Microsoft Translation (MsT) and Human Translation (HT) on Translating Thirukural into English Language

Microsoft Translation (MsT) and Human Translation (HT) on Translating Thirukural into English Language

Gomathi Batumalai¹, Manimangai Mani², Veeramohan Veeraputhran³

^1,3Department of Foreign Languages, Faculty of Modern Language and Communication, 43400 Universiti Putra Malaysia, Serdang, Selangor

²Department of English Language, Faculty of Modern Language and Communication, 43400 Universiti Putra Malaysia, Serdang, Selangor

DOI: https://dx.doi.org/10.47772/IJRISS.2024.803444S

Received: 17 November 2024; Received: 30 November 2024; Accepted: 03 December 2024; Published: 31 December 2024

ABSTRACT

Thirukural is an ancient literature that is found in Tamil. This is a famous piece of literature that people worldwide use as guidance. Thirukural has been translated into more than 82 languages, its English renditions by various translators serve as a crucial medium for global audience. This study mainly discusses the implementation of Microsoft Translation (MsT) and Human Translation (HT) in translating Thirukural into English. Aim of the paper is to identify the translation techniques for Microsoft Translation (MsT) and Human Translation (HT). This study also analyzed the accuracy of Microsoft Translation (MsT) and Human Translation (HT). Peter Newmark’s Translation Theory (1988) was used to conduct the study. This translation theory explains the translation procedures and methods for data analyses. This is a qualitative study. All the data was collected with text analyses and survey methods. The data were discussed briefly in the findings and discussion. This study found that translating literature with Artificial Intelligence and Human Intelligence brings many changes in Targeted Text (TT).

Keywords: Translation, Thirukural, G.U. Pope, Microsoft Translation, Human Translation

INTRODUCTION

Translation technology refers to special computer tools that help translate written words from one language to another. Just like how technology can help people do their work faster and better. These translation tools also make the translation easier and more accurate. When we talk about translation technology, we mean different types of helpful tools. Some examples are programs that manage special words, machines that do translations, virtual helpers for interpreting, and even systems that change spoken words into written ones. In the past, people used to do translations by hand. They would read big dictionaries made of paper and use their best judgment to change words from one language into another. But now, with the help of technology, things have become much smoother.

Over the last ten years, translation technology has improved a lot. We can now use small devices like apps, headphones, and pocket-sized gadgets to carry translation tools with us wherever we go. This is great because even if the internet isn’t very strong, we can still understand people from different countries. In the future, we might find even more ways to use this technology. For example, there’s a lot of interest in combining automatic lip-reading with translation. People are also working hard on making translation tools better for understanding different ways of speaking in various regions. Some of this work is already making devices and services better, but in the future, we might have even better tools that can handle different ways of talking. It’s cool to see how technology is part of our everyday lives. In today’s smart world, technology is like the strong backbone of humanity. It’s important in every area, especially in things like literature.

Literature is the fancy term for all sorts of writing like stories, poems, and plays. Thanks to translation, literature from different parts of the world can reach people all over. Translators and writers are the heroes who make this possible. When we talk about translating literature, we’re talking about something special. It’s not just about changing words; it’s about keeping the feelings, style, and meaning of the writing intact. This can be a bit tricky because sometimes a direct translation doesn’t capture everything. Our talented translators don’t simply change words from one language to another. They rewrite the text to make sure it sounds just right and still means the same thing. They think about the author’s unique way of writing, the words they use, and even things like jokes and rhymes. It’s not an easy job, but it’s important to keep the heart of the original writing.

Literary translation is even more special because it’s not just about meaning. It’s also about making the writing beautiful and relatable. This means understanding the special cultural things and getting the humor, emotions, and other delicate parts of a work just right. Sometimes, sticking strictly to the words might not capture all the complexity and importance of the original work. This is why literary translation can involve some changes from the usual rules.

Because of this, our skilled translators create a version that’s not exactly like the original but still similarly captures the meaning. How they do this depends on how they understand the writing. It’s a good idea to translate into a language you’re good at writing in because it’s almost like writing a new piece. We’re skilled writers too, and this makes us great at translating. We don’t just focus on meaning; we also think about the author’s style, the words they use, and even how things sound when spoken. It’s quite a challenge, especially when things like rhymes, puns, and metaphors come into play.

