1. Introduction

The reception of a translation is entangled with the ontology and epistemology of meaning. In fact, a major facet contributing to the complexity of the phenomenon of translation is the philosophy of language and meaning. Notwithstanding its long historical precedent, the controversy surrounding the relationship between author, text, and its meaning continues to generate debate. Those subscribing to a conduit metaphor of language would have us believe that the text is an autonomous, neutral carrier of meaning intended by the author (Turner 2018). Accordingly, a given text is associated with one universal meaning that is decodable across readers. In this view, form and content are not only separable but independent (Arrojo 2010; Chan 2016).

Recent advances in sociocultural theory, hermeneutics, and discourse and cognitive processing, however, challenge the view that meaning resides in the text alone, by recognizing the role of the reader and their agency in constructing meaning. As such, a cluster of views, generically termed reader response theory, posits that meaning is constituted in the interaction between the reader and the text (Beach 2013; Chan 2016). This view concerning the locus of meaning has also been extrapolated to how translated texts are made sense of and comprehended. Nevertheless, in translation, matters become even more complex, since, in addition to the translated text and the reader, there is a third force: the source text, whose meaning is subject to the same controversies as that of the translation. Hence, we are left with a host of factors influencing the meaning of the translated text: the reader, the text itself, the source text, and the culture.

This complexity is further compounded when it comes to reading the translation of sacred texts, especially the Qur’an, the very translation of which is surrounded by controversy (Fawcett & Munday 2009; Mustapha 2009; Peachy 2013; Spolsky 2004). The untranslatability argument of the Qur’an is founded on the Muslims' belief that the Qur’an is the very word of God, and its genuine meaning can never be reproduced in human language. Notwithstanding divergent stances concerning issues of untranslatability (Peachy 2013), the Qur’an has been and continues to be translated into various languages because the majority of Muslims across the globe do not speak Arabic, nor do non-Muslim readers who take an interest in reading the Qur’an. Given that the majority of readers, both Muslim and non-Muslim, access the meaning of the Qur’an through translation (Elimam 2017), the diverse backgrounds that these readers have across different linguistic and cultural contexts—and the acknowledged role of reader characteristics in the construction of meaning—make research on how readers' characteristics bear on their reception of the Qur’an translations highly warranted.

In sum, although the role of the reader in giving meaning to translation was recognized quite early on with the work of Nida on dynamic equivalence (as cited in Pym 2009) and there has been even stronger recognition of the reader in subsequent theories, this theoretical recognition did not culminate in empirical research in the reception of Bible translations (Gambier 2018). Nor has there been a robust research program on the reception of the Qur’an translations and the reader characteristics that bear on their reception.

This interdisciplinary study draws on psychometrics and insights from language policy to help narrow the noted gap regarding readers' linguistic background and their reception of the Qur’an translation. We pursued this goal in the multilingual, multicultural context of Khuzestan, a province in the southwest of Iran, to explore how readers' language attitudes might be related to Farsi translations of the Qur’an.

In the remainder of this paper, after discussing the theoretical framework of the study (i.e., translation reception), the related literature on translation reception in general and the reception of Qur’an translations specifically is reviewed. This is followed by a description of the context of the study. Next, the design of the study and data collection procedures are explained. Subsequently, analytic procedures and the results are presented, and the paper concludes with a discussion of the theoretical and practical implications of the study, as well as its limitations.

2. Theoretical Framework

 Translation reception, alternatively called translation audience studies, has its precedents in other disciplines such as rhetoric and literature. In rhetoric, the role of the audience has always been a major concern (Berlin 1982). In literature, reader response, or the reader’s subjective and unique appreciation of a literary text, has been a major concern (Beach 2013). This concern is rooted in our expanded view of the nature of the relationship between text and meaning. While in formalist and New Criticism approaches meaning was imagined to reside in the text, according to Reader Response Criticism, textual meaning emerges in interaction with readers’ subjectivities (Beach 2013; Brems & Pinto 2013).

