpus Linguistic Perspective on the Lexicon of Islamic Family Law in English: Legal Communication or Cultural Discourse?’ by Roshdy. It seeks to examine the discursive norms specific to the translation of Islamic family law, making use of a mixed methods approach built on ‘qualitising’ quantitative data, which is an approach through which quantitative data are qualitatively interpreted. The analysis of the paper controls the power of corpus processing, allowing the study of ten iconic lexical profiles (individual concepts/conceptual systems and their corresponding Arabic and English terms), which are peculiar to Islamic family law based on a self-built monolingual English corpus covering a set of different genres. The paper also employs a profile-based correspondence to measure the probability of lexical differences in encapsulating conceptual categories, in an attempt at designating the same concept. The corpus analysis aims to pinpoint the crucial loanwords which typify cultural icons included in English discourse of Islamic family law and investigate the distinct mechanisms employed to translate concepts specific to Islamic Law. The paper concludes that Islamic family law hinges substantially upon Arabic loanwords to preserve the linguistic heritage particular to Shari’a Law, which can be viewed as a kind of ‘working for one’s own good’, thus having the ethics of self-interest, which are akin to postcolonial agendas.

The third paper is entitled: ‘Phraseology and Terminology Challenges and Approach to Translating Divorce Decrees’ by Halimi. It seeks to explore translation challenges encountered in rendering a specific type of Arabic civil status deed known as divorce decrees. To achieve this, the paper addresses with detail the most crucial concepts particular to certified translation and specific to Arabic divorce decrees. It offers a descriptive analysis of multiple Arabic divorce decrees to identify shared linguistic aspects and delineate the appropriate translation techniques in coping with religion-based terminology and phraseology. This is followed by a specific assessment of the relevance of the current analysis through investigating a French certified translation of an Arabic divorce document to arrive at a valid conclusion about such approach to rendering such legal genre.

The title of the fourth paper is: ‘Exploring Translator Choices Facing the Possible Death of Truth in Translating Family Law of Algeria’ by Goui. It explores at the outset the challenges involved in the translation of the Algerian family law, which was officially translated into French and unofficially rendered into English. It then examines the accuracy level and the unelectable gaps, aiming to trace the death of the truth contained in the original rulings. Following this path, the translators’ choices can possibly be explored and investigated, in an attempt at reducing such translation loss and examining the reasons behind it. The paper makes use of the Algerian government’s officially translated version of family law into French and published on the official website of the People’s Democratic Republic of Algeria as its main corpus. The paper claims that translating family laws originally written in Arabic is mostly considered a thorny operation. In particular, translating Algerian family law from Arabic into French, or even into English, is imbued with translational and translatorial challenges due to the differences in legal backgrounds, religion, culture, environment, and so on.

The fifth paper is entitled: ‘Translation of Arabic Legal Documents into English: A Semiotic Analysis of the Translation of the Saudi “Basic Law of Governance”’ by Alowedi. It aims to analyse the English translation of the Arabic legal document pertaining to the Saudi Basic Law of Governance. It seeks to assess the accuracy of preserving the semiotic connections between the signifier and the signified within the corresponding seme or sign, considering the Arabic context and Saudi culture. The English target text is assessed against the Arabic source text in view of the appropriate use of lexicon and word order. The paper makes use of a semiotic approach, in an attempt at analysing the legal document in question. The paper argues that to achieve a high translation quality, certain procedural adjustments should be made as a sign is not born with a certain meaning, though it can change its referent through the presence of adjacent signs. It also claims that semiotic analysis can assist in demonstrating how the rendition of legal documents can preserve specific beliefs, while dispensing with others.