Thiruvalluvar, one of the greatest poets ever, wrote an ancient classic in Tamil called Thirukural about 2,000 years ago. This amazing work talks about important values and good behavior that apply to everyone, whether they’re rich or poor, regular people, or even rulers. It’s relevant no matter where you live. This study aims to explore the nuances of translating ancient Tamil literature using MsT and HT, focusing on the preservation of cultural and semantic integrity. With AI-powered tools increasingly used in translation, there is a need to critically assess their effectiveness in conveying the depth of classical literature. This research highlights the challenges and advantages of machine translation for ancient texts, providing insights for educators, linguists, and AI developers.

Thirukural

Thirukural is an ancient literature written by Tamil scholar Thiruvalluvar. Thirukural was founded in BC 16 Century (Staff, 2009). The most well-known of the Patiren-kirkkanakku in Tamil literature, Tirukural, often spelled Tirukural or Thirukural, also known as Kural, is a work that has had a significant impact on Tamil culture and daily life. Although some academics place the authorship of this work earlier in the first century BC, they generally credit the poet Tiruvalluvar, who is considered to have lived in India in the sixth century.

The Tirukural has been compared to the great literature of the world’s major religions because of its pragmatism, aphoristic observations on daily life, and timeless outlook. The three main parts in Thirukural, Aram (virtue), porul (government and society), and inbam (love) include 133 sections of ten couplets each. The first section, called virtue (aram), spans chapters 1 through 38. The moral principles of an individual and philosophy are mostly explained in this section. Wealth is the second component (porul). Chapters 39 to 108 in this section describe socioeconomic ideals, politics, and society. The final section, which speaks about love (inbam), spans chapters 109 to 133.

Thiruvalluvar wrote these 1333 couplets as a service to the general populace. This Thirukural is widely used by people all over the world in different languages. Translation of Thirukural has brought the beauty of Tamil literature worldwide.

LITERATURE REVIEW

There are various studies carried out regarding the translation of Thirukural. Scholars from different perspectives have carried out the analysis of the translation of Thirukural.

The first study is ‘Critical Discourse Analysis of a Native and Non-Native Translator of Thirukural’ (Chris Tina, 2021). This study mainly compares the translation of Thirukural on critical discourse by native translator Sri Venkatesa Subramaniam Aiyar and the non-native translator G.U.Pope. This study focused on the second section of the Thirukural, Porul/Wealth. This study found that G.U.Pope translated the Thirukural in poetic form while sticking to the line division-like source text. However, he is a complex word which hard to understand. On the other hand, Aiyar used prose format, not sticking to line division; however, he used simple words to make readers understand the translator in a single read. Both translators made the translations based on their knowledge and proficiency. This study is based on Human Translation (HT).

The next study is ‘Thirukural Translation of GU Pope and Rajaji: A Comparative Study’ (Jaya et al., 2018). This study compares G.U.Pope’s and Rajaji’s English translations of Thirukural. This article proves that GU Pope made word-for-word translations and found a need for more clarity in the meaning. However, Rajaji made sense-to-sense translation. This translation was very plain. Rajaji’s translation found some need for more beauty in Thirukural. This study is also based on Human Translation (HT).

Other than that, the study ‘Building Discourse Parser for Thirukural’ (Anita & Subakakitha, 2019). This study is based on constructing the Thirukural discourse parser, which finds its semantic relations in hidden meaning. Various technological applications and software were used to collect data for this study. This study is based on Machine Translation. This study proves that translation brought damage to the Thirukural.

Most of the previous studies used human translation in translating Thirukural. On the other hand, only a few translations were made by machine translation. All the studies above study the contrast in the message’s meaning in translation; however, this study is translated with Microsoft translation software to translate Thirukural into English.

METHODOLOGY AND SCOPE

Qualitative research involves collecting and analyzing non-numerical data, such as text, video, or audio, to understand ideas, opinions, or experiences fully. It can unearth complex circumstance information or generate original research ideas. The opposite of qualitative research is quantitative research, which gathers and examines numerical data for statistical analysis. Qualitative research is often used in the humanities and social sciences in disciplines like anthropology, sociology, education, the health sciences, history, etc. This is a qualitative study. All the data were collected via text analysis and surveys. The collected data were explained well in the findings and discussion.

