A Comparative Study between Chat GPT and Human Translation in Translating English Proverbs into Arabic
Maram R. El-Saadany
Assistant Professor of Translation Studies
Faculty of Social Sciences& Humanities
Galala University, Egypt.
DOI: 10.21608/JSSA.2024.257874.1592
Abstract
Chatbots have been increasingly applied in language learning. It has a perceived effectiveness on the resulting knowledge of knowing the meaning of words. Proverbs are crucial in all languages and cultures. A proverb is defined in Collins dictionary as a short sentence that is often quoted by people to express something about life or to give advice. The proverbs used in this study are gathered from “One Thousand and One English Proverbs translated into Arabic” by Omar Jabak. This research aims to pinpoint the differences between human translation and chatbots while translating twenty proverbs from English to Arabic. Moreover, this study aims to detect the most frequent errors indicated by Chat GPT as a translation strategy. The results reveal that special kinds of errors are performed by some machine translation chatbots more than others. They fail to translate the cultural aspects of the language. Moreover, choosing wrong lexical items and too literal translation are commonly detected errors more than others when analyzing the machine translation techniques as they fail to connote the intended meaning of the proverbs. The findings reveal the fact that some kinds of errors frequently occurred more than others. Also, it shows that the kinds of mistakes detected when using partial equivalence and paraphrasing are very similar when translating proverbs. Distortion of meanings and “comprehension errors” are the most common traced errors when translating proverbs using “paraphrase”. Providing the correct translation without fully comprehending the source text is impossible. Furthermore, this study sheds light on the problems of translations provided by machine translations and the drawbacks of chat Gpt as a translation tool. Finally, further studies are needed in this field of research urgently.
Keywords: Chat Gpt ; proverbs; translation; corpus linguistics; machine translation; error analysis
- Introduction
Proverbs are critical aspects of each language and culture. Nevertheless, translators usually misuse the lexical terms that translate them. It is difficult to translate proverbs to characterize them. Ghazala (2014) said that “Proverbs are metaphors that stand for something else. Besides, they are culture-specific”(p. 138). Researchers have suggested that many strategies have to be used while dealing with proverbs in different languages. For instance, Baker (2011) has suggested certain techniques and strategies for translating proverbs. They are “total equivalence, partial equivalence, paraphrase, omission, and translation by compensation”. Proverbs may sometimes have figurative language. That is to say, giving the exact meaning of the words can hinder the understanding of the whole proverb. Therefore, this paper focuses on the translation quality assessment of proverbs translated by Chatbots in comparison to Omar Jabak in his book “One Thousand and One English Proverbs” which is translated into Arabic and was revised by (Alfaleh, 2017) depending on Na Pham’s (2005) error analysis model.
Practitioners have given chatbots great attention as they can deal with their users in the target language at the same moment (Fryer et al., 2020; Lee et al., 2020). Chatbots are an electronic program that supplements similar human words through words or voices (Ashfaque et al., 2020). Moreover, pre-defined rules have been designed to be used by chatbots. Also, computer applications are not intelligent and cannot answer queries that they were not prepared for. Chatbots deal with more advanced artificial intelligence (AI) techniques such as natural language processing, and deep learning. These AI-powered chatbots are created to translate, and answer queries relying on a huge human language data set (Jiang et al., 2022). This is why these programs can communicate intellectually with users as they learn from their errors, and maintain over time (Fryer et al., 2019; Kim, 2018). They can emphasize important information and answer all queries that help computer users deal with any linguistic feature (Schmidt, 1990).
Chatbots are valid tools that learners use all over the world instantly (Haristiani, 2019; Winkler and Soellner, 2018). These chatbots can provide a variety of answers, translations, and vocabulary that learners might be unfamiliar with (Huang et al.,2022), and explain a real language-learning experience (Chiu et al., 2023). Furthermore, chatbots decrease (Fryer and Carpenter, 2006) the gap between learners and teachers (Kohnke, 2022a). AI maintains automated speech recognition, NLP, to do simple tasks. Chat GPT’s ability to imitate human speech opens a new era in educational tools. On the other hand, it is not accessible in every country. The Foreign Service Institute has declared Arabic aa a” super hard language.” Khoury (2008) emphasized that Arabic is considered the third most difficult language in the classification as the super hard category of languages that “are exceptionally difficult for native English speakers” (Language Continuum: 45). Conversation practices can be handled by AI-powered chat bots. ( Huang et al., 2022) Furthermore, they can increase the learners’ interest ((Fryer et al., 2017; Gallacher et al., 2018; Kohnke, 2022a) and help in their overall growth. (Kim, 2019; Kim et al., 2019; Kohnke, 2022b). It can relate the meaning of a word in a certain text and elaborate on language mistakes, produce texts in various types such as emails, and offer meanings for dictionaries, thesaurus, and translations.
