Serat Yusup is a Javanese literary work that tells the story of Joseph, the son of Jacob, composed in the Javanese poetic form known as macapat. It was originally written in Javanese script and later also in pegon, an Arabic script adapted for writing Javanese. Pigeaud suggested that Serat Yusup dates back to the seventeenth century (Pigeaud 1967, 1:217), yet this should apply only to the versions written in Javanese script. The pegon versions likely emerged about a century later, coinciding with the period of the two earliest translations of the Quran being discussed.

Pigeaud further suggested that the Javanese Serat Yusup was shaped by the Malay model Hikayat Yusuf, with which it interacted alongside the original Arabic texts (Pigeaud 1967, 1:217). According to Majid Daneshgar, who closely examined the oldest version of Hikayat Yusuf in Erpenius’ collection at Cambridge University Library, this text is rooted in a Persian tafsir (Daneshgar 2024). This connection places the Javanese Serat Yusup within the tafsir tradition. In this study, I aim to further extend this lineage by illustrating how the Javanese Serat Yusup has also played a role in the production of Quran translations. To support my argument, I present three instances where Serat Yusup has left its mark on interlinear translations of the Quran. These cases are derived from two manuscripts: A.54, currently housed in the National Library of the Republic of Indonesia, and Or.2097, which is preserved in Leiden University Library. Both manuscripts were written in the 18th century.

The first instance is the spelling of Zulaykha’s name. Zulaykha (spelled زليخا in Arabic), the wife of Potiphar, is a character featured in a Quranic story, but her name is never mentioned. In the Quran, she is only referred to as imraʾat al-ʿAzīz (the wife of al-ʿAzīz). In other words, the Quran doesn’t tell its readers how to spell her name. In order to learn that, Javanese Muslims—who are not native Arabic speakers—have to rely on derivative texts, such as the Serat Yusup. In these Javanese texts, her name is spelled as Jaleka in Javanese script or Zalika/Zaleka in pegon.

Figure 2. MS Dd.5.37 of Cambridge University Library, a 17th-century Malay Hikayat Yusuf, f. 21v. The spelling of Zulaykha’s name.

Figure 3.MS  AW 97 of the National Library of the Republic of Indonesia, undated, original place unknown. The spelling of Zulaykha’s name in a pegon version of Serat Yusup.

Figure 4. MS A.54b of the National Library of the Republic of Indonesia, an 18th-century Bantenese interlinear Quran translation. The spelling of Zulaykha’s name.

Figure 5. MS Or.2097 of the Leiden University Library, an 18th-century Central Javanese interlinear Quran translation. The spelling of Zulaykha’s name. 

Interestingly, this spelling is also found in both of the oldest Quran translation manuscripts (Figure 4 and 5). This is a first signal for the connection between the Serat Yusup and the Quran translations.

Strengthening that signal is the second instance which concerns the translation of the word al-dhiʾb. This word—translated as “wolf” in modern dictionaries—refers to the animal that is said to have killed the Prophet Yusuf while he was playing with his brothers. The Quran again does not provide any further details about this animal. And again, Javanese Muslims have to rely on derivative texts for clarification. In MS. Dd.5.37, the Cambridge manuscript of the oldest Malay Hikayat Yusuf, the animal is always identified as harimau (tiger); in all the Javanese Serat Yusup manuscripts I’ve read, it is presented as macan (tiger), and interestingly, this macan presentation is also used in both Quran translations, A.54 and Or.2097. While it is important to investigate whether the Persian tafsir provides a Persian equivalent for ‘tiger’ in this context, that lies beyond my expertise. Nonetheless, this example further highlights the relationship between Serat Yusup and the interlinear Quran translations.

The third instance, however, shows a more complex relation between Serat Yusup and the interlinear Quran translations. It highlights something more subtle than the previous two, i.e. the levels of Javanese language. Broadly speaking, Javanese has two speech levels: the low one (ngoko) and the high, refined one (krama). The krama level is used when addressing someone respected, be it due to their age, social/political/religious status, or other factors. The shift in speech levels is usually done through word choices or adding specific elements to the root words. All types of words—nouns, verbs, adjectives, etc.—may undergo this transformation. What is particularly relevant here is the level of the word used to describe Zulaykha’s actions. When news of her affair with Joseph spread through the city and became the talk of the town, Zulaykha became aware of it and sought to clarify the situation. The story continues with the well-known scene where women, mesmerized by Joseph’s beauty, unwittingly cut their fingers. However, I want to focus on the moment when Zulaykha became aware of the gossip about herself and Yusuf.