This study used Thirukural as a source of data. From 133 chapters, this study focused on the chapters of Love from Ibnathupal. Each couplet from Chapter 109 to Chapter 133 was used to analyze the data. These 24 couplets were chosen to collect the data for this study. Each couplet from all chapters was inserted in Microsoft Translation (MsT) and Human Translation (HT) to get the study’s findings.

Microsoft Translation (MsT) is a software that people use to translate a word or phrase into another language in Microsoft software. We can access This free software in Microsoft software such as Microsoft Word. This is a famous application too. G.U.Pope’s English translation of Thirukural has been used as the study’s Human Translation (HT) data. G.U.Pope is a famous Tamil scholar who spent 40 years translating Tamil literature, such as Thiruvacagam and Naladiyar, in Tamil Nadu. His translation, made in 1988, has been used for the study. Both translations were used to collect data for this study. All the data were analyzed in findings and discussions.

Peter Newmark Translation Theory (1988)

Peter Newmark’s Translation Theory (1988) is pivotal in the article’s analysis of Microsoft Translation (MsT) and Human Translation (HT). His framework distinguishes between semantic translation, which prioritizes preserving the original text’s meaning and cultural nuances, and communicative translation, which emphasizes readability and accessibility for the target audience. MsT predominantly employs a literal or word-for-word approach, a subset of semantic translation, ensuring linguistic accuracy but often failing to capture Thirukural’s cultural and metaphorical depth, leading to contextual distortions. Conversely, HT, exemplified by G.U. Pope’s translation, blends semantic and communicative methods, preserving the cultural richness and poetic nuances of Thirukural while ensuring comprehension for English readers. However, Pope’s use of complex vocabulary occasionally hinders readability. Newmark’s theory underscores the trade-offs between fidelity to the source and audience engagement, revealing that while MsT offers efficiency, HT’s nuanced approach is better suited for translating intricate literary texts like Thirukural.

Diagram 1: Newmark V-Diagram (1988)

FINDINGS AND DISCUSSION

Contrast in Translation by MsT and HT

Based on the data collection via MsT and HT, the data are shown in the tables below. The translation with two methods shows contrast in the process of the translation.

Table 1: Microsoft Translation (MsT) of Thirukural

Table 1 displays the Thirukural translation created using Microsoft Translation (MsT). This translation process utilized a laptop and Microsoft Word document software. This freely available software supports translation across several languages. However, the artificial intelligence (AI) employed in this translation procedure exhibits limited accuracy. Within this research, Table 1 showcases the translation of the initial 24 couplets from Thirukural’s Love/Inbam section, spanning chapters 109 to 133. Examination of the table brings to light numerous disparities in the MsT rendition of Thirukural. The portions highlighted within the table denote the errors identified in the translation. These inaccuracies significantly alter the intended meaning. An imperative aspect to recognize is that MsT lacks a predefined target audience, resulting in a translation process devoid of focused tailoring. Notably, the Tamil language boasts the unique trait of encompassing words with multifaceted meanings. Regrettably, this complexity was misconstrued in the translation, leading to misinterpretation and a disconnect from the appropriate context. The MST diverges from a translation that accurately captures the intended meanings.

In simpler terms, the data presented in the research table illustrates the shortcomings within Microsoft’s Thirukural translation facilitated by MsT. The translation’s precision is compromised, and its contextual understanding is limited. As a consequence, incorrect meanings are attributed to various segments of Thirukural. The AI underlying this translation neglects the specific readership, thereby resulting in these errors. Furthermore, the intricacies of the Tamil language’s lexicon, encompassing words with diverse interpretations, were not adequately accounted for, further underscoring the disparities between MsT and accurate translations.

Table 2: Human Translation (HT) of Thirukural

As presented in Table 2 above, the Human Translation (HT) of Thirukural was undertaken by George Uglow Pope, renowned as GU Pope. A distinguished Tamil scholar and academic, GU Pope resided in Tamil Nadu for a span of 40 years. His notable accomplishment includes translating the entirety of the Thirukural text into the English language. Within Table 2, the translation of the initial couplets from chapters 109 to 133 is exhibited. These initial couplets, selected from each chapter, were specifically chosen for analysis. The highlighted segments within these translated couplets accentuate the contrasts evident in the HT. This data-driven presentation effectively underscores the considerable disparities between Human Translation and Machine-based Translation (MsT). GU Pope’s rendition of Thirukural carried out with his profound proficiency in both Tamil and English, stands in stark contrast to automated translations like MsT.