The title of the sixth paper is: ‘Semiotic Study of the Gaza Conflict on Social Media’ by Hurley. It carries out a Peircean semiotic analysis of social media influencers’ variation in representation of Palestinian and Israeli in the current Gaza conflict. The paper undertakes at the outset a thematic review of relevant semiotic literature, representation, social media and conflict. It then sets a novel framework specific to the Middle East semiotics and is used to conduct a comparative semiotic analysis of how Palestinian and Israeli social media influencers represent the conflict on Instagram. The paper concludes that the dimensions of the Middle East semiotics’ framework may be adopted to underscore and analyse the intricate semiotic implications involved in the opposing depictions of conflict at epistemological, ontological, technological and axiological levels. This would offer a view that goes beyond individual social media influencers’ analysis to take into consideration the axiological impacts and social media platforms’ legal responsibilities of how to deal with such conflict.

The seventh paper is entitled: ‘Wearing the Veil in the Web: Transformations of Social Norms and Everyday Practices in the Digital Sphere within the MENA Region’ by Greco and Leone. It seeks to explore the veiling practices, which are traditionally linked to specific concepts of modesty as well as religious identity, and which are currently governed by new translations and dynamic reinterpretations within digital sphere. Drawing on Lotman’s [43] theory of cultures, such dynamics are likely to be viewed as cultural translation processes, navigating semiotic boundaries. Through examining these dynamics by analysing semiotically a corpus of digital representations of the veiling practices across the MENA region, the paper contributes to a deeper understanding of the cultural adaptations, which give rise to identity negotiation, providing a perspective on the way in which the developing nature of cultural symbols translates into societal values and legal frameworks.

The title of the eighth paper is: ‘Integrating Corpora and AI Tools in the Teaching, Learning and Assessing of Arabic-English Legal Translation’ by El-Farahaty. It seeks to present the use of corpus tools and AI tools to enhance research-driven approaches and consolidate blended learning methods in Arabic-English legal translation. It makes use of advanced corpus tools and multiple corpora in Sketch Engine, such as the Leeds Monolingual and Parallel Corpora of Arabic and English Countries’ Constitutions as well as the UN Parallel Corpora. The paper also utilises ChatGPT in developing teaching activities using legal documents. Such a mixed methods approach may enhance interactive learning of legal texts, underscoring educational and research innovation. The paper acknowledges the limitations of AI tools in guaranteeing precision and accuracy in translating complex legal texts.

The ninth paper is entitled: ‘AI Adoption in Legal Translation: A Study of Intrinsic Motivation among Saudi Students’ by Brashi. It examines the interplay of intrinsic motivation, perceived ease of use, perceived usefulness and behavioral intention, with special emphasis on ChatGPT adoption when rendering legal texts by Saudi students at a university level. The paper seeks to extend the Technology Adoption Model through investigating the internal drivers of adapting technology (intrinsic motivation). It offers insights into AI appropriate design and implementation strategies, which can particularly be useful for both higher education institutes and AI tools’ developers. The findings of the paper confirm the significance of the role played by intrinsic motivation led by positive perception and enjoyment students find in using such AI tools, which may serve as a key mediator in strengthening ChatGPT adoption in translation.

The title of the tenth paper is: ‘Evaluating Free Legal Translation Tools between Arabic and English: A Comparative Study of Google Translate, ChatGPT and Gemini’ by Farghal, Shraideh and Al-Omari. It investigates the complex area of MT used for rendering legal texts, in an attempt at discussing crucial linguistic aspects specific to legal discourse. The paper makes use of a corpus of legal partnership contracts issued by the Jordanian Hashemite Fund for Human Development Organisation. It also employs a number of automatic evaluation metrics, including BLEU, ChrF and METEOR as well as adequacy and fluency as human evaluation criteria. The paper compares the translation output quality of Google Translate, ChatGPT and Gemini as well as human translation. The findings of the paper indicate that the translation output quality of ChatGPT is higher than both Google Translate translation output quality and that of Gemini in view of all the said metrics, with regard to the level of precision, the rate of the edits required and the level of semantic accuracy. Such results confirm the strength of ChatGPT-generated translations and highlight the importance of choosing the optimum MT/AI tool based on specific factors, such as meaning preservation, accuracy and fluency.