Initially focused on the final output of translation, the field of Translation Studies has evolved to become more aware of the complexity of the translation phenomenon. As such, translation is now seen as a social event with various historical, political, social, cultural, linguistic, and individual dimensions. This expansion of the concept of translation (Tymoczko 2010) has contributed to a deeper understanding of translation as a social event in which not only the source and target languages, but also the translator and the reader, play crucial roles (Khoshsaligheh et al. 2020). This awareness has contributed to both translator and reader visibility. In reception studies, the focus shifts away from the translated text and translator toward the reader (Brems & Pinto 2013). That is, how a text is made sense of depends on readers’ horizons of expectations, which are determined by their membership in interpretive communities (Kotze et al. 2021).

In the literature on translation reception, a distinction is made between social and individual levels of analysis. The former “looks at the reception of translations at a social level and focuses on theoretical readers.” This line of reception is concerned with the broad impact of translation: how it is received in the target culture and the influence of translated literatures on national literature. In contrast, the latter level of analysis “looks at reception at a more individual level and focuses on real readers” (Brems & Pinto 2013, 143). In this approach, research is often concerned with one of the following areas: a) the mental processes triggered when people receive translated content, b) how readers evaluate the various strategies used in translation, and c) how different contextual, sociological, technical, or linguistic factors influence translation reception (Kotze et al. 2021; Brems & Pinto 2013, 145-146). In terms of its theoretical orientation, the current study is situated within the latter area of the micro approach to translation reception that focuses on the real, individual reader.

3. Literature Review

The emerging literature on translation reception has addressed cognitive processes (Kruger & Crossley 2018; Chen 2023), contextual factors (D’Angelo 2019; Khoshsaligheh et al. 2020; House 2015; Ghamsarian & Sanatifar 2020), and the assessment of translation strategies (Venuti 1995; Munday 2012; Kotze et al. 2021). Regarding cognitive strategies, Kruger and Crossley (2018) examined how readers process translations compared to original texts, highlighting differences in comprehension and engagement. Chen (2023) further explored the cognitive strategies readers use when processing complex translations, emphasizing the role of memory and attention.

With regard to contextual factors bearing on translation reception, technical and linguistic aspects have been addressed. D’Angelo (2019) found that digital translations might offer interactive features that enhance reader engagement. Khoshsaligheh et al. (2020) examined the impact of subtitling on audience reception, highlighting the importance of context in audiovisual translations. Specifically, they found that Iranian readers give priority to faithfulness to the tone of the original and immersion in the translation.

Other scholars have addressed language-related factors that influence the reception of translation. As such, linguistic factors, such as language proficiency and familiarity with the source language, play a crucial role. House (2015) emphasized the importance of considering the target audience’s linguistic competence to ensure effective translation. Ghamsarian & Sanatifar (2020) explored how dialectal variations influence reader reception, stressing the need for context-specific adaptations.

The third strand of research on translation reception deals with the assessment of translation strategies. This involves assessing the effectiveness of different approaches, such as literal translation, adaptation, and localization. Venuti (1995) argued for the importance of recognizing the translator’s visibility and the need for balance between foreignization and domestication strategies. Empirical studies, such as those by Munday (2012), have explored how these strategies affect reader satisfaction and comprehension. Kotze et al. (2021) analyzed the reception of translated literature and found that strategic choices significantly impact reader enjoyment and understanding. Other scholars have examined the reception of more specialized translations, such as medical translation (Valdez 2022) and the reception of Chinese classic literature (Zheng & Fan 2023).

Compared to other genres, less scholarly attention seems to have been paid to the reception of sacred texts. Further, in comparison to the Bible, the translations of the Qur’an have witnessed far less empirical scrutiny, possibly for the following two reasons. One relates to the very notion of the Qur’an’s untranslatability, noted earlier, and the second has to do with the status and legitimacy accorded to the Qur’an translations. As Peachy (2013, 37) notes, in the eyes of Muslims, “the Qur’an in any other language than the original Arabic is not the Qur’an.” Accordingly, the Qur’an translations are not given a similar legitimacy and authority that, say, Bible translations receive. “There is, therefore, no translation of the Qur’an with any status or influence comparable to, for example, the Authorized Version of the Bible in English, or Martin Luther's version in German” (Barnes 2011).

From a reader response theoretical perspective, Farghal and Al‐Masri (2000) examined readers’ responses to the English translation of the Qur’an verses with referential gaps, and they found that the presence of referential gaps creates comprehension problems for readers who read the English translations of the Qur’an. Further, they concluded that readers of translations commonly overestimate their understanding of the Qur’an.