GU Pope’s illustrious background as a Tamil scholar equipped him with a deep-rooted understanding of the intricacies embedded within the Thirukural’s verses. His translation efforts were not mere word replacements, but a testament to his adeptness in grasping the essence and cultural significance of each couplet. This proficiency enabled him to achieve a translation that is faithful to the original while effectively conveying the intended meanings in the English language.

Table 2 serves as a validation of the superiority of Human Translation, particularly when handled by experts like GU Pope. The highlighted divergences within the translated couplets are a reflection of the nuanced contextual awareness that a human translator brings to the task. GU Pope’s rendition, born out of his linguistic and cultural expertise, outshines automated translations by accurately capturing the essence of Thirukural’s teachings.

In conclusion, Table 2 stands as a tribute to the meticulous efforts of GU Pope, a seasoned scholar who bridged the gap between Thirukural’s Tamil origins and its English-speaking audience. His translation work, characterized by an in-depth understanding of both languages, resulted in a translation that goes beyond mere words, encapsulating the profound wisdom and cultural nuances of Thirukural. This contrast between HT and MsT reaffirms the indispensability of human interpretation, especially in the realm of translating intricate literary masterpieces.

Translation Contrast in MsT and HT

Table 3: Accuracy in MsT and HT

Table 3 gives us a clear picture of how well Machine-based Translation (MsT) and Human Translation (HT) worked in turning Thirukural into English. This table helps us see that MsT and HT are quite different when it comes to translating Thirukural. The differences start with the words used. MsT can sometimes pick different words, which can change the meaning. Machines follow patterns, but they can’t understand the full meaning like humans can. So, when translating a deep work like Thirukural, MsT can make mistakes. Another thing to think about is the context of the words. Thirukural has cultural and metaphorical stuff that needs understanding. People get this, but machines have a hard time. This means that MsT might not catch all the special meanings in Thirukural.

Grammar, the rules of how words fit together, is also tricky. Thirukural’s words are like an art, and machines can’t do it as well. MsT might sound okay, but it won’t sound as nice as Thirukural. So, MsT and HT both have problems when translating Thirukural, but MsT has bigger problems. Thirukural is important, and its deep ideas need careful translation. Machines can’t do this well. Thirukural is not just about words, it’s about feelings and culture too. This is where people are better. They know the language, and the culture, and can capture the heart of Thirukural. They can see what’s hidden in the words and make sure the translation feels right. Cultural things in Thirukural can be small but important. People who translate know these things and can make the translation fit. They know how to keep the style of the original while making it work for the readers.
In the end, MsT and HT show us that translating Thirukural is tough. MsT makes mistakes because it’s a machine. HT is better because people understand Thirukural’s depth. Thirukural needs a translation that keeps its special meaning and soul, and people can do that better than machines.

Translation Techniques of MsT and HT

Translation is the process of changing a message from one language to another. Different strategies are used in Machine-based Translation (MsT) and Human Translation (HT), which were based on Peter Newmark’s Translation Theory from 1988. These strategies include word-for-word translation, literal translation, faithful translation, semantic translation, adaptation, free translation, idiomatic translation, and communicative translation. Both MsT and HT use these approaches to translate.

In this study, both machines and humans were used to translate Thirukural. For MsT, artificial intelligence (AI) with a database was used to translate Thirukural into English. This type of translation is exact and tries to match words directly from one language to another. However, literal translation doesn’t always capture the full meaning of the text and can result in strange sentences or phrases that don’t make sense. It might also use idioms that don’t work well in the new language. MsT often produces mechanical and robotic-sounding translations because it’s done by algorithms that work on individual words. MsT in this study used literal translation, which focused on the meaning of individual couplets. This led to mistakes in examples, word choices, meanings, and context. As a result, the exact translation approach used by MsT didn’t give accurate results for Thirukural. On the other hand, HT done by GU Pope is better than MsT. When humans translate, they use their language skills to create a better translation. Word-for-word translation is a method used in HT. HT relies on human understanding to translate from one language to another. It’s the oldest and most commonly used translation method, coming in many forms. In this study, GU Pope used the word-for-word method to translate Thirukural into English.

