Overcoming error fossilization in academic writing: strategies for Saudi EFL learners to move beyond the plateau

Abstract

The current study explores error fossilization in the academic writing of Saudi English as a Foreign Language (EFL) undergraduate learners. A manual textual analysis approach, employing corpus content analysis on writing across diverse genres including argumentative, expository, narrative, and descriptive essays was conducted to discover the fossilized errors. The results showed that certain errors, such as idiomatic language usage and subject-verb agreement, appear to be fossilized, while others diminish over time. The study highlights specific instances of error fossilization, including challenges related to articles, idiomatic language, plural markers, negative markers, and prepositions. These errors are attributed to factors such as linguistic differences between Arabic and English, limited exposure to authentic language materials, and cultural influences. This study provides valuable insights into the complexities of error fossilization among Saudi EFL undergraduates. By understanding the factors contributed to these errors and their underlying causes, educators can develop tailored approaches to language instruction, ultimately enhancing English proficiency in similar contexts.

Introduction

Writing skills have received paramount concern of pedagogical attention in the realm of Second Language Acquisition (SLA) due to its complexity task of mastering the complex rules and structures, including grammar, syntax, punctuation, and vocabulary. From then on, interlanguage fossilization has garnered significant attention in the sphere of SLA. It often singles a deficiency in language learning where a particular language component is acquired inaccurately or erroneously. The notion of fossilization, as described by Selinker (1972), occurs when learners' language acquisition process is influenced by their native language (L1), input from the target language (TL), and various cognitive, social, and affective factors. Once fossilized, these errors resist quick correction or replacement with the correct language form. The root cause of fossilization in SLA is frequently attributed to negative transfer from the learner's mother tongue, leading to persistent systematic errors that persistently affect language users (Selinker, 1978).

Evidence from a number of experimental studies has established the issue of poor English writing among English as a Foreign Language (EFL) students as a common challenge faced by educators and learners alike. Several factors contribute to this issue, including language proficiency levels, lack of exposure to authentic English writing models, limited vocabulary, cultural differences in writing conventions, and insufficient writing practice (Al-Dulaimi, 2022). Since then, experimental studies on error analysis particularly in academic writing by EFL Saudi learners, have received considerable attention (e.g., Alqefari, 2023; Altowity, 2021; Khaleghi et al., 2024; Khan, 2022). For instance, Altowity (2021), based on error analysis, investigated the academic writing of Saudi undergraduate EFL learners. The study revealed a phenomenon where learners' persistent errors become ingrained and resistant to correction despite ongoing instruction and feedback. Khan (2022) examined errors in the writing of undergraduate Saudi EFL learners and found that both groups of the participants committed different grammatical errors, such as ‘wrong word order’ and wrong use of article’.

Moreover, EFL teacher should discern students' linguistic errors and comprehend the instances in which these errors arise. Ellis and Barkhuizen (2005) postulated that a comprehensive examination of students' errors, whether systematic or sporadic, justifies examination, aiming to discern underlying causes and contributing factors. Aligned with the perspective of Divas and Hedari’s, we propose more intriguing and nuanced investigation. It would be beneficial to exclude beginner learners from investigating grammatical errors. It is likely to exhibit a higher frequency of errors by beginner, which might lead to a potential skewing the analysis. In this study, we focus on intermediate to advanced leaners, hoping to offer deeper insight into the causes of error fossilization by English for florigen language learners (EFL).

Although the above studies have greatly enriched our knowledge of the specific factors contributing to error analysis in academic writing among Saudi EFL learners, the overall error fossilization of undergraduate Saudi EFL in terms of investigating the potential impact of fossilization seems to be completely neglected. The study investigates if fossilization is a cause of the prevalence of errors in the writing of even intermediate to advanced Saudi EFL learners. It depends on Khatter's (2019) study in analyzing the most common fossilization errors as the two studies were conducted in the same context: Saudi Arabia.

In the current study, we aim to contribute fresh perspectives on the potential impact of fossilization on the persistence of writing errors among intermediate to advanced EFL Saudi learners. The chosen methodology for this investigation entails an in-depth examination of writing proficiency over a span of twelve weeks, as this particular skill holds paramount importance for Saudi university-level EFL learners. The aim is to delve into the dynamics of language fossilization and its potential influence on persistent writing errors among learners at an intermediate to advanced proficiency level. The research questions are as follows:

• i. Can the errors in the academic writing of Saudi EFL undergraduates be attributed to the phenomenon of fossilization?