Framing his argument in terms of why, by whom, for whom, and how, Peachy (2013) argues that for a translation of the Qur’an to succeed, the translator cannot afford to ignore the intended audience of translation. Only if such aspects are taken into account would a translation accomplish its goal of “conveying the Qur’anic message.” In particular, the question 'for whom' the translation is to be done is of central importance. To meet the needs of the audience of English translations of the Qur’an, Peachy and his co-translator make accommodations, which would have been considered unnecessary or even problematic from the standpoint of those endorsing overt translation of the Qur’an. Peachy's (2013) approach to the translation of the Qur’an is reminiscent of the shift from text and translator to the reader in translation reception studies (Brems & Pinto 2013). The argument is that without more empirical data on readers' responses and assessments of translated texts, current translation strategies and tactics would continue lacking empirical testing; the process of audience design would continue to be unable to address the needs and expectations of real readers; and finally, translators would continue to be left to their own devices and to work based on assumptions often grounded on individual stereotypes and prejudices.

Notwithstanding Peachy's (2013) strong support for considering the reader in translating the Qur’an, his arguments and subsequent translation of the Qur’an do not seem to have been based on empirical field data about the target readers.

To our knowledge, the only study with an explicit, empirical focus on the readers of the Qur’an translations is Elimam (2017), who empirically surveyed the preferences and expectations of non-Arab Muslim and non-Muslim readers of the Qur’an translations. He designed a questionnaire and administered it to a small sample, the majority of whom were Muslims and were accessed at mosques in the UK. Elimam found considerable diversity in readers' preferences for the Qur’an translations to the extent that they agreed on almost none of the items on the questionnaire. He concluded that with this diversity in readers' preferences, none of the existing English translations of the Qur’an “ticks all boxes for all readers.”

It should be noted that none of the studies reviewed above directly address the role of readers’ linguistic background on their reception of the Qur’an translations. Farghal and Al‐Masri’s (2000) study focused on readers’ comprehension of translations, Peachy’s study was mainly theoretical, and Elimam's study concerned the general expectations and preferences of the readers; it did not address linguistic or other social differences among the readers and how they might moderate their receptions of translations. Furthermore, all the mentioned studies are about the English translations of the Qur’an. There is thus a lack of research on the reception of the Persian translations of the Qur’an.

The current study narrows this gap by focusing on the readers' language beliefs and attitudes about the target and source languages (i.e., Farsi and Arabic). Another unique feature of the study has to do with the very context in which it is conducted. Whereas most studies on the Qur’an translations are concerned with non-Arab readers, this study zeroes in on readers whose first language is, in fact, Arabic, but since their schooling and literacy practices have been in Farsi, they have to rely on Farsi translations to understand the Qur’an. The next section elaborates on this context further.

4. Context of the Study 

Farsi, also called Persian, the language of present-day Iran, is the first language into which the Qur’an was translated (Mustapha 2009) and is rightly known as the second language of Islam (Perry 2012). This historical precedence adds to the significance of Farsi translations of the Qur’an. Currently, in Iran, many Farsi translations of the Qur’an are in circulation, which, by convention, are not published as stand-alone volumes but accompany the main Arabic text, often in a smaller, less visible font.

Though Farsi is the official language of the country, Iran is multilingual both in terms of its sociolinguistic landscape and in its official language policies as stated in the constitution. According to Ethnologue (www.ethnologue.com), there are currently 75 languages in use in Iran. Despite this language diversity and the constitutional support for multilingualism, a de facto language policy of institutional monolingualism (Meylaerts 2011) prevails in Iran. Subsequently, regardless of the native language of people in different parts of the country, schooling and literacy practices are in Farsi across Iran. As a result, literacy in the local languages is not encouraged or fostered.

The present study was carried out in Khuzestan, a multilingual province in the Southwest of Iran. Bordering on Iraq, the province prides itself in its multicultural, multilingual fabric where Farsi, Arabic, and several other language varieties are used in parallel. Similar to other provinces, Farsi is the lingua franca of the province. The province has a sizable population of Arabic speakers, to the extent that some cities in the province are predominantly Arabic-speaking.