- Literature Review
Many studies have investigated the translation of idioms and proverbs. An analysis of translating dialectal proverbs was published by Almutairi (2013). His main concern was on Baker’s strategies (1992) for idiom translations. The research focuses on investigating whether these strategies can be applied to proverbs or not. The results pinpointed that the translator used literal translation, paraphrasing, and omission. In other words, Baker’s model (1992) for translating idioms also applies to translating proverbs.
Alshammari (2015) investigated Nida’s translation theory in translating Arabic proverbs into English. He examined the ‘accuracy’ of Nida’s theory (1964) for the Arabic proverbs’ translation into English. The results revealed that “dynamic equivalence” is used to make communication easier in the target language whereas “formal equivalence” keeps the culture of the source language.
The previous studies mentioned above have not dealt with mistakes that occurred while translating proverbs which is the main reason for carrying out this research. The current research focused on investigating the errors located when translating proverbs. It also offers a translation quality assessment based on NA Pham’s (2005) model. Furthermore, it assesses the appropriateness of the model to the proverb’s translation.
This paper focuses on investigating the errors obtained from the translation of proverbs by relying on comprehension, linguistics, and translation which were suggested by Na Pham (2005).
- Methodology
The researcher used twenty proverbs to represent the sample of this study. These proverbs are extracted from “One Thousand and One English Proverbs” translated by Omar Jabak. The twenty proverbs are translated using AI-powered Chat GPT and assessed against Omar Jabak’s translation. Na Pham’s error analysis model is used as an assessment tool.
3.1 Na Pham’s Error Analysis Model
Na Pham emphasized ted that “Translation errors show the inability of the students in expressing the meaning of the source text in the target language, or some distortion of the source text, even though the sentence may be grammatically correct”. She mentioned nine translation problems namely: pragmatic errors, inaccurate rendition of individual lexical items, distorted meaning of the source text, omission, addition, free translation, too literal translation, wrong lexical choice, and wrong focus of attention.
Figure 1.1Na Pham’s (2005) error analysis model
In Figure 1.1 the arrows indicate that every comprehension error can cause a translation error. You cannot give a suitable translation without a full understanding of the main text. In some cases, linguistic errors and translation errors are connected. For instance, inaccurate meaning is caused by committing a linguistic error.
3.2 The Sample
The sample of this study is represented by twenty proverbs to be translated by two methods: Human Translation (HT) and machine translation ( MT) or (Chat GPT). These proverbs are taken from an English Arabic dictionary for proverbs “One thousand and one proverbs”. They have been translated using the machine program, Chat Gpt, and Omar Jabak’s assessed translation for “One Thousand and One English proverbs”.
3.3 Research Question
- Is Chat Gpt capable of translating cultural aspects as human translators?
- Is there a difference between human translation and Chat Gpt translation of proverbs from English to Arabic?
- Method of Analysis
The researcher used chatbot technology presented by Chat GPT 3 to translate some proverbs taken from “One thousand and one English proverbs”. Second, twenty proverbs only are included in this study as examples of cultural problems that can face Chat Gpt while translating these proverbs. Next, the researcher investigated the translation of every proverb concerning monolingual and bilingual dictionaries. Moreover, an analysis of Chat GPT translation has been conducted and compared to the correct translations’ given. This unique step added another dimension to the research which is comparing human translations to machine translation, especially chat Gpt which is considered a trendy way of translation that is used nowadays by all users all over the world. The cultural aspect is a major challenge that we all face, so translators and linguists need to highlight and shed light on as machine translation techniques are not programmed to avoid such problems. Finally, a translation assessment is carried out by using Na Pham’s error analysis model.
4.1 Error Analysis
4.1.1 Using Partial/ Close Equivalents
1- ST: Where ignorance is a bliss, it is folly to be wise
MT: حينما يكون الجهل سعادة، فمن الحماقة أن تكون حكيما
HT : وأخو الجهالة في الشقاوة ينعم ذو العقل يشقى في النعيم بعقل
Translate the following:
Spears (2005) stated that the proverb means that “If knowing something makes you unhappy, it would be better not to know it” (p. 753). This implies that this proverb is discussing “how” happy you would be when you are ignorant and not about” ignorance” in general.
It has an Arabic equivalent which says ” وأخو الجهالة في الشقاوة ينعم ذو العقل يشقى في النعيم بعقله “ (Baalabki, 2005, p. 94). Thus, we can say that the chat Gpt has given an inappropriate translation of the source text proverb. This shows that Chat Gpt cannot preserve the meaning of the main text when giving a “translation shift in structure” (Na Pham, 2005, p. 212).