• ii. What categories of errors are found in the writing of Saudi undergraduate EFL learners?

Literature review

Theoretical perspectives on interlanguage fossilization

Interlanguage fossilization is a common linguistic phenomenon in second language acquisition. Popularized by Selinker (1972), Interlanguage works as a unique intermediate stage of linguistic, development of a learner where the native language (NL) interacts with the target language (TL), forming the third language. Interlanguage consists of elements from both the native and target languages, along with unique elements absent in either. These distinct non-target-like components emerge during spontaneous and significant exchanges. Though, Selinker (2006) estimates that only 15% of adults successfully learn a second language, the process of fossilization is not global. It changes with contexts and learner type, so the term ‘a fossilized speaker’ is not only visually but also inaccurate. Conversely, Brown (2001) states that EFL learners go through quasi-systematic growth to reach a whole competence level in the TL. Effective and cognitive feedback for a successful Interlanguage development has the potential to reverse fossilization, just like ‘cognization’ or the process of freezing matter at low temperatures.

According to Selinker (2006), a majority of second language learners (SLLs) lag behind in achieving the target language competence due to a substantial disparity between their native language and the foreign or target language (TL). The reasons for these disparities include the learners' poor retention, lack of focus, and insufficient understanding of the language, as observed by both learners and teachers. Selinker posits that the inadequacy in students' grasp of the second language, combined with the ongoing impact of their native language, results in writing that is incoherent and incohesive, marked by syntactic and morphological errors.

Nevertheless, Dina (2017) asserted that most fossilized errors among post-graduate EFL students were that of an article, number, clause, and writing style. However, it is essential to note that error analysis merely deals with data rather than language acquisition. Therefore, it can neither explain the origins of these errors nor provide remedial measures. Moreover, identifying these errors and restating learners' intended messages is purely subjective and factually objective (Cook, 1993).

Corder (1981) articulated the value of recognizing the learners' interlanguage and the teachers’ interpretation of the errors. Alternatively, he suggested the teachers ask the learners for interpretations (authoritative reconstruction) or use plausible interpretations, but their knowledge of the learner and their L1 is essential to connect with them. The findings on learner errors and the relationship between grammar and lexis led Salem (2007) to conclude that the errors made by some advanced learners could not be labelled as either grammatical or lexical. Thus, understanding the differences between the two could lead to a pattern of linguistic awareness, while the error analysis could help teachers provide error-centered feedback.

Factors contributing to interlanguage fossilization

Furthermore, Brown (2001) added that while most errors were visible or overt, some were covert or grammatically correct but uninterpretable in each context, which might produce misleading information about the learners' intrinsic knowledge. Han (2009) agreed with several researchers that the learning patterns of the first language continue to present challenges to the sequential learning of the second language. So, when the language learner is fluent in the second language, despite making frequent syntactical and pronunciation errors, they need to be more motivated to achieve perfect language use. Researchers have established the consistent use of non-target features in the SLLs' interlanguage as the main factor for fossilization, but these L2 errors are mere ‘candidates for fossilization.’ In isolation, these errors do not substantiate a robust argument for the concept of fossilization. Primarily, these errors are characterized as inferential, contrasting with those that are directly observable or exhibit temporary stability, as delineated by Han (2004). The phenomenon of stabilization leading to fossilization is contingent upon three pivotal factors: a temporary hiatus in L2 acquisition, a restructuring of the interlanguage, and a protracted gap in the interlanguage developmental process. This extended hiatus in the interlanguage developmental trajectory serves as a precursor to the eventual occurrence of fossilization, as expounded upon by Han (2004, 2009), Selinker and Han (1996).

Han (2004) argued that fossilization is more local or relates to some features of the interlanguage systems rather than global or the entire interlanguage system since fossilization is more of an ongoing process than a finished product. According to Han, fossilization mainly occurs due to learners’ maturational constraints and interference with their native language without other variables. Spolsky and Hult (2008) claimed that interlanguage transfer is key to foreign language acquisition and can have both negative and positive impacts. While positive transfer eases the learning process, negative transfer, also called interference, leads to erroneous and inappropriate target language use. Newport (1990) added that language learners with different maturational states do not consistently achieve the same output even in similar conditions. The study found some constraints that ensure success in language acquisition do not remain constant for a lifetime. Also, both first language learning and second language learning are influenced by the learner’s maturational stage. To illustrate this, learners who begin acquiring a language during an early maturational state usually perform significantly higher proficiency that those who stat later. However, while language learning progress remains consistent throughout childhood, the rate of improvement tends to flatten in adulthood. All in all, these studies highlighted the complex interplay between maturational constraints, interlanguage transfer, and fossilization in second language acquisition. They emphasized the importance of considering individual learner differences and developmental factors when examining language learning processes and outcomes.