Though teaching the Qur’an commences in primary school, Arabic, as a school subject, is not taught until junior high school. However, the diglossic nature of Arabic levels the ground for Arabic- and Farsi-speaking students because the variety of Arabic taught at schools is different from the Arabic variety used for everyday communication in Khuzestan. In addition to the institutional monolingualism policy noted above, the diglossia situation of Arabic also contributes negatively to the literacy rates among Arabic-speaking citizens across countries (Myhill 2014). The implication is that when it comes to reading the Qur’an in the context noted, readers with both Arabic and Farsi backgrounds need to rely on the Farsi translations of the Qur’an for further comprehension and understanding.

Given that language policies influence translation policies (Haddadian-Moghaddam & Meylaerts 2015; Meylaerts 2011) and that language beliefs and practices constitute an important component of language policies (Spolsky 2004), it follows that language beliefs and views about translations might be closely interrelated. Therefore, differential language backgrounds in a multilingual site offer a rich ground for studying how translations from the classic Arabic into the official language of the country might be viewed and received by readers. It is within this context of cultural diversity that this study seeks to examine how attitudes toward Arabic as the source text language and toward Farsi as the official language of the country might bear on the readers’ receptions of the Qur’an Farsi translations. More specifically, this study is intended to develop a reliable measure of the Qur’an Translation Reception and Language Attitudes (QTRLA). In so doing, it seeks to answer the following research questions:

• Are readers' receptions of translations and their attitudes about Persian and Arabic languages considered separate factors according to exploratory factor analysis?
• What are the pathways of relationships between attitudes about Arabic, attitudes about Farsi, and the reception of Qur’an translations?

5. Methods: Initial Scale Design and Participants

As we sought to design and validate the QTRLA, we began with a data-driven approach as part of an exploratory-sequential, mixed-methods design (Creswell & Plano Clark 2011). In the first phase of the study, in-depth interviews were conducted with three participants who were frequent users of Farsi translations of the Qur’an. The interviewees were asked about their attitudes towards Farsi translations of the Qur’an, how often they would refer to them while reading the Qur’an, their reasons for doing so, and their overall assessments of the usefulness of Farsi translations in easing the comprehension of the Qur’an. In addition, the interviewees were asked about their beliefs and attitudes about Arabic and Farsi languages. The interviews led to a number of tentative ideas about the readers' attitudes and uses of the Farsi translations, which were subsequently used in generating an item pool for the scale we intended to develop. We also drew on the literature as well as on our own experience with reading and using the Qur’an and its Farsi translations.

The generated items were then subjected to an initial screening. Items with more direct relevance to our research purpose were selected. To have an idea of whether the instrument is efficient in eliciting the right data on readers' reception of the Qur’an translations, it was piloted with a small sample of participants. Subsequently, a few items were dropped, and the wordings of several items were revised. After doing the needed revisions in light of the pilot study, 16 items survived. A five-point Likert type scale, ranging from strongly agree to strongly disagree, was used to quantify the respondents’ views. These items were then administered to the main study's participants.

A total of 96 college students, 69 females and 27 males, constituted the participants of the study on a voluntary basis. Their age range was between 18 and 34. Ninety three were undergraduate students, one was an M.A student, and two participants were PhD students. Concerning their first language, there were 46 Farsi speakers, 37 Arabs, one Kurd, and one Turk. The questionnaires were administered to the participants in a distance-learning higher education institution called Payame Noor (literally meaning in Farsi the message of light) in Ahvaz, the provincial capital of Khuzestan.

6. Analysis and Results

To analyze the data, we used SPSS version 18, FACTOR software, and Smart PLS3. SPSS was used for performing exploratory factor analysis (EFA) and related descriptive statistics. Based on patterns of correlations among responses to items, EFA identifies clusters of items that vary together, which are taken to be measuring the same dimension of the construct under investigation. The constructs in this study, as noted earlier, are language attitudes and the reception of the Qur’an translations. At issue here is to see whether and the extent to which items meant to measure certain aspects of the construct – called factors – statistically cluster together on the same intended factor.

Subsequently, FACTOR was used to validate the number of extracted factors found via SPSS. Finally, we investigated the factorial structure of the questionnaire using SmartPLS, which is software used for conducting Partial Least Squares Structural Equation Modelling (PLS-SEM). In comparison with the more common covariance-based SEM, PLS-SEM makes more liberal assumptions about the nature of the data used (Ravand & Baghaei 2016).