2- ST: A friend is never known till needed
MT: لا يُعرف الصديق إلا في حالة الحاجة
HT : الصديق إلا عند الشدةلا يُعرف
Translate the following:
The proverb means that you will never understand the real qualities of your friend until you go through a hard time. On the other hand, the machine translation Chat Gpt translation of the proverb means that your real friend is the friend who stands by you in your hard times or when needed. The proverb here concentrates on the time of knowing your real friend whereas the Chat GPT focuses on the qualities of a real friend. Attia (2004, p. 61), emphasized that the translation of this proverb should be ” الصديق إلا عند الشدة لا يُعرف”. Chat Gpt’s error is on the “Wrong Focus of Attention”.
3- ST: To kick against the Pricks
MT: ان يركل بقوه ضد الشوكه
HT : لا تضرب الحائط بنفسك
Translate the following:
Attia, (2004, p. 36) has translated this proverb to “لا تضرب الحائط بنفسك”. Although Chat Gpt here is aware that this is a proverb, and it is culture-specific; it has provided a wrong translation. It distorted the meaning of the proverb as actually, it does not give comprehensive meaning to the proverb. A comprehension error is the main reason for the wrong translation. The meaning of the source text proverb is “It is futile to fight a losing battle by resisting authority” (Proverb Hunter Dictionary).
4- ST: Between two evils, it is not worth choosing
MT: بين شرين، لا يستحق الاختيار
HT : الاختيار بين شرين لا جدوى له
Translate the following:
Here, it means that it does not matter if you choose between two evils. The correct translation can be “الاختيار بين شرين لا جدوى له”. The problem here with chat Gpt is that it does not concentrate on the rules that govern the construction of sentences in Arabic. In Arabic, we either have nominal or verbal sentences. That is why the translator started with a verbal sentence to produce the punctual meaning needed. Chat Gpt “distorted the meaning” of the proverb due to a lack of understanding of the text.
5- ST: As the tree falls, so shall it lie
MT: كما تسقط الشجرة، هكذا تبقى
HT: يبعث كل عبد علي ما مات عليه
Translate the following:
Speak (2015) has explained that the given proverb is taken from the Bible. The equivalent Arabic proverb to this one could be ” ما مات عليه يبعث كل عبد على “. Here chat Gpt has provided a wrong translation due to a lack of comprehension of the English proverb, which is far away from the real intended meaning of the English proverb.
Translation by Paraphrase:
6- ST: Everybody’s business is nobody’s business
MT: شأن الجميع لا يكون شأن أحد
HT: ما حك جلدك مثل ظفرك فتولَّ أنت جميع أمرك
Translate the following:
Speake (2005) states that this proverb means “If a matter concerns a wide range of people but there is no person specifically in charge to deal with it, people neglect it” ( p. 97). Chat Gpt has provided a paraphrase for the given words. It translated it as “ شأن الجميع لا يكون شأن أحد” . Baalbaki (2005) emphasized that the equivalent Arabic proverb to this one is
ما حك جلدك مثل ظفرك فتولَّ أنت جميع أمرك (p. 37). Here, the chatbot distorted the meaning as it failed to comprehend the meaning. Thus, the meaning is distorted.
7-ST: The child is the father of the man
MT: والد الرجل الطفل هو
HT : من شب على شيء شاب عليه
Translate the following:
Speake (2015) claimed that the meaning here is “The character, interests, and
abilities of the adult are foreshowed in those of the child” (p. 50). Also, it can be stated as
” The child’s character gives insights into the kind of the man he will grow up to be” (L. Flavell &R. Flavell, 2006, p. 47).
This proverb talks about the child and the grown-up he becomes. It does not talk about the son and the father. The Arabic equivalent to this proverb is ” من شب على شيء شاب عليه ” (Attia, 2004). In conclusion, Chat Gpt has distorted the meaning. It translates the proverb by paraphrasing its words.
8-ST: No pains, no gains
MT: لا ألم، لا مكاسب
HT: لا شيء يأتي بسهوله
Translate the following:
Another vivid example that reveals the distortion of the meaning is revealed in this proverb. Here chap Gpt failed to translate the correct meaning as it has provided a paraphrase or word-by-word translation which does not reflect the real intended meaning of the proverb. The connoted meaning is provided by Attia 2004 as he translated it as “لا شيء يأتي بسهوله”.
9-ST: The Dutch has taken Holland
MT: الهولنديون استولوا على هولندا
HT: خبر قديم جدًا
Translate the following:
In the Farlex Dictionary of Idioms, the proverb is “A sarcastic phrase said in response to outdated news”. Therefore, it cannot be translated as “ خبر قديم جدًا”. This proverb exemplifies how proverbs are culture- specific. It also shows that the translator has to be aware of the history and the culture at the same time. Such a sarcastic attitude cannot be connoted by a machine and that is why it has translated a literal translation as its dictionary does not have such intended meaning. Therefore, chat Gpt has failed to provide the correct translation due to a lack of comprehension.
Literal Translation:
10-ST: Give a dog a bad name and hang him
MT: إعطِ كلب اسمًا سيئًا واشنقه
HT: لا تصاحب الأحمق
Translate the following:
Speake (2015) claimed that the meaning of this proverb is, “A person’s plight is hopeless once his reputation has been ruined”. Therefore, the word ‘name’ can be translated as ‘reputation’ or ‘accusation’. Attia (2004) has translated it as” ألق على الكلب تهمة سيئة ثم اشنقه” (p. 63). Here Chat Gpt has provided a wrong lexical choice. When Chat Gpt was asked about his choice of translation, it stated that it had provided a literal translation. Although it is also aware that this is an idiomatic expression and these expressions can vary across languages and cultures, it could not translate it correctly.
11- ST: A forced kindness deserves no thanks
MT: تفرض اللطف لا تجد له شكرا
HT: المعروف الذي يأتي كرهًا لا يستحق الشكر
Translate the following:
In the Proverb Hunter Dictionary, this proverb means “The doer of kindness deserves thank you, but he does not deserve it if he has done it against his will”. Attia (2004) claime that the translation can be ” المعروف الذي يأتي كرهًا لا يستحق الشكر ” .Again a literal translation has been provided by chat Gpt. A wrong lexical choice has been provided for the word “kindness” which has distorted the meaning. Also, the word تفرض” “ has no meaning in Arabic. A morphological error appeared here in the derivation of the word. Even the alternative substitution of the word which is المفرض” “ is a wrong morphological choice as it should be “المفروض” . The choice of this word is a lexical choice that should be avoided. The right translation is “ كرها”.
12-ST: Imitation is the sincerest form of flattery
MT: التقليد هو أبدع صورة للمديح
HT : التقليد هو أصدق أنواع الإطراء
Translate the following:
Flavell L. & Flavell R. (2006) emphasized that meaning here is “Copying someone
pays an implicit compliment to that person “. Therefore, the translation could be ” التقليد هو أصدق أنواع الإطرا” (Attia, 2004). On the other hand, chat Gpt used the word ‘flattery’ as ‘ للمديح ‘ which is a wrong lexical choice because it failed to understand the meaning of the lexical word.
13-ST: Don’t meet trouble halfway
MT: لا تلتقِ المشاكل في منتصف الطريق
HT : لا تبحث عن المشاكل قبل ان تقع
Translate the following:
Ghazala (2014) stated, “Proverbs should not be taken literally and directly”. Speake (2015) states that it is near in meaning to “never trouble the trouble untill trouble troubles you” (p. 206). It can also be stated as “There is no benefit in actively seeking out problems or difficulties”. It can be as “ لا تبحث عن المشاكل قبل أن تقع”. (Attia, 2004). Again, chat Gpt has mistranslated the words “meet” and “halfway”. A wrong lexical choice has been provided by Chat Gpt which has distorted the meaning.
14-ST: One man’s meat is another man’s poison.
MT:طعام رجل هو سم رجل اخر
HT: الاذواق تختلف
Translate the following:
Spears (2005) stated that the meaning of this proverb is “Something that one person likes may be distasteful to somebody else”. Therefore, it is applicable when people differ in their taste. As seen in the chat above, Chat Gpt failed to deliver the meaning of the proverb. In addition, it gave un meaningful translation as it has no meaning to say
“طعام رجل هو سم لرجل اخر”
In addition, when asked chatbot about the right translation, it admitted that the machine was wrong in translating the intended meaning. Overall, the phrase “الأذواق تختلف” highlights the importance of embracing diversity, respecting individual differences, and promoting harmonious interactions within Arabic-speaking societies. It serves as a reminder to be tolerant, understanding, and accepting of others’ unique perspectives and preferences.
15-ST: He that has not silver in his purse should have silk in his tongue
MT:من لا يملك فضه في جيبه يجب ان يمتلك حريرا في لسانه
HT: من لا يملك مالا يجب أن يكون له كلاما معسولا
Translate the following:
The proverb Hunter Dictionary illustrates that a man who does not possess money cannot offend those whom he looks for financial support. Therefore, we can conclude that this proverb targets poor people. The cultural significance of using silk as a metaphor in Arabic has been ignored by chat GPT. The program provided the literal meaning of the word “silk” in its translation. According to Attia (2004), the meaning can be addressed in Arabic as
“ من لا يملك مالا يجب أن يكون له كلاما معسولا“.So, Chat Gpt did not express the right meaning in the given translation.
16- ST: Good wine needs no bush
MT: جيدة النبيذ لا يحتاج إلى شجيرة
HT: البضائع الجيدة تعلن عن نفسها بنفسها
Translate the following:
Speake (2015), illustrated that it means that we do not need advertisements for good products. Baalbaki (2015) translated it as البضائع الجيدة تعلن عن نفسها بنفسها
In Chat GPT’s translation, it wrote النبيذ الجيد لا يحتاج إلى شجيرة
The machine failed to convey the message of the idiomatic expression and the meaning of the translated proverb is distorted. Also, the morphological derivation used by the chatbot has caused a comprehension problem. Chat Gpt proved that it failed to translate the English proverb.
17- ST: No man is indispensable.
MT: لا يوجد رجل لا يمكن الاستغناء عنه
HT: ليس هناك من لا يمكن الاستغناء عنه
Translate the following:
According to the Proverb Hunter Dictionary, this proverb is mentioned in business companies and it means, anyone may lose his or her job if the company sees that he/she is unnecessary.
So, Attia (2004) stated that the best translation for it is ” ليس هناك من لا يمكن الاستغناء عنه “
Chat GPT failed to give this accurate translation and translated it literally. However, when providing the right translation, it admitted that this translation conveyed the meaning “effectively”.
18- ST: Charity covers a multitude of sins.
MT: الصدقة تغطي الكثير من الخطايا
HT: الصدقة تخفي العديد من الآثام
Translate the following
Atia (2004) mentioned that it is taken from a religious virgin, the bible, so the best translation offered is ” الصدقة تخفي العديد من الآثام “
As for Chat Gpt, it translated the religious part as it is but it failed to notice the difference between الاثم و الخطيئه
The former الخطيئهis a minor mistake whereas the latter الاثم is a major one and there is a big difference between them in the meaning and their religious penalties.
19- ST:A good name is sooner lost than won
MT : اسم جيد يضيع أسرع مما يُكتسب
HT: السمعة الطيبة تُكتسب بصعوبة وتضيع بسهولة
Translate the following:
Attia (2004,p..69) translated the proverb as ” السمعة الطيبة تُكتسب بصعوبة وتضيع بسهوله “
The word “name” here connotes the meaning of the “image” or the “reputation’. However, Chat Gpt failed to represent this figurative meaning. A wrong lexical choice leads to the distortion of meaning and the message was not delivered appropriately because of the misunderstanding of the proverb.
20-ST: Give the devil his due
MT: امنح الشيطان حقه
HT: أعط الحق لأهله حتى لو كان شيطانا “
Translate the following
The figurative meaning here is misinterpreted in the word” devil. The exact translation should be ” أعط الحق لأهله حتى لو كان شيطانا ” (Attia, 2004). Chat Gpt failed to deliver the exact meaning of the translated proverb as it is written to show the extent to which whom we should “give the due”, even if it was given to the devil himself.
- Conclusion & Discussion
The present study examined the differences between machine translations and human, correct translation. The findings of this study shed light on the strategies used by Chat Gpt.
Two research questions have been investigated and the findings concerning each question will be discussed in the coming section. Proverbs exist in all languages. Chat GPT is a well-known tool with significant goals to increase our usage and engagement in language learning. Also, it is a computerized tool for language learning and has been used in educational aspects. (Bailey et al., 2021). Chat Gpt faces many challenges that all engineers have to take into consideration. The biggest challenge is the cultural aspect of any language. Compared to educational channels, chatbots can facilitate people’s interactivity with the computer, and support learner’s autonomy (Nguyen & Sidorova, 2018); it is expected to become a pioneer learning tool as an alternative to other websites (Heo & Lee, 2018) but this could not solve the problem of the cultural aspect of the language. A comparative human translation is added to reveal the problematic areas that the machine has. Table (1) shows all the proverbs translated by the Chat GPT and human translation.
Translators have to be aware of the new technologies and what they offer to people. It has been claimed that machine translation can replace human translators and can eliminate such jobs. This research is clear proof that machines cannot replace humans as cultural aspects and dialects are real challenges for all computer programs. This paper is so important as we are in an urgent need to assess the quality of translated texts produced by machine translation. The translation of proverbs is assessed from two perspectives: human and machine translations. The assessment process was based on comparing proven and documented translations provided by human translators and chat Gpt as the newest machine translation tool. Giving an inaccurate meaning, meaning distortion and comprehension errors were the traced errors. Above all, wrong lexical and morphological derivations are the most common problems that face ChatGpt.
The main problem that we face nowadays is that new generations lack any cultural aspects in their lives. This can lead us to a generation that can rely on Chatbots. So practitioners and software developers must be aware of these cultural and linguistic problems that need language specialists to locate and identify. Therefore, more concentration should be given to the meaning. To sum up, the key point is that chat Gpt-produced translations are not texts that we humans consider “meaningful”. To get such texts we need to build on all that “human context” and other materials that we humans have written. The “row computational system” will just do “raw computation”. Finally, this paper is written to highlight the importance of the work of a “language specialist” concerning Engineers and they are complementary sides to achieve success in AI Technology.
Research Questions:
1- Is Chat Gpt capable of translating cultural aspects as human translators?
This study has given a complete analysis that reflects the fact that Chat Gpt is incapable of translating cultural aspects as accurately as humans and that is why a “Language Specialist” is needed to work side by side with the programmers to avoid such problems in the future. Most of Chat Gpt’s translations have lexical and morphological errors that have distorted the meaning of the translation.
- Is there a difference between human translation and Chat Gpt translation of proverbs from English to Arabic?
The study revealed that Chatbots cannot translate proverbs or idiomatic expressions accurately as they are culture-specific and they need a Linguist practitioner to help with the data provided to the machine to correct any derivational problem either lexically or morphologically.
Wrong lexical and morphological derivations and too-literal translations were the most frequently traced problems presented by Chat GPT. Consequently, Language practitioners should be working with the programmers to make sure of the lexical choices built into the program to be used when translating proverbs or any translated sentence. In other words, having a Language Specialist to assist the programmers is a must as proverbs do have a figurative meaning which implies that they should not be translated literally. Translators should first search for the meaning of the proverb in dictionaries to decide which translation should be given.
Last but not least many people stated that there are some drawbacks to Chat GPT. First, is the responses given by Chat Gpt original? Chat GPT does not provide references or citations. Some people may criticize its responses as they are not genuine and it is plagiarism. The next question is How accurate is the Chatbot? Bowman (2022) emphasized that in many incidents in which you ask Chat GPT a question, it’ll give you a very effective impressive answer that’s just totally wrong. Lastly, it provides a literal meaning as it lacks all the cultural aspects related to all languages and not only Arabic.
- Recommendations
Language practitioners or specialists should work parallel to programmers to avoid all the cultural problems raised in this paper. Machine translation programmers should be aware that some proverbs do not have a straightforward meaning, meaning they should not be translated word for word. Therefore, translators should first search the meaning of the proverb in dictionaries and enhance their programs with these translated proverbs to provide more accurate data with the aid of the “Language Practitioners”. More acquaintance with the system of the language program that is added to the machine has to be our future concern. Although chatbot is one of the newest machine tools used nowadays by everyone, more concern must be given to each language’s problematic areas. Chat GPT-given rules for any language have to be fed into the systems under the supervision of language specialists. Universities and schools must map guidelines for using it properly in their classrooms. They should prepare their students for a world where technology dominates their lives. Last but not least, it is not easy to be a translator. There are lots of obstacles that can hinder you while performing this job. Translation errors occur because of these problems and obstacles. That is why translation quality assessment is important. Other researchers should work more on this topic and the merging roles of the “Language Specialist” and the “Programmer”.
- References
Alfaleh, B. A. (2017). Translation Quality Assessment of Proverbs from English into Arabic: TheCase Study of One Thousand and One English Proverbs Translated into Arabic. Department of English Language and Literature, College of Languages and Translation, Al-Imam Mohammad Ibn Saud Islamic University
Almutairi, E. S. (2013). Translating dialectical proverbs with particular reference to Bedouin
Arabic: Abdelrahman Munif’s Mudun Al-Milh as case study (Master’s Thesis). University of Salford Manchester. Retrieved from The Saudi Digital Library. Retrieved from
http://search.sdl.edu.sa/Dissertation/ucform.aspx?q=%23dialectal&sort=
Alshammari, J. N. M. (2015). Examining Nida’s translation theory in rendering Arabic proverbs
Into English. A comparative analysis study. International Journal of English Language
and Linguistics Research, 3(8), 45-75. Retrieved from
http://www.eajournals.org/wpcontent/uploads/Examining-Nidas-Translation-Theory-in-
Rendering-Arabic-Proverbs-into-English-A-Comparative-Analysis-Study.pdf
Ashfaque M W,Tharewa lS, IqhbalS, etal.(2020) A review on techniques, characteristics and approaches to an intelligent tutoring chatbot system. In 2020 International Conference on Smart Innovations in Design, Environment, Management, Planning, and Computing (ICSIDEMPC) (pp. 258–262). IEEE.
Attia, M. (2004). Dictionary of common English Proverbs. Retrieved from http://www.attiaspace.com/Publications/CommonProverbs.pdf
Bailey, D., Southam, A., & Costley, J. (2021). Digital storytelling with chatbots: Mapping L2 participation and perception patterns. Interactive Technology and Smart Education, 18(1), 85–103. https://doi.org/10.1108/ITSE-08-2020-0170
Baalbaki, M. (2005). Al-Mawrid English-Arabic dictionary. Beirut: Dar El-Elm.
Baalbaki, R. M. (2013). Al-Mawrid a dictionary of idiomatic expressions English-Arabic. Beirut: Dar El-Elm.
Baker, M. 1992. In Other Words. London: Routledge.
Baker, M. (2011). In Other Words. Retrieved from https://www.slideshare.net/abdullahktk2/mona baker-in-other-words-a-coursebook-on-translation routledge-2011
Bowman E (December 19, 2022) A new AI chatbot might do your homework for you. But it’s still not an A+ student. NPR. Retrieved January 5, 2022
Chiu TKF, Moorhouse LM, Chai CS, et al. (2023) Teacher support and student motivation to learn with Artificial Intelligence (AI) based chatbot. Interactive Learning Environment. Epub Online. DOI: 10.1080/10494820.2023.2172044
Flavell , L., and Flavell, R. (2006). Dictionary of proverbs and their origins. Britain: Kyle Books
Fryer L and Carpenter R (2006) Bots as language learning tools. Language Learning& Technology 10(3): 8–14.
Fryer LK, Ainley M, Thompson A, et al. (2017) Stimulating and sustaining interest in a language course: an experimental comparison of Chatbot and human task partners. Computers in Human Behavior 75: 461–468.
Fryer LK, Nakao K and Thompson A (2019) Chatbot learning partners: Connecting learning experiences, interest and competence. Computers in Human Behavior 93: 279–289.
Fryer LK, Coniam D, Carpenter R, et al. (2020) Bots for language learning now: Current and future directions. Language Learning & Technology 24(2):8–22. Retrieved from http://hdl.handle. net/10125/44719
Gallacher A, Thompson A, Howarth M (2018) “My robot is an idiot!” – Students’ perceptions of AI in the L2 classroom. In: Taalas P, Jalkanen J, Bradley L, et al. (eds) Future-proof CALL:
Language Learning as exploration and encounters – short papers from EUROCALL 2018
(pp. 70–76). Research-publishing.net. DOI: 10.14705/rpnet.2018.26.815
Ghazala, H. (2014). Translation as problems and solutions. (10th ed.). Jeddah: Konooz Elmarefa.
Haristiani, N (2019) Artificial intelligence (AI) chatbot as language learning medium: an inquiry. Journal of Physics: Conference Series 1387(1): 202.
Heo, M., & Lee, K. J. (2018). Chatbot as a new business communication tool: The case of NaverTalkTalk. Business Communication Research and Practice, 1(1), 41–45. https://doi.org/10.22682/bcrp.2018.1.1.41
Huang W, Hew KF and Fryer LK (2022) Chatbots for language learning–are they really useful? A systematic review of chatbot-supported language learning. Journal of Computer Assisted Learning 38(1): 237–257.
Jiang H, Cheng Y, Yang J, et al. (2022) AI-powered chatbot communication with customers: dialogic interactions, satisfaction, engagement, and customer behavior. Computers in Human Behavior 134: 107329.
Khoury,G.(2008).Vocabulary acquisition in Arabic as a foreign language: The root and pattern Strategy.(Unpublished doctoral dissertation), Boston University: United States.
Kim H, Shin D, Yang H, et al. (2019) A study of AI chatbots as an assistant tool for school English curriculum. Journal of Learner-Centered Curriculum and Instruction 19(1): 89–110.
Kim H, Yang H, Shin D, et al. (2022) Design principles and architecture of a second language learning chatbot. Language Learning & Technology 26(1): 1–18. http://hdl.handle.net/10125/73463
Kim NY (2018) A study on chatbots for developing Korean college students’ English listening and reading skills. Journal of Digital Convergence 16(8): 19–26.
Kim NY (2019) A study on the use of artificial intelligence chatbots for improving English grammar skills. Journal of Digital Convergence 17(8): 37–46.
Kohnke L (2022a) A pedagogical chatbot: A supplemental language learning tool. RELC
Journal.Epub Online. DOI: 10.1177/00336882211067054
Kohnke L (2022b) A qualitative exploration of student perspectives of chatbot use during
emergency remote teaching. International Journal of Mobile Learning and Organisation
16(4):475–488. DOI: 10.1504/IJMLO.2022.125966
Lee JH, Yang H, Shin D, et al. (2020) Chatbots–technology for the language teacher. ELT Journal 74(3): 338–344.
Mackey A (2012) Interaction, Feedback and Task Research in Second Language Learning: Methods and Design. Cambridge: Cambridge University Press.
McMinn S (2023) ChatGPT killed the classroom star: AI’s rise means it’s time to rethink teaching and testing. South China Morning Post Online. Available at https://www.scmp.com/comment/ opinion/article/3206436/chatgpt-killed-classroom-star-ais-rise-means-its-time-rethink-teaching and-testing (Accessed 28/1/23)
Na Pham, P. Q (2005). Error analysis in Vietnamese – English translation: pedagogical implications (doctoral dissertation). University of Western Sydney. Retrieved from http://handle.uws.edu.au:8081/1959.7/20242
Nida, E. A. (1964). Toward a science of translating: with special reference to principles and
procedures involved in Bible translating. Brill Archive
Nguyen, Q. N., & Sidorova, A. (2018). “Understanding user interactions with a chatbot:
A self-determination theory approach.” In AMCIS 2018 Proceedings, 3.
https://aisel.aisnet.org/amcis2018/HCI/Presentations/3
Schmidt RW (1990) The role of consciousness in second language learning. Applied Linguistics 11(2): 129–158.
Speake, J. (ED). (2015). Oxford dictionary of proverbs. (6th ed.) New York, NY: Oxford University Press.
Spears, R. A. (2005). McGraw-Hill’s dictionary of American idioms. Retrieved from
http://english4success.ru/Upload/books/959.pdf.
Thalji, M. B., and Dweik, B. S. (2015). The translation of proverbs: obstacles and strategies
(Master’s thesis). Middle East University. Retrieved from
http://www.meujo.com/uploads/1/5870d3856b3ac_1.pdf
The Dutch have taken Holland. (n.d.) Farlex Dictionary of Idioms. Retrieved
from https://idioms.thefreedictionary.com/the+Dutch+have+taken+Holland
Winkler R and Soellner M (2018) Unleashing the potential of chatbots in education: A state-of-the-art analysis. In Academy of Management Annual Meeting Proceedings. Academy of Management.
دراسة مقارنة بين Chat GPT والترجمة البشرية في ترجمة الأمثال الإنجليزية إلى العربية
د. مرام السعدني
مدرس لدراسات الترجمه
كليه العلوم الانسانيه، جامعه الجلاله، جمهورية مصر العربية.
المستخلص:
معرفة معاني كلمات المثل الشعبي أمر ضروري في جميع اللغات والثقافات. ويُعرَّف المثل الشعبي في قاموس كولينز كجُملة قصيرة يُنقلها الناس في كثير من الأحيان للتعبير عن شيء ما عن الحياة أو لإعطاء نصيحة. وتم جمع المثل الشعبي المستخدم في هذه الدراسة من كتاب “ألف مثل شعبي إنجليزي مترجم إلى العربية” لعمر جابك. تهدف هذه الدراسة إلى تحديد الاختلافات بين الترجمة البشرية والدردشة الذكية أثناء ترجمة عشرين مثلاً من الإنجليزية إلى العربية. علاوة على ذلك، تهدف هذه الدراسة إلى اكتشاف الأخطاء الأكثر تكرارًا التي يشير إليها Chat GPT كاستراتيجية ترجمة. تشير النتائج إلى أن هناك أنواعًا خاصة من الأخطاء يؤديها بعض دردشة الذكاء الاصطناعي للترجمة أكثر من غيرها. إنها تفشل في ترجمة الجوانب الثقافية للغة. علاوة على ذلك، اختيار العناصر اللغوية الخاطئة والترجمة الحرفية هي الأخطاء الشائعة التي يتم اكتشافها بشكل أكثر تكرارًا من غيرها عند تحليل تقنيات الترجمة الآلية حيث يفشلون في توصيل المعنى المقصود من المثل. تكشف النتائج عن حقيقة أن بعض أنواع الأخطاء يحدث بشكل متكرر أكثر من غيرها. كما تظهر أن أنواع الأخطاء المكتشفة عند استخدام المعادلة الجزئية وإعادة الصياغة متشابهة جدًا عند ترجمة المثل. إن تحريف المعاني و”أخطاء الفهم” هي الأخطاء الأكثر شيوعًا التي يتم اكتشافها عند ترجمة المثل باستخدام “إعادة الصياغة”. فإن توفير الترجمة الصحيحة دون فهم النص المصدر بشكل كامل أمر مستحيل. علاوة على ذلك، تسلط هذه الدراسة الضوء على مشاكل الترجمة التي تقدمها الترجمة الآلية وعيوبها.
الكلمات الرئيسية:Chat GPT ,المثل الشعبي، الترجمة، اللغويات النصية، الترجمة الآلية، تحليل الأخطا