Cognitive development

This section discusses various aspects of cognitive development and its impact on language learning, including short-term memory and information rage, information processing capacity, age and language acquisition, cognitive and motivational factors, and interlanguage and fossilization.

The literature on cognitive development states that adults have better short-term memory than children, and their capacity to store information increases over maturation (Kail, 1984). It is still being determined whether this is an increase in capacity to store information or to use that capacity effectively. Newport (1988) alternatively suggested that successful language acquisition thrives on a young child's limited information processing capacity as opposed to adults. Hyltenstam and Abrahamsson (2003) summed up the constraints of second language acquisition by indicating that learners who begin at an early age have a similar linguistic achievement as those native learners. However, for those who start the process at an older stage, the road to acquiring native-level proficiency becomes more difficult due to age, cognitive impairments, motivational levels, or formalized training. Krashen (1982) suggested that adults fare poorly in language learning because their functional abilities interfere with learning strategies conducive to second language learning.

The studies examined various aspects of language acquisition and error fossilization. They explored the differences in language learning abilities between adults and children, the impact of age on second language acquisition, the concept of Interlanguage and its contribution to fossilization, as well as methods to address fossilized errors in language learning. These research studies act as an initial direction for the current study in examining the persistence of developmental errors among both monolingual and bilingual learners and propose innovative techniques, such as the audio-articulation method, to correct fossilized pronunciation mistakes. Additionally, it emphasizes the ongoing nature of language learning and the absence of a definitive "end-stage" of language proficiency.

Again, this study is the first of to investigate if fossilization is a cause of the prevalence of errors in the writing of even intermediate to advanced Saudi EFL learners. The approach in this study is to examine spontaneous writing over twelve weeks as this skill is most needed by Saudi university-level EFL learners and is also the one in which they receive a substantial number of years of exposure and training.

Methodology

Study design

This study employs a manual textual analysis approach, conducting corpus content analysis on writing samples from English major students. The collected samples encompass various writing genres, such as argumentative, expository, narrative, and descriptive essays. One notable benefit of manual textual analysis is its ability to sidestep the challenges posed by automated correction software, which often fall short in achieving a high level of accuracy (Humphreys & Wang, 2018). Another merit of using this approach, stated by Huckin (2003) that “content analysis is the identifying, quantifying and analyzing of specific words, phrases concepts or body of texts with the aim of uncovering some underlying thematic or rhetorical pattern running though these texts” (p. 14). This approach would allow us with its flexibility to detect the most common error in our study samples.

Setting and participants

This research was carried out at the Department of English and Translation College of Science and Arts in Mithnab, Qassim University, Saudi Arabia. The Department grants a bachelor's degree in English language and translation where the students complete a four-year program by passing eight levels. The Department offers a wide range of courses, including several essential disciplines in the English language, namely linguistics, applied linguistics, translation, literature, and English language skills.

The study samples assigned 32 intermediates to advanced English primary EFL students at Qassim University, Saudi Arabia. It was carried out during the Spring Semester of 2022 for twelve weeks, which consisted of controlled intervention in the form of teacher inputs that used an adapted version of the Coding Rubric of Swan and Smith (2001). All the 32 participants were strictly Arabic-speaking males, had EFL learning experience of at least nine years during intermediate and high school, and scored between intermediate and advanced levels on the Standardized Test of English Proficiency (STEP) which they are required to take to enter higher education. The median age of the group fell to 21.6 years. The STEP is an accredited English language proficiency test similar to TOEFL and IELTS but simpler. It consists of approximately 100 multiple-choice questions across four sections: Reading Comprehension (40%), Sentence Structure and Grammar (30%), Listening Skills (20%), and Writing Analysis (10%). These sections collectively measure proficiency in English from various aspects, as well as in testing. The samples were convenience-based as the researchers had prior teaching and research experience at Qassim University, and the study sample had been taught one TEFL course by them in the previous semesters. Requisite permission from the Qassim University was obtained for research purposes, and open-ended consent from the participants was also obtained, which allowed the researchers to requisition their writing strictly for study purposes.

Procedures

The participants were encouraged to engage in self-diagnosis of errors based on a performance rubric they maintained during this period. This required them to keep a record of the errors marked by the researchers on their essays writing outputs. The error categories in this performance rubric replicated the