Our choice of PLS-SEM was driven by the nature of our sample and hence our data; covariance-based SEM makes stringent assumptions about the data and requires large sample sizes (Byrne 2013; Phakiti 2018). In addition, PLS-SEM is recommended when theory regarding the domain of the study is not well established (Hair Jr. et al. 2016), which is the case in this study. As noted above, EFA is used to explore the dimensions in the data in an exploratory manner. In contrast, Confirmatory Factor Analysis (CFA), carried out via SmartPLS in this study, is used to collect more solid evidence regarding factors that underlie the data. The logic behind factorial analysis is to see if theoretical expectations (i.e., deductive reasoning) are verified by statistical analysis (i.e., inductive reasoning). More specifically, we wanted to know whether and the extent to which items we designed to measure, say, attitudes about Farsi, do in fact load on the same factor. In simpler terms, both EFA and CFA are ways of knowing if items intended to tap into the same factor are statistically demonstrated to group together.

To check the normality assumptions required in parametric statistics, Kurtosis and skewness values of each item were checked, leading to the elimination of one item, which was about readers' vertical motivation (Zayed 2017), on the grounds that it had values beyond the required range of +2 to -2 (Bachman 2004). The remaining 15 Likert scale items were submitted to Principal Component Analysis (PCA).

Prior to conducting factor analysis, the factorability of the data was assessed by inspecting the correlation matrix, communalities, the KMO index of sampling adequacy, and Barlett's test of sphericity. There existed many high correlations in the correlation matrix, and the communalities of all items were high. The KMO index exceeded the minimum required. The KMO index ranges from 0 to 1, with 0.6 suggested as the minimum value for a good factor analysis (Pallant 2013). Bartlett's test of sphericity must be significant, which in this case was at the 0.001 level, attesting further to the factorability of the data.

Table 1. KMO and Bartlett's Test

To decide on the number of factors underlying the items, several criteria are recommended in the literature, including Cattell's scree plot test (retaining all factors above the elbow of the plot), Kaiser's criterion (eigenvalues higher than 1), and parallel analysis. The scree plot has been criticized for the subjectivity associated with identifying where the elbow of the plot lies (Beavers et al. 2013). According to Kaiser's criterion, there were four factors with eigenvalues above 1, which explained 48.876 percent of the variance in the data (Table 2).

Table 2. Eigenvalues and total variance explained after rotation

However, Kaiser's criterion has received criticism for its overestimation of the number of factors (Baglin 2014). Parallel analysis (PA) is a more robust alternative to the other two criteria (Pallant 2013). Table 3 illustrates the actual eigenvalues from PCA and criterion values from parallel analysis.

Table 3. Comparison of eigenvalues from PCA and criterion values from PA

To decide on the number of factors to retain based on parallel analysis, the actual eigenvalues from PCA should be higher than their corresponding criterion values from parallel analysis. As Table 3 demonstrates, for the first three factors, the actual eigenvalues are larger than the criterion values obtained in parallel analysis. For the fourth factor, however, the eigenvalue (1.117) is smaller than the criterion value (1.235), and hence, it was dropped from the final solution.

As noted earlier, attitudes towards Farsi, Arabic, and the Qur’an translations were measured on a Likert scale. Though commonly practiced, factor-analyzing Likert scale data using SPSS has been criticized because SPSS uses Pearson correlation, which results in an underestimation of the dimensions underlying the data (Baglin 2014). It has been argued that Pearson correlation is not compatible with the nature of Likert scale data, which are ordinal. Alternatively, Baglin recommends using FACTOR software, which incorporates built-in assumptions congruent with Likert data. To overcome this limitation, we used FACTOR software (Lorenzo-Seva & Ferrando 2006) to determine the number of factors to retain.

Table 4. FACTOR output: eigenvalues and mean of random variance explained

  Table 4 is the output form FACTOR software, with the second column showing the proportion of variance explained by each factor in the data. “FACTOR compares the mean or the 95th percentile of the factor’s percentage of common variance explained from the randomly permutated data to the observed explained common variance from the sample” (Baglin 2014). For each factor, if the variance explained in the real data exceeds the variance explained from the randomly permutated data, the factor should be retained. Accordingly, the first three f actors must be retained and the remaining five must be excluded from the solution. The FACTOR solution confirms the final solution reached via Parallel Analysis, which together provide good grounds in support of a three-dimension solution.