GU Pope’s translation focused on understanding the meaning of each word in Thirukural and translating it as accurately as possible. Even though his translation had a few mistakes, it’s better than MsT. GU Pope used complex words, which might be hard for readers to understand easily. In simpler terms, translation means changing a message from one language to another. Different methods are used for machine and human translation. MsT uses a direct approach that matches words, but it can sound strange and not capture the full meaning. Human Translation, done by GU Pope, is better because it uses language skills to create a good translation. It’s an older method and often produces better results. GU Pope’s translation focused on understanding each word, but it had some mistakes and used complex words.

CONCLUSION

In conclusion, translating literature is the hardest task because it holds the richness of the SL. Translating with human and artificial intelligence brought a translation with a lot of contrast. The result of both MsT and HT shows a difference in the result. However, both brought less accurate translations into the English Language. Machine translation was created to save us from having to manually translate things using dictionaries and other resources. Naturally, the translation outcomes are also precise. But even so.

Additionally, machine translation offers benefits and drawbacks. However, machine translation has a significant impact on our lives since, with machine translation, it is easier for us to read, hear, and discuss foreign languages. Human translation will be more accurate than machine translation. This MsT and HT brought about a lot of contrasts, but the MsT brought about many changes. Machine translation can be less reliable than human translation, especially when translating complex or context-dependent information like Thirukural. Machine translation may have difficulty understanding the nuanced meanings and cultural allusions that human translators can understand. The appropriate diction or tone, which is important for particular texts, could be missed by automated translations. As a result, MsT’s translation of Thirukural needs to be more accurate. Conversely, human translators can ensure correct translations in the right context. The slight cultural variations that could affect the translation are known to human translators. The translation might be tailored to the target audience or kept true to the original author’s writing style. A better translation was made possible through human translation.