In The scene is depicted in Q. 12:31 where, the Quran uses the word samiʿa (fa-lammā samiʿat, “so when she heard”), literally meaning “to hear,” to indicate Zulaykha’s awareness of the gossip. Iand it is this literal meaning meaning that is reflected in the two 18th-century Javanese interlinear translations of the Quran: amiharsa in A.54 and midhanget in Or.2097. Interestin—gly, both terms are considered refined in Javanese. It is also important to point outintriguing that the very verse Q. 12:31 is never translated literally in the Serat Yusup, and Zulaykha is accordingly never associated with the verb for hearing. Where then did the Quran translators get amiharsa and midhanget from? While I haven’t come across the word midhanget in the Serat Yusup, I often find amiharsa, which is associated with highly respected figures (e.g. God [Sang Yang Widi], the king [Sribupati and Srinalendra], the Prophet Jacob, and the Prophet Joseph himself); Zulaykha is never linked to this term. Other words for “hearing” used in Serat Yusup are a refined word pireng attributed to the king (ratu), and a lower one angrungu attributed to figures of lower social status.

By attributing amiharsa and midhanget to Zulaykha, the two interlinear Quran translations seem to place Zulaykha in a position equal or at least close to God, kings, and prophets. This is particularly intriguing, as it suggests an elevation in her rank—in Serat Yusup Instead, she is often only depicted as “speaking,” “dreaming,” and “seeing,” using with words of the low level: sumaur/angucap, angimpi, and tumingal, suggesting her low position. . Given this, the Quran’s use of amiharsa and midhanget in relation to Zulaykha is unique, as it conveys a different nuance from what we see in the Serat Yusup. This raises theone may raise a question of whether the Quran translators intend to show more respect for Zulaykha than the composers of Serat Yusup did. Put theThis question aside, this example illustrates how the interlinear Quran translations not only dreaw from and get were influenced by Serat Yusup but also engaged with it critically.

Through these three instances—highlighting spelling, translation, and more nuanced linguistic features—I aim to demonstrate how the Javanese Serat Yusup has contributed to the production of a Quran translation. In addition to Serat Yusup, there are numerous other literary works to investigate, such as Serat Enoh and Sūrat Nūḥ (Q. 71), as well as Hikayat Israʾ Miʿrāj and Sūrat al-Isrāʾ (Q. 17), to name just two. Examining the connections between these serats and their corresponding sūrat counterparts not only acknowledges the serats’ significance within the tafsir tradition but also, as I mentioned in my earlier post, gives the Javanese Quran translation its due as part of the Javanese literary tradition.

Sources

MS. PNRI A.54a-e, The National Library of the Republic of Indonesia collection.

MS. Or. 2097, Leiden University Library.

Daneshgar, Majid. 2024. “Translating Persian Tafsir in Aceh: The Oldest Malay ‘Story of Joseph’ at Cambridge University Library.” Cambridge University Library Special Collections (blog). 14 Oktober 2024. https://specialcollections-blog.lib.cam.ac.uk/?p=26005.

Pigeaud, Theodore G. Th. 1967. Literature of Java. Vol. 1. The Hague: Nijhoff.

I will begin this blogpost by answering the question posed in the title. Yes, and thus the translations of the Quran into vernacular languages need to be treated as part of their own local literary traditions. With particular regard to translations in interlinear form, the limited space available to translators has made them more creative in playing with subtleties not found in the Quran’s original language, Arabic.

To make my point clearer, I will take an excerpt from two Javanese interlinear Quran translations from the eighteenth century: a manuscript from Banten, MS A.54 now kept in the National Library of the Republic of Indonesia, and a manuscript from Semarang, MS Or.2097 of Leiden University Library. To be more specific, I will discuss a group of words that are derived from the stem r-w-d (which generally means ‘to desire’) from Q. 12 (Sūrat Yūsuf).

The verses I pick here describe scenes when Zulaykha tried to seduce Yūsuf (v. 23); when  news of the affair  then spread all over the city and the women were gossiping about it (v. 30); and when Zulaykha proved to the women that Yūsuf’s beauty is so very exceptional that everyone will be captivated by it (v. 32). The last verse (v. 51) is depicting a scene where the king (malik) was helped by Yūsuf and, upon Yūsuf’s request, asked Zulaykha and the women of the town to clarify what happened between them and Yūsuf.

In these verses, there is one word in the form of fiʿl muḍāriʿ (present and future tense), turāwidu (v. 30), and three other words in the form of fiʿl māḍī (past tense), i.e., rāwadat (v. 23), rāwadtu (vv. 32 and 51), and rāwadtunna (v. 51). The actor (fāʿil) of the first three words (turāwidu, rāwadat, and rāwadtu) is the wife of Yūsuf’s master identified in the books of tafsīr as Zulaykha, while that of the fourth word (rāwadtunna) is given in female plural form referring to Zulaykha and women of the town.