Regarding the loadings of individual items on factors, as Table 5 illustrates, the first five items loaded on the first factor, the next five items on the second and the last three items loaded on the third factor. Now that we knew which items flock together, the next step was to figure out which items tap into the translation reception dimension of the scale and which items were targeting the language attitude construct.

Table 5.  Descriptive statistics and factor loadings of items of the scale

Factors were named on the basis of the meaning of items that clustered together. Items under the first component touched on participants' views about the Qur’an translations.  Accordingly, this factor was named attitudes about Translation Quality (TR) factor. Items clustering on the second component all in one way or another carried some favourable attitude towards Farsi language. Thus, this component was termed ProFarsi factor. Similarly, items loading on the third factor all pertained to some positive attitude or belief towards Arabic, hence, named ProArabic Factor. Overall, then the answer to our first research question is that the factorial structure of the scale comprises of a translation reception judgment dimension and two attitudinal dimensions about the source and target languages of translations.

To answer the second research question, which concerns the paths of relationships among the three factors underlying the scale, two models were postulated and statistically examined using PLS-SEM. In the first model, figure 1, translation reception is taken as the endogenous variable and the other two factors as two exogenous variables (In PLS-SEM, these are factors with no arrows pointing towards them). As can be seen, the two factors explain one-fourth of variation in the outcome variable, with the Arabic Endorsement dimension negatively correlated with receptions of Farsi translations of the Qur’an. To see whether the negative contribution of the Arabic Endorsement factor to the outcome variable was direct or mediated, the second model was postulated (figure 2). In other words, we wanted to see whether the unfavourable reception of translations is about the Qur’an translations per se or whether such it has to do with attitudes towards Farsi language.

Figure 1. Relationship between Factors with Two Exogenous Factors

As can be seen in figure 2, almost all the contribution of the Arabic Endorsement factor is transferred to the path between Arabic Endorsement and Farsi Endorsement factor (i.e., the path coefficient from ProArabi to Translation quality is 0.392 in figure 1 and it is 0.031 in figure 2). This change in the path coefficients indicates that attitudes towards Farsi explain the negative attitudes towards Farsi translations of the Qur’an.

Figure 2. The model with Indirect Path Relationships

In summary, in response to the second research question, PLS-SEM analysis revealed that attitudes towards Farsi mediate the relationship between attitudes towards Arabic and receptions of translation. In other words, it is mainly language attitudes that determine perceptions of translations not the translations per se.

7. Conclusions

This study addressed two related research questions. First, we examined the dimensionality of a scale developed for assessing attitudes towards source and target languages in Qur’an translations and second, we explored the direction of relationships among the three dimensions.

To reiterate, the answer to the research question is in the affirmative, that is, the three variables of attitudes about Farsi, attitudes about Arabic, and translation reception constitute distinct dimensions in the questionnaire. Regarding the second research question, which concerns the direction of relationship among the three noted factors, it seems that negative attitudes towards translation is mediated through the attitudes towards Farsi, That is, it is the negative attitudes toward the language of translation NOT the translations themselves which explain the negative path coefficient between ProArabic factor and perceived quality of Qur’an translations (i.e., reception).

Now what do these statistical analysis mean to us? Findings related to the first research question mean that in assessing readers’ reception of the Farsi translations of the Qur’an, we should keep in mind the linguistic fabric of the context where translations are used. Accordingly, quality of the Qur’an translations is not a universal or national characteristics. It is likely that people across different social and cultural settings in Iran might have different or even opposing receptions of the same translation.

 Secondly, when translations of the Qur’an are not received favourably, it does not necessarily mean the there are issues with the quality of the translations. Unfavourable attitudes towards translation might be due to readers’ attitudes towards the target language – Farsi in this case – in general not as it is used in translations. In such cases, the phenomenon of translation reception gets entangled with issues in language policy. That is, for policy makers who seek to improve citizens’ religious literacy, they should be mindful of the powerful influence of language beliefs (see Spolsky 2004).