REFERENCES

1. Al Mahasees, Z. (2020). Diachronic evaluation of Google Translate, Microsoft Translator, and Sakhr in English-Arabic translation. Unpublished Master’s Thesis, the University of Western Australia, Australia.
2. Baker, M. (1992). In Other Words: A Coursebook on Translation. Routledge. Bassnett, S., & Lefevere, A. (Eds.). (1998). Constructing Cultures: Essays on Literary Translation. Multilingual Matters.
3. Crystal, David, and Simeon Potter. “English Language | Origin, History, & Characteristics.” Encyclopædia Britannica, 19 Oct. 2018, www.britannica.com/topic/English-language.
4. Fengling, L. (2017). A Comparative Study of Nida and Newmark’s Translation Theories. International Journal of Liberal Arts and Social Science, 5(8). Retrieved from https://www.ijlass.org/data/frontImages/gallery/Vol. 5 No. 8/4.31-39.pdf field: the 12th
   International Conference on Translation, 2009. Kuala Lumpur
5. Gallagher, U. (2019). French Translations of Thirukkural – A Comparative Study. Language in India, 19(1930-2940). Retrieved from http://languageinindia.com/aug2019/drumafrenchtranslationsthirukkural.pdf
6. Georges Mounin, and Azizah Hj Ahmad. Masalah Teori Terjemahan. Dewan bahasa dan pustaka KPM Kuala Lumpurr, 1991.
7. Haroon, Haslina, and Hasuria Che Omar. Asas Terjemahan Dan Interpretasi (Penerbit USM). Penerbit USM, 24 Apr. 2014.
8. Hasuria Che Omar, Haroon, H., & Ghani, A. (2009). The sustainability of the translation
9. Hermans, T. (Ed.). (1985). the Manipulation of Literature: Studies in Literary Translation. Croom Helm. Humanities, Arts and Science, 81.
10. Jayashree, D., & Gajjam, A. (2022). A Thirukkural Way of Intellectual Humility. Language in India, 22(1930-2940). https://doi.org/1930-2940
11. Junczys-Dowmunt, M. (2019). Microsoft translator at WMT 2019: Towards large-scale document-level neural machine translation. arXiv preprint arXiv:1907.06170.
12. K, S. (2017). Thirukkural Translation in Deutsch (German) – Karl Grauls Translation Work. Journal of Tamil Peraivu, 5(1), 10–19. https://doi.org/10.22452/jtp.vol5no1.1
13. Klappenbach, Anna. “Most Spoken Languages in the World 2019 | Busuu Blog.” Busuu Blog, 7
14. Jan. 2022, blog.busuu.com/most-spoken-languages-in-the-world/. Malaysian TranslatorsAssociation.
15. Malmkjær, Kirsten. Linguistics and the Language of Translation. Edinburgh Textbooks in Applied, 2005.
16. Mandarić, K. (2022). The Impact of Artificial Intelligence on the Translation Profession. A Case study of Microsoft Translator (Doctoral dissertation, Josip Juraj Strossmayer University of Osijek.
17. Faculty of Humanities and Social Sciences. Department of English Language and Literature). Md, Spring,. Handbook of Supersonic Aerodynamics. Muhibbah Publication, 1950.
18. Morini, Massimiliano. The Pragmatic Translator. A&C Black, 1 Jan. 2013.
19. Munday, J. (2008). Introducing Translation Studies: Theories and Applications. Routledge.
20. Munday, Jeremy. Evaluation in Translation: Critical Points of Translator Decision-Making. Milton Park, Abingdon, Oxon; New York, Routledge, 2012.
21. Munday, Jeremy. Introducing Translation Studies: Theories and Applications. London, Routledge, 2008.
22. Muniapan, B., & Rajantheran, M. (2011). Ethics (business ethics) from the Thirukkural and its relevance for contemporary business leadership in the Indian context. International Journal of Indian Culture and Business Management, 4(4), 453. https://doi.org/10.1504/ijicbm.2011.040961
23. Nagarajan, K. V. (2005). Thiruvalluvar’s Vision: Polity and Economy in Thirukkural. History of Political Economy, 37(1), 123–132. https://doi.org/10.1215/00182702-37-1-123
24. Newmark, P. (1988). A Textbook of Translation. Prentice Hall.
25. Nida, E. A., & Taber, C. R. (1982). The Theory and Practice of Translation. Brill.
26. Priya, S. R. (2020). The Challenges of Translation. International Journal of Research in
27. Putri, T. A. S., & Nugroho, R. A. (2022). TRANSLATION TECHNIQUES USED BY GOOGLE TRANSLATE AND MICROSOFT TRANSLATE IN TRANSLATING ABSTRACT TEXT.
28. UNCLLE (Undergraduate Conference on Language, Literature, and Culture), 2(01), 182–193. Retrieved from http://publikasi.dinus.ac.id/index.php/unclle/article/view/6264/2984#
29. Pym, A. (2010). Exploring Translation Theories. Routledge.
30. Rescigno, A. A., Vanmassenhove, E., Monti, J., & Way, A. (2020). A case study of natural gender phenomena in translation a comparison of Google Translate, Bing Microsoft Translator, and DeepL for English to Italian, French, and Spanish. Computational Linguistics CLiC-it 2020, 359.
31. S.P.Visalakshi. (2017, August 5). International Journal of Trend in Research and Development. Retrieved November 8, 2022, from www.ijtrd.com website: http://www.ijtrd.com/Default.aspx
32. Schäffner, C., & Adab, B. (Eds.). (2001). Developing Translation Competence. John Benjamins Publishing.
33. Smaik, A. (2022, March 2). Literal Translation: Definition, Drawbacks & Examples. Retrieved from Translation Agency – Renaissance Translations website: https://renaissance-translations.com/avoid-literal-translation/
34. Staff. “திருவள்ளுவர் தினம் வந்தது எப்போது?” Https://Tamil.oneindia.com, 26 Nov. 2009, tamil.oneindia.com/religion/hindu/tiruvalluvar.html. Accessed 1 May 2023.
35. Tymoczko, M., & Gentzler, E. (Eds.). (2002). Translation and Power. University of Massachusetts Press.
36. Venuti, L. (1995). The Translator’s Invisibility: A History of Translation. Routledge.
37. Translation, Thirukural, G.U. Pope, Microsoft Translation, Human Translation