The eleventh paper is entitled: ‘The Quality of Translated Legal Terms Regarding Children in the MENA Region: A Comparative Study’ by A’al Jehjah. It seeks to explore challenges specific to the quality of legal terms rendered from Arabic into English in Laws peculiar to children in five countries, namely Iraq, Libia, Saudi Arabia, Sudan and the United Arab Emirates in the MENA region. It particularly addresses the terms that describe children’s behaviors, which are deemed socially and legally unacceptable and adult behaviors that breach laws which preserve the rights of children in both local as well as international contexts. The paper develops an analytical model known as CCFND, using a mixed methods approach, consisting of descriptive and analytical methods, comparative legal research tools as well as critical discourse analysis. The findings of the paper suggest that the translation of the aforementioned terms from Arabic into English may be of a better quality if such translation of these terms hinges upon a diverse set of translation strategies. It is recommended that the King Salman Global Academy for the Arabic Language compiles a bilingual contextual dictionary of appropriate legal translation techniques for terminology and phraseology particular to children in both local as well as international laws.

6 Dynamism of MENA’s Semiotic Landscapes

Problems of meaning, communication and interpretation intensify within the processes of translation. These issues are not entirely resolved by computer-assisted translation tools or machine translation systems due to the semiotic complexities of context. While the present paper does not propose to resolve the challenges of legal translations in the Middle East context, it has engaged with these linguistic, technological and semiotic complexities in relation to legal matters. MENA is a distinct area where several laws, ranging from sacred laws to secular laws, are applied to different institutions in the area. Moreover, the conceptual and legal asymmetries, translations and tensions between laws within and outside the MENA region make the process of legal translation a daunting yet crucial task that needs to be assessed by professional and expert legal translators as well as semioticians who possess the relevant knowledge, competence and awareness of the dexterity of signs. Moreover, we need to appreciate how the processes of legal translation occur in relation to the contextual range of meanings and interpretation. Hence, English translations of legal terms across Arabic-speaking MENA countries can differ according to the political, social, cultural, and economic circumstances of the country in question. In addition, this can result in varying legal provisions having different wording, technical expressions and outcomes. Consequently, failure in conveying the nuanced legal meaning of any legal term may affect the practical application of the target legal text, thus impacting the entire legal effect of the law in question.

Concurrently, semiotic assemblages in MENA are underpinned by the region’s differing sociocultural trajectories, including varying semiotic resources, historical and contemporary objects and interpretations of the semiotic landscape. This involves a range of social actors who are immersed and absorbed in spaces defined by hybrid legal frameworks as well as those confronting the challenges of humanitarian crisis, war and conflict. To explore the unquestionable interpretative challenges of MENA’s semiotic legal assemblages, semiotics facilitates an expanded version of meaning which attends not only to the borders between languages, but also to the borders between semiotic modes, legalities and their applications. The semiotic perspective of language, meaning and translation as a dynamic process, rather than merely a fixed system of knowledge, enables far deeper engagement with MENA’s multifaceted legal context as well as tangible strategies for not shying away from the propensity for misinterpretations of the source legal texts. When working with texts that were written in a foreign language, scholars and students of law and legal translation need to consider the practice of semiotic translanguaging, interpreting and explaining the source legal text, including diverse semiotic assemblages of different terms, concepts and imaginaries for the purpose of achieving acceptable legal communication.

Overall, semiotic research can help to shift our understandings of language from a focus on linguistic signs in the public domain to include greater contextual (ethnographic), historical and ethical understandings of texts in the landscape. This also raises questions about who put them there, how they are interpreted and what role they play in relation to space, race, ethnicity, gender, class, migration, mobility, technologies and law. Finally, in relation to questions of AI-generated translations, this holistic semiotic perspective does not debunk the importance of human actors since they remain pivotal to the semiotic context of meaning production, application, consumption and ethical interpretation.