Generally speaking, these findings suggest that translations as social artefacts should be viewed and assessed within a broad socio-cultural perspective incorporating the interests and efforts of multiple agents (Xu & Yu 2019). Put it another way, regardless of the technical quality of a translation from the expert perspective, a translation may fail to flourish due to linguistic and socio-cultural considerations that are external to the translation per se. Accordingly, the findings of the current study underline the central role that is assigned to the reader in translation studies. More specifically, the findings underscore the role of language attitudes in shaping or mediating the reception of translations. Given the rather clear pattern of perceptions that emerged from this study among Farsi and Arab readers of the Qur’an translations, the results of this study run counter to Elimam (2017). In his study, Elimam came across a rather messy and unpredictable picture of readers' preferences. Perhaps, this discrepancy has to do with the profile of participants in the two studies; in Elimam's case, the participants were more heterogeneous in terms of nationality, reading experience, and cultural capital. However, in this study, the research participants had more in common in terms of linguistic and cultural background.

The findings of this study also point to the usefulness of research tools developed in neighbouring disciplines in empirically investigating translations in general and translations of the Qur’an in particular. The broad implication is that we would benefit from crossing insulated disciplinary borders to seek insights from the repertoire of human knowledge in its wide ecology (Cronin 2017). Another implication of the findings is that given the centrality of the Qur’an in Muslims' life, it is important to research the actual readers of the Qur’an translations, their preferences (Elimam 2017; Peachy 2013) and their language ideologies. To do so, having the right measurement tools to assess users' attitudes towards the source and target languages is essential to gathering insights into the readers' language beliefs and ideologies. In turn, such an awareness of the reader characteristics highlights the inadequacy of the dominant mentalistic subjective approaches (House 2014) to assessing the translations of the Qur’an.

8. Limitations and Further Research

As with any research, the present study had its own share of limitations. First, the rather limited sample of participants may compromise the generalizability of the psychometric properties of the designed scale. Therefore, a larger sample size allowing the use of covariance-based SEM would be a more robust test for confirming the factor structure of the scale. Secondly, the sample was one of convenience not a random sample, further alerting us to approach the findings with caution. Future studies free of the noted limitations would yield more valid scales of translation reception of sacred texts. In addition, had we measured the participants' literacy in Arabic and Farsi, we could have built a more fine-grained scale sensitive to participants' level of literacy. One issue complicating research on the reading of a sacred text in the source or in the target language is reader motivation. Whereas in ordinary reading, motivation is usually horizontal, readers of the Qur’an and those of other sacred texts might be driven by vertical motivation: reading for divine reward (Zayed 2017). As such, further inquiry into readers' motivational profiles would boost the construct validity of the measurement of translation reception (Messick 1996). 

Translators may sometimes assume a homogenized readership with similar tastes or expectations. In keeping with Elimam (2017), the current study demonstrates that readers across cultural and linguistic communities have different perceptions and expectations, indicating that translators of Qur’an should be informed of the local needs of the target readership of the Qur’an. Perhaps in a diglossic context like that of the current study, a translation intended for the non-Arabic speaking readership may feature too much redundancy, inspiring boredom and disinterest in the reader. Being aware of divergent readers’ backgrounds and preferences, the translator must make a conscious attempt to adapt the text to “linguistic universe of its intended readers” (Evers et al. 2010, cited in Elimam 2017). It is likely that for the Arabic speaking readers, whose written literacy is in Farsi but are at the same time fluent speakers of an Arabic vernacular, the problem of untranslatability of the Qur’an (Abdul-raof 2001) becomes more salient. One possible strategy is for the translator to view the translation more as a practice in translanguaging (Sato 2019) where the boundaries between the source and target texts are blurred and the process is rather seen as an opportunity to make optimal use of all the available semiotic repertoire to reinforce communication. 

It should also be noted that attitudes and beliefs about languages stem from wider social, historical, and political considerations, adding further complexity to the study of the reception of the Qur’an translations. Therefore, for the translator to adapt the translation of the Qur’an to the expectations and preferences of the readers, one must go deeper into the socio-historical profile of the target reader (Chan 2016).

This study was exploratory in the sense that it highlighted how readers’ semiotic and linguistic repertoires bear on their perceptions of Qur’an translation. It remains to be seen what causes such differential reaction. It might be that such different stances have to do with factors external or internal to the translated text. This however awaits further research.  Ethnographic studies on how, how often, why, and under what circumstances the readers do attend to the Qur’an's translation will furnish us with more nuanced understanding of the reception of the Qur’an translations.