Content uploaded by Mohammed Shormani
Author content
All content in this area was uploaded by Mohammed Shormani on Sep 14, 2024
Content may be subject to copyright.
1
Translating Arabic free poetry texts into English by ChatGPT: Success and
Failure
(To appear in IJT&IS)
Nasim Alwagieh
PhD Student, Ibb University
Mohammed Shormani
Professor of Linguistics, Ibb University
email: shormani@ibbuniv.edu.ye
Abstract
The purpose of this study is to examine the success and failure of ChatGPT in translating some
Arabic free poetry texts selected from Wahib's 'Hulm 'ala Nasiat Aljabin 'A Dream on the Corners
of Nostalgia’. In this study, we examined ChatGPT’s ability in translation field. In particular, the
focus is on translation in Arabic context. The free poetry texts were first translated by the authors,
then by ChatGPT. Both translations were then analyzed and compared to find out to what extent
ChatGPT translation deviates from human translation. Qualitative approach is adopted, with the
help of comparative and analytical methods. The results show that ChatGPT is to some extent
successful in producing a natural English translation of the Arabic poems in this study. While
human adopts translation strategies like paraphrasing, adapting, recreating, transliteration.
ChatGPT adopts translation strategies like paraphrasing addition, word-for-word, literal
translation. But there is no care for the aesthetic values in ChatGPT’s translation, and many types
of translation errors result from applying these strategies.
Keywords: ChatGPT, Translation, Arabic, free poetry/verse, translation strategies
1. Introduction
Translation is a process of transferring information, experience and knowledge from generation to
another (Newmark, 1988). It has several theories including linguistic theory (Catford, 1965),
cultural theory (Geertz, 1973; Conway, 2012; Shormani, 2020), Skopos theory (Flynn, 2004),
among others. All these theories help in shaping translation as both product and process. Lexical
equivalence is not everything in translation (Nida, 1964). With the developments of technology,
translation industry witnesses an important shift and enters computer world, what has been known
as Computer-assisted translation (CAT), which has several tools or models such as Google
Translate. CAT tools develop paralleling technology developments. Artificial intelligence (AI)
developments result in several models used in translation industry. One of these models is
ChatGPT which has been developed by OpenAI company in 2022. It is a language learning model
(LLM) which outperformed and surpassed CAT tools.
In this study, we examined ChatGPT’s ability in translation field. In particular, we focus on
translating a free poetry text in Arabic context. The text is taken from a collection of poems entitled
Hulm 'ala Nasiat Aljabin ‘A Dream on the Corners of Nostalgia’ by Najwa Kinana Wahib. The
2
text was first translated by the authors, then by ChatGPT. We analyzed and compared both
translations to find out to what extent ChatGPT translation deviates from human translation, where
ChatGPT performed like humans, what are the literary aspects that ChatGPT did not capture, and
how it can be improved.
The rest of the paper is organized as follows. In section 2, we provide a literature review of the
relevant research. In section 3, we present the research design. In section 4, we tabulate the
important results. In section 5, we elaborate on the most important findings and compare both
translations: human and ChatGPT. Section 6 concludes the article.
2. Literature Review
In human life, the term ‘intelligent’ is used to refer to someone who is special in doing complex
tasks. Nowadays, it is also used to refer to machine. AI “refers to intelligent devices and software
that can reason, absorb, gather knowledge, interact, control, and distinguish between objects”
(AlAfnan et al., 2020, p. 61). This is the main concern of this paper, focusing on the intelligence
of ChatGPT (Chat-Generative Pre-Trained Transformer), an intelligent language model, launched
by OpenAI company in 2022. This AI model can learn and perform tasks like humans. As the name
suggests, ChatGPT can be involved in chats with humans. Though ChatGPT has been investigated
by a considerable number of scholars, studies concerning ChatGPT use in translation are still very
few, especially in Arabic context, and more especially in literary sphere. Therefore, this literature
review will focus on studies concerning AI use in translation industry.
Grace et al. (2018), for example, focused on the role of AI in reshaping of the future by the
upcoming years. They insisted on high-level machine intelligence (HLMI) achievement as unaided
machines accomplishing tasks better and more cheaply than human, expecting that AI machines
may outperform human. They gave a timespan for each and every field of life and study to replace
human within a couple of years. For example, AI would cover translating languages by 2024, and
AI would also enter academia by 2026, whereby student in high school academic writing will
change by 2026 than before. By 2049, there will be no need for authors and writers as writing a
bestselling book will be by the help of AI, stating that “[r]esearchers believe there is a 50% chance
of AI outperforming humans in all tasks in 45 years and of automating all human jobs in 120 years”
(Grace et al., 2018, p.1). They also go a step further Asians expected changes 44 years before North
Americans, “displays a similar gap between the two continents with the most respondents in the
survey: China (median 28 years) and USA (median 76 years)” (Grace et al. 2018, p.5).
As for the use of ChatGPT, Songsiengchai et al. (2023) explore the potential of AI and to compare
the English language learning effectiveness among Thai students after implementing AI models
like ChatGPT to facilitate English language learning. The participants were students aged 19-20
from first-year pre-service teachers in Bangkok. 120 students participated, 60 in the control and
60 in the experimental group. The feedback from the focus group interviews stated that the real-
time feedback and interactive exercises offered by ChatGPT helped the students to understand and
apply to language concepts more effectively. They concluded their study with “students who used
ChatGPT exhibited high motivation, self-confidence, and a positive attitude shift” (Songsiengchai
et al., 2023, p.75). One of the main obstacles faced by students is related to the equitable access to
AI technologies. It is not the same in all over the country. Students in the countryside had little
time to be able to access to the internet as oppose to the students in the cities.
3
Dergaa et al. (2023) conducted a study, exploring the potential benefits and threats of ChatGPT
and other NLP technologies in academic writing and research publications, highlighting the ethical
considerations involved in using these tools, and considering the impact they may have on the
authenticity and credibility of academic work. In an answer to the question: Will ChatGPT replace
academic researchers? they argued that ChatGPT is a robust LLM with various capabilities, yet it
also possesses limitations that make it unsuitable for certain academic research types. There are
also researchers who strongly advocate the use of ChatGPT in the academic writings. For instance,
Zohery (2023) argues that ChatGPT as a powerful IA tool can be used in first draft writing, article
summarization, as well as translation into any intended language. There are several advantages of
ChatGPT in academic writing and publishing including speedy drafting, stimulating creativity,
providing feedback and suggestions, and create more fun and engaging scientific writing by
imitating a natural dialogue format between writers and AI assistants (Brady et al., 2023; Fitria,
2021).
Fitria (2021, p. 45) discussed the wide spread of ChatGPT in Twitter as a social network, stating
that AI could be “one of the technological trends which are relevant to be developed for any
purpose’ … a technology that is undergoing rapid development”. Chatbots as conversational tools
help students and all people to practice more activities and be more productive. One of the main
benefits of the appearance of AI chatbots was their ability to generate texts, emphasizing the
easiness of using Chatbot as ChatGPT.
ChatGPT has developed rabidly, and these developments have come up with several models such
as ChatGPT-1, GPT-2, GPT-3 and the latest is ChatGPT-4 (Brady et al., 2023; Cascella et al. 2023;
Sallam et al. 2023). ChatGPT-4 is considered the latest model, which has been reported to be the
best model for translation (Jiao et al. 2023). ChatGPT versions especially GPT-3 “have garnered
significant attention and have the potential to revolutionize a wide range of language-related tasks”
(Brady et al., 2023, p.5). Furthermore, ChatGPT is accused of plagiarism, however, they stressed
‘ChatGPT-generated texts are based on published literature, and citation practice is encouraged in
scholarly publishing’ (Brady et al., 2023, p.10). However, the problem is who cites ChatGPT, a
dangerous phenomenon for ethics and scholarship in general.
As for the performance of ChatGPT, it was reported that ChatGPT can perform competitively
better than human adjudicators. As reported by Kung et al. (2023, p. 4) in a study protocol, two
physician adjudicators, who were blinded to each other, utilize criterial data for accuracy,
concordance, and insight. When ChatGPT was fed with the same criterial data, it performed better
than the two adjudicators, thus demarcating “high internal concordance”. Figure 1 showcases the
United States Medical Licensing Examination (USML) utilized by human adjudicators and
ChatGPT.
4
Figure 1: Workflow for sourcing, encoding, and adjudicating (adopted from Kung et al.’s (20023: 4),
abbreviations: QC = quality control; MCSA-NJ = multiple choice single, answer without forced justification; MCSA-
J = multiple choice single answer with forced justification; OE = open-ended question format. These abbreviations
are Kung et al.’s)
In this complex task, ChatGPT performs better than human. This experiment reveals that ChatGPT
has the potential for learning in medical education provided that it is trained on the tasks it is
intended to do.
2.1. Technology and Translation
The development in technology leads to more advancement in the life of translators. The
translation process involves “all the processes and methods used torender and/or transfer the
meaning of the source language text into the target languageas closely, completely and accurately
as possible” (Ghazala, 1995, p.). technology in translation begins be employing CAT tools, which
has different types such as ‘Machine Translation (MT), Machine Aided Translation (MAT),
Machine Aided/Assisted Human Translation (MAHT), Human Aided/Assisted Machine
Translation (HAMT), Fully Automatic High-Quality (Machine) Translation (FAHQT/FAHQMT)
and Computer Aided/Assisted Translation (CAT)’ (Quah, 2006, p. 6). Technology integrates with
all the fields of human life to support humans and to facilitate their work. Palacz (2003, p.4) defines
CAT tools as “Computer-Assisted Translation software comes into play and offers the opportunity
to meet these demands. Powerful computer technology can enhance the uniquely human abilities
of translators by coupling them with raw computing power”. They are further defined as “any type
of computerized tool that translators use to help them conduct their jobs”. There are particular
programs produced to support translators in their job such as; Trados Workbench, Wordfast,
memoQ, DéjàVuX, SDLX, Star Transit, MultiTrans, Similis, MetaTexis, OmegaT (Bowker and
Ciro, 2019, p. 184).
However, the advent of Artificial Intelligence (AI) has affected translation greatly, as a direct way
to support human to fulfil their jobs in less time and efforts. But does AI going to support human
or to replace human in doing different tasks as translation? According to AlAfnan et al. (2020, p.
61) AI “refers to intelligent devices and software that can reason, absorb, gather knowledge,
interact, control, and distinguish between objects”. As AI development is very huge, the lay man
cannot imagine the huge development in AI technology. AI interaction with translators is active.
5
Some researchers (Kirov & Malamin, 2022) refers to AI as ‘one of the major technologies leading
the digital transformation and the fourth industrial revolution 1, and machine–human–computer
symbiosis is a major feature of the phenomena, such as virtual reality, computer games,
smartphone apps, smart watches, and information searches through voice recognition, etc.’ (Kirov
& Malamin, 2022, p.2). AI development leads to several Deep Learning Models (DLMs) including
ChatGPT. ChatGPT is one of the AI DLMs which provide best result to translation industry.
As a result of all these AI developments, ChatGPT comes to the surface. We will try to test its
ability to translate a free poetry text.
3. Study design
3.1. Approach
This research is qualitative. It adopts the comparative and analytical approach.
3.2. Procedure
The study took place in three stages. The first stage was the data collection, in which we selected
the Arabic free poetry text. The second stage is the translation of the text: first, the text was
translated by the authors themselves, and second it was translated by ChatGPT. The third stage is
data analysis in which we analytically compared both translations, viz. the human translation and
ChatGPT translation. As a result of the appearance of different versions of ChatGPT. The version
used in translating this text is ChatGPT 3.5 as it is the easiest to be accessed. In the same time, it
is the most common to be used.
3.3. Research Sample
The text selected to be translated by human and ChatGPT is some excerpts and poems from a
collection of poems in Arabic by Najwa Kinana Wahib, entitled ‘A Dream on the Corners of
Nostalgia’ ‘Hulm 'ala Nasiat Aljabin’. The text involves the title of the book Item 1. some titles of
poems items 2-5, from acknowledgements items 6-8, and some excerpts from the poems entitled
in items 2-5. ChatGPT translation of these texts are given in Appendix A.
4. Results
The results of this study are tabulated in Table 1, which presents the verses of the poems in
Arabic Language, human translation, and ChatGPT translation.
Table 1: Arabic text, Human Translation and ChatGPT translation
Item
No.
Arabic text
Human translation
ChatGPT translation
1
A Dream on the Threshold of
Nostalgia
‘A Dream on the Corner of
Longing.’
6
2
A Longing Fit
‘Nostalgia Episode’
3
A Prayer on the Shore of the
Dream
‘Prayer on the Shore of Dream’,
4
A Tale of a Woman with a Letter
The story of a woman and a letter
5
From the Evenings’ Inspirations
Inspired by the Evenings
6
I dedicate this to anyone glimpsing
the shadow of the dream
Dedicated to all who caught a
glimpse of the dream's shadow,
7
Lightening a star in his sky
And scattered a star in its sky...
8
..
To anyone planting a thorny tree in
his way..
To all who planted a thorn in their
path...
9
.
But he has a heart traveling on foot
without destination
But he possesses a heart that travels
on two feet to an unknown
destination.
10
.
Its shells, with my pen, its solid
and hard rocks
Its shells, even my pen, its steadfast
and stubborn rocks
11
and my corny looks
And my features, its shells
12
Enough! You will break my pride!
Enough! You will break the scales
of my ladder
13
She is distinguished by her lively
fabulous characteristics.
It is characterized by precise
description and magnificent vivid
imagery.
14
Either elicited by symbol, and
event or a consensual issue
Either deduction through a symbol,
event, or consensus-backed issue,
5
that overcomes its flaws
forgiving its stylistic flaws
16
...
Or seduction through disparaging
religious and worldly symbols
It's either the method of inference
through symbols, events, or issues
that have a consensus forgiving its
linguistic slips, or it .....
17
Arguefor the sake of arguing
dissent to know’
7
5. Discussion
ChatGPT has a conversational nature. It focuses on dialog and conversational ways of contact and
communication with humans. It is easily used; it starts by telling it to do any command or prompt,
for example; ‘Translate the following English text into Arabic’. Then it will do so. To facilitate
the analysis of the data collected in the experiment, the translation generated by ChatGPT will be
analyzed in one category, namely lexical problems.
As Table 1 shows, the first title 'Hulm 'ala Nasiat Aljabin' was translated by
ChatGPT as ‘A Dream on the Corner of Longing’. We ask ChatGPT to translate the title another
time, the generated translation is ‘Dream on the Edge of Longing (Figure 2).
18
I wanted to write to you .. but in
my voyage, I lost my way. So, I
returned back
And how I wished to write to you...
but, in a moment of travel, I got lost...
then hesitated. And how I wished to
write to you... but I lost my way in a
moment of departure...
9
I could not find letters..having the
shape of a pen
Then I retreated. And now I can't find
letters
20
Fit the space of the heart and
satisfies its arrogance
Worthy of the heart's space... or
satisfying the pride
21
The space of the papers .. Or the
ink .. to suit the pen ….
Of the paper's expanse. Nor ink...
fitting for the pen's form.
22
Simulate in its features the cheeks
of the roses paralleling in its smell
the gardens of basil
In his features, he resembles the
rose's cheeks,
and in his scent, he competes with
the gardens of lavender.
23
Farther than its writer closer to its
inspirer
And farther than its writer... and
closer than its inspirer,
24
It is a moment of permanent
revelation …
It's a moment of eternal
confession…
25
The heart signs, referring to a date
of
Where the heart's hands point to the
time,
26
?azan for silence.
The call of silence.
8
Figure 2: Screenshot of the translation generated by ChatGPT
ChatGPT has the ability to translate texts and to generate more translations for the same text. The
same title translated for the third time and produced the following translation ‘Dream at the corner
of Yearning’, as shown in Figure 3 below.
Figure 3: A third translation generated by ChatGPT
Although the three generated translations are somehow accepted, the human translation of the title
‘Dream on the Corners of Nostalgia’, is more touching and more attractive. As we are dealing with
a literary text, both sense and meaning are important in the translation produced.
The second title is Nawbat Hanian , translated by ChatGPT as ‘Nostalgia Episode’, while
human translate it as ‘A Longing Fit’. Although the title was translated by ChatGPT may be
accepted, it does not perfectly match the source text; the word Nostalgia has perfectly matched
Hanian However, the word ‘Episode’ Nawbat does not perfectly match the Arabic word ‘’.
It is better if it is translated as ‘Fit’. Additionally, the third title was translated
by ChatGPT as ‘Prayer on the Shore of Dreams’. This translation is similar (but not identical) to
human translation. The fourth and fifth titles and were translated by
human as ‘A Story of a Woman with a Letter’, and ‘From the Evenings’ Inspirations’, respectively,
and ChatGPT translated both as ‘A tale of a woman and a letter’ and ‘Inspired by the Evenings’,
respectively. Although both translations in both human and ChatGPT are somehow acceptable,
9
there is an aspect worth shedding light on. In Arabic, ‘’ can mean ‘and’ and ‘with’, the latter of
which is what is intended in the Arabic text . This was not grasped by ChatGPT.
In the excerpts taken from the acknowledgements of the book under consideration, item 6
.. was translated by human as ‘I dedicate this to anyone glimpsing the shadow of
the dream’ while ChatGPT translated as ‘Dedicated to all who caught a glimpse of the dream's
shadow’. ChatGPT used an adjectival expression, while human used a verbal one. Item 7
..was translated by human as ‘lightening a star in his sky’ while ChatGPT translated it as
‘And scattered a star in its sky...’. Human translation is more appropriate and reflects the sense of
the poet. But ChatGPT translation is far from this sense, using the word ‘scattered’ because ‘a star’
can not be ‘scattered’. Item 8 .. was translated by human as ‘To anyone
planting a thorny tree in his way..’, ChatGPT translated it as ‘To all who planted a thorn in their
path...’. Again, ChatGPT translation is far from the sense of the SL text. The author intends to say
‘even those who plant thorny trees, their efforts will not go in vein’ because one day they will
harvest fruits and flowers from this thorny tree.
In the excepts from the poems, we find that item 9 (from a
Prayer on the shore of dream) was translated by human as ‘But he has a heart traveling on foot
without destination’ and by ChatGPT as ‘But he possesses a heart that travels on two feet to an
unknown destination’. ChatGPT translation is inappropriate because it used the verb ‘possesses’
with ‘heart’. A heart cannot be ‘possessed’ or not possessed by human, because it is simply there;
living creature has a heart. The translations item 10 by both
human and ChatGPT are acceptable. Item 11 was translated by human as‘and my
corny looks’ while ChatGPT translated it as ‘And my features, its shells’. Translating as
‘features’ by ChatGPT is not appropriate because the intended meaning in the SL text is ‘looks’
not ‘features’; ‘looks’ is an outer thing while ‘features’ are an inner thing. Item 12
was translated by ChatGPT as ‘Enough! You will break the scales of my ladder’. This
translation is not appropriate, because it does not convey the intended meaning as it is in the SL
but it will be better if the translation was without the word ‘ladder’, as in the human translation of
this sentence as ‘Enough! You will break my pride!
In item 13, ChatGPT translation of the first sentence used the pronoun ‘it’ but the description is
about a woman who said ‘Enough! You will break my pride!’ The correct pronoun here is ‘she’,
which is not grasped by ChatGPT. The word , it is one word but translated into English by
ChatGPT as ‘magnificent vivid imagery’, while what is intended is ‘characteristics’. In item 14,
the repetition of the linking words like ‘or’ in ‘or consensus-backed issue forgiving its stylistic
flaws, or seduction through insulting religious and worldly symbols, or’ weakens the meaning. In
Item 17 ‘’ was translated by ChatGPT as ‘dissent to know’.’ However, it was translated
by human as ‘Arguefor the sake of arguing’ a proverb is translated by another proverb in the TL.
This happens due to the use of word for word translation by ChatGPT.
In item 18, there is mismatch between a word and its meaning. The word ‘translated by
ChatGPT into ‘hesitated’ does not reflect the intended meaning. However, human translation for
that word is ‘returned back’. The sentence ’,was translated by ChatGPT as
‘Nor ink... fitting for the pen's form...’. The translation here is word for word. The human
10
translation here is ‘Or the ink ... to suit the pen ….’. In item 19, ‘ was
translated by human as ‘I could not find letters.. having the shape of a pen’ while ChatGPT
translated it as ‘Then I retreated. And now I can't find letters’. In item 22, the word ‘’, which
refers to a plant was translated by humans as ‘basil’ but it is was translated by ChatGPT as
‘lavender’ which does not reflect the intended plant. Finally, in item 26 was translated
by human as ‘?azan for silence’ while ChatGPT translated as ‘The call of silence’. ChatGPT
translation is not appropriate because ‘’ cannot be translated as ‘call’. This is because the word
‘’ is used only in a prayer context, silence is an attribute of prayer. During prayer, speech is
prohibited, and silence here is perhaps the same silence practiced during prayer. Human translated
it as ‘?azan’ by transliteration technique.
In brief, this work presents a preliminary study of ChatGPTas a modern way to translate Arabic
free poetry texts into English. We find that ChatGPT provides a good and competitive product
while translating literary texts. Although the text selected is a free poetry/verse, which is much
easier than formal verse, there is still a need for improving ChatGPT translation to overcome the
problems encountered. There are serious errors in ChatGPT translation resulted mainly from word-
for-word technique, which is not appropriate in poetry translation. Thus, the role of the human
translator is to review the translation for the emotional and sensational parts and to make sure if
there is a mismatch of particular words and to make sure of the meaning as a whole.
6. Conclusion
In brief, we noticed that grammatical errors are lesser than lexical errors. There are words with
more than one meaning. This paper aims to investigate the problems encountered by translators
when they use ChatGPT in translating free poetry texts. Is ChatGPT usage sufficient to be used by
translators? Or there is a need for human translators to translate or to post edit texts after translating
texts by ChatGPT. The data are derived by different pieces from Arabic collection of poems by
Najwa Kinana Wahib, named 'Hulm 'ala Nasiat Aljabin'. Human translation seems to adopt
blended types of strategies: adaptation, recreation, omission, addition, transliteration. However,
ChatGPT seems to adopt word-for-word, addition (more than omission, literal translation, which
are not appropriate for free poetry translation. Human role in translation is needed to aid ChatGPT,
especially while translating literary texts such as poetry. In this respect, some researchers such as
Kim (2023, p. 2) stressed that “any new ideas generated by ChatGPT should be validated through
actual experiments and their results should be verified by humans”. For him there is still a need
for human in the translation process. Human may pre edit or post edit the texts. That is to say,
human role will integrate with machine till they reach perfection in their products. However, there
are other researchers as (Brady et al., 2023, p.6) who stress ‘GPT is a versatile tool that has been
designed to perform various language-based tasks such as text generation, question answering, and
translation’. Here, machine can replace humans in fulfilling different tasks.
References
AlAfnan M. et al. (2023) ChatGPT as an Educational Tool: Opportunities, Challenges, and
Recommendations for Communication, Business Writing, and Composition Courses. In
Journal of Artificial Intelligence and Technology, 2023, 3(pp. 60-68)
https://doi.org/10.37965/jait.2023.0184
11
Bowker, L. & Ciro, J. B. (2019). Machine Translation and Global Research: Towards Improved
Machine Translation Literacy in the Scholarly Community. UK: Emerald Publishing
Limited.
Brady et al.(2023) ChatGPT and a new academic reality: Artificial Intelligence-written research
papers and the ethics of the large language models in scholarly publishing. JASIST Journal
of the Association for Information Science and Technology.
https://doi.org/10.1002/asi.24750
Cascella, et al (2023) Evaluating the feasibility of ChatGPT in healthcare: an analysis of multiple
clinical and research scenarios. J Med Syst. https://doi.org/10.1007/s10916-023-01925-4
Catford, J.C. (1965). A Linguistic Theory of Translation. London: Oxford University Press.
Dergaa, I. et. al. (2023). From Human Writing to Artificial Intelligence Generated Text:
Examining the Prospects and potential threats of ChatGPT in Academic Writing. Biology
of Sport. 40. (615-622). 10.5114/biolsport.2023.125623.
Fitria, T. N. (2021). The Use Technology Based on Artificial Intelligence in English Teaching and Learning).
ELT Echo : The Journal of English Language Teaching in Foreign Language Context, 6(2), Article 2.
https://doi.org/10.24235/eltecho.v6i2.9299
Flynn, P. (2004). Skopos theory: An ethnographic enquiry. Perspectives-Studies In Translatology. 12(4). 270–
285.
Ghazala, H. (1995). Translation as Problems and Solutions. ELGA.
Grace, K., Salvatier, J., Dafoe, A., Zhang, B., & Evans, O. (2018). When will AI exceed human
performance? Evidence from AI experts. Journal of Artificial Intelligence Research. 62,
729-754.
Jiao, W., Wang, W., Huang, J. T., Wang, X., & Tu, Z. (2023). Is ChatGPT a good translator? A
preliminary study. arXiv preprint arXiv:2301.08745, 1(10).
Kim, S.G (2023) Using ChatGPT for language editing in scientific articles. Maxillofacial Plastic
and Reconstructive Surgery (MPRS). Volume 45, article number 13, (2023).
https://doi.org/10.1186/s40902-023-00381-x
Kirov, V. & Malamin, B. (2022). Are Translators Afraid of Artificial Intelligence? Societies. 12,
70. https://doi.org/10.3390/soc12020070.
Geertz, C. (1973). The Interpretation of Cultures: Selected Readings. New York: Basic.
Conway, K. (2012). A Conceptual and Empirical Approach to Cultural Translation. Translation
Studies. 5 (3): 264–279.
Kung, J. E., Marshall, C., Gauthier, C., Gonzalez, T. A., & Jackson III, J. B. (2023). Evaluating
ChatGPT performance on the orthopaedic in-training examination. JBJS Open Access, 8(3),
e23.
Newmark, P. (1988). A Textbook of Translation. New York, London: Prentice Hall.
Nida, A. (1964). Toward a Science of Translating. Leiden: E. J. Brill, 1964.
Palacz, B. (2003). A Comparative Study of CAT tools (MAHT workbenches) with Translation
Memory Components. Adam Mickiewicz University, The School of English. Poznan.
Poland.
Sallam et.al. (2023). Almanac - retrieval-augmented language models for clinical medicine. NEJM
AI 1(2). https://doi.org/10.1056/aioa2300068
12
Shormani, M. (2020). Does culture translate? Evidence from translating proverbs. Babel, John
Benjamins, 66(6), 902 – 927.
Songsiengchai, S. et al. (2023). Leveraging Artificial Intelligence (AI): Chat GPT for Effective
English Language Learning among Thai Students. English Language Teaching. 16(11): doi:
10.5539/elt.v16n11p68 URL: https://doi.org/10.5539/elt.v16n11p68.
Zohery, M. (2023). ChatGPT in Academic Writing and Publishing: A Comprehensive Guide. In
Zohery, M. Artificial Intelligence in Academia, Research and Science: ChatGPT as a Case
Study. (pp.10-61). Achtago Publishing.
Appendix A The chosen texts and their translation by ChatGPT
‘A Dream on the Corner of Longing.’
‘Dream on the Edge of Longing.’
Another translation for
‘Dream at the Corner of Yearning.’
13
.....
..
‘Nostalgia Episode’
There, at the shore of days...
Upon the warm sands,
I sneak every dusk...
Heart bare, soul stripped.
14
.
...
..
.
.
‘And dreams still soak through.
... My grandfather hasn't been a sailor for a long time, and I don't recall
my father being one either. But I am the daughter of the sea, and its salty water is my
ink
.. and my features, its shells,
15
And my pen, its steadfast and stubborn rocks.
I've learned to love the storms and to master the art of struggling with
monsters.. and to roam and wander in the fields of life,
And to be a sailor.. not just a visitor in a port.’
...
.
.
And in complete happiness,
Removing the lifebuoy from his chest.
Despite my intense fear of it and the darkness it harbors,
And tales of bloodthirsty monsters lurking and lying in its depths,
Still, it has always been the chalice of my inspiration and the abyss I love
to extract pearls from the darkness of its caves.. so I am fully convinced and have faith
That literature devoid of ignorant literature.’
16
.. ‘ ‘ ..
.
..
ChatGPT:
Here's the translation of the poem into English:
‘Prayer on the Shore of Dream...
‘The profession of the writer is not a golden bracelet’...
And his pen may not be a silver necklace.
But he possesses a heart that travels on two feet to nowhere...
He wanders aimlessly, casting his soul into the expanse of the sea,
Unconcerned with a compass or even an insurance policy... or even a return ticket.
He takes the shortest path towards complete misery.’
17
...
..
...
..
From the inspiration of evenings,
And this evening, and every evening...
I am seized by an intense desire to catch up with the sun's rays,
And to cling to its rays...
Perhaps to discover the sea it departs to at twilight,
So that I may draw from it the longing to return every morning.
18
.
..
.
.. ..
..
‘And lost it to the footsteps of the path.
Rather, a heart that treads with its feet upon scorching sands... stretching
along the shoreline... its particles gathered from
every Arab horizon,
so they embrace, supplicating to their Creator... reciting the prayer of unity...
echoing the hymns of hope... and kissing the homeland's forehead.
... Then it dips them into the salty sea waters...
feeling its past wounds and accumulated scars.’
19
...
.
..
.
..
‘And the wound that hasn't healed yet, he feels a tremor...
In which the scent of cedar, coffee, wheat, palm trees,
and willows permeates,
causing the sun to blush in embarrassment, shedding a final tear
from her sleepy eyes... then throwing her shawl over her temple.
As long as I stood on dry land, and as long as the seas
resided there.
So after a long wait, I finally realized...’
20
Literature is not weighed in kilograms nor measured in metric units. Literature is
Literature is measured by an ascending tremor on richer scale of the soul. As you read, savor,
and your soul quiver, Mr. Richter shout: ‘Enough! You will break the scales of my ladder’
‘It is characterized by precise description and magnificent vivid imagery that transports
you to the heart of the scene and doesn't leave you as a mere observer... a watcher... a
reader... And it distinguishes itself by avoiding the shortcomings of inference and
temptation that many Arab literary figures have fallen into.’
21
.
Either resorting to symbolism, event, or issue that has consensual forgiveness for its linguistic
slips, or seduction through disparaging religious and worldly symbols... or transgressing against
what people universally agree upon under the guise of ‘dissent to know’... or delving into the realm
of explicit sexuality, which holds a magnetic allure for people, oscillating between praising and
criticizing! A seduction that ‘forgives’ the shallowness of writing and the laxity of language...
22
..
..
..
‘Dedicated to all who caught a glimpse of the dream's shadow,
And scattered a star in its sky...
And planted a smile in its field...
To all who planted a thorn in their path...’
‘And how I wished to write to you... but I
23
Lost my way in a moment of departure...
Then I retreated. And now I can't find letters
Worthy of the heart's space... or satisfying the pride
Of the paper's expanse. Nor ink... fitting for the pen's form.’
‘In his features, he resembles the rose's cheeks,
and in his scent, he competes with the gardens of lavender.
For the greatest love letters are those yet unwritten, and unsent to anyone.
Those that were deeper than their moment, and truer than their writing.’
24
‘And farther than its writer... and closer than its inspirer,
It's a moment of eternal confession,
Where the heart's hands point to the time,
The call of silence.’