In translating the above r-w-d derivation, the two Javanese translators used different words. The Banten manuscript (PNRI A.54) uses three words (akarĕp, arĕp, and anĕkani), while the Semarang manuscript (Leiden Or.2097) uses only two (angarĕpi and adhĕmĕn). Angarĕpi is used in the Semarang manuscript to translate all derivations of r-w-d except for the first rāwadtu in verse 32 which it translates as adhĕmĕn. It is interesting that the same rāwadtu in this verse is also treated differently by the Banten manuscript; it is the only word translated with the word arĕp. The second rāwadtu in verse 51 is translated with anĕkani, the same word it uses to translate rāwadtunna in the very verse. The words rāwadat and turāwidu were both translated as akarĕp.

What is the difference between these Javanese words employed in the translations? Why did the two Javanese translators use the same Javanese words to translate different Arabic words (rāwadat and turāwidu) on the one hand, and different Javanese words for the same Arabic word (the two rāwadtus) on the other? Why did the Bantenese translator introduce a totally new word for the two r-w-d derivations in verse 51, while the Semarang translator went back to the words that he/she used in the beginning (verse 23 and 30)?

In order to answer these questions, one cannot merely rely on one’s knowledge of Arabic. The distribution of verb actors (fāʿil) in Arabic I mentioned above, for example, shows a pattern different from that of the translation. The philological analysis of some works of tafsīr specializing in linguistics also cannot give us clear answers—all more so because we do not know whether such tafsīrs were accessible to the two Javanese translators. What helps us more is linguistic and literary analysis of the Javanese.

In two aggregator websites for Javanese dictionaries (www.sealang.net and www.sastra.org), we find that arĕp (also the root of akarĕp and angarĕpi) comes from the Old Javanese language, harĕp. Besides ‘to desire something/someone’ which fits the topic of desire addressed here, arĕp also has the meaning of ‘to want, to wish’ and ‘in front of, fore part’, the latter being quite far from desire. Tĕka also comes from Old Javanese and also has a meaning quite distant from desire, that is ‘to come to’. The addition, the prefix a- and suffix -i for angarĕpi and anĕkani aim to make the verbs transitive. Although not listed in the sealang website as a derivative form of harĕp, akarĕp is recorded in three other sources that I consulted with a meaning not far from its root: ‘to want, wish, desire for’.

The nuances of arĕp, angarĕpi, anĕkani, and akarĕp are quite different from adhĕmĕn. This last word also comes from Old Javanese and is included in the four sources above. Compared to the previous words, dhĕmĕn has less variations in meaning, either ‘to like something very much’ or ‘to love, to be pleased with someone/something’ in association to lust.

What can we make of these explanations? I suggest this has to do with the speech contexts of each word. The first two words (rāwadat and turāwidu) appear when the Quran speaks to its readers and women of the town speak to other women. In these two conversations, Zulaykha was referred to in the third person. For Javanese, in this kind of context, the choice of words to attribute to the third person (whether to use formal, polite words or the informal, casual ones) depends on the relation between the third person and both the speaker and interlocutor. In this case, the relation between Zulaykha and the speakers (Quran and the women) and interlocutors (Quran’s readers and other women) is equal, so the Javanese Quran translators did not have to use a formal or polite word.

A different feeling emerges in the first appearance of rāwadtu (verse 32). In this verse, Zulaykha is speaking in the first person to the women of the town. Even though there is no clear explanation regarding the identity of these interlocutory women, the mention of Zulaykha as the wife of al-ʿAzīz (lit. a respected person) in the Quran is enough to give Zulaykha a position higher than the women. Moreover, in this verse Zulaykha has succeeded in convincing these women of Yūsuf’s extraordinary beauty. In such a context, Zulaykha finds a kind of safe space that allows her to speak more openly about what she feels for Yūsuf. The safe space was then lost when Zulaykha and the women of the town had to go before the king (malik) and explain what they had done to Yūsuf (verse 51).

The identification of these three situations may help explain the choice of words to translate derivatives of the Arabic r-w-d. In the first neutral situation, the Javanese translators had no reason to speak very directly about desire, so they chose akarĕp and angarĕpi which hint quite clearly at desire but not as straightforwardly as adhĕmĕn. The safe space and power relations in the second context allow Zulaykha to use the direct word adhĕmĕn to express her feelings. However, she had to quickly change course and speak indirectly when discussing this matter with the king. According to the Banten translator (A.54), Zulaykha even had to resort to innuendo by using a word completely unrelated to desire, anĕkani.

Could we reach this understanding by reading tafsir works? I would rather suggest delving into Javanese literary works to identify and further comprehend its subtleties, as shown above. To put it another way, one should give the Javanese Quran translation its due as part of the Javanese literary tradition.

MS. PNRI A.54a-e, The National Library of the Republic of Indonesia collection.

MS. Or. 2097, Leiden University Library.

www.sastra.org

www.sealang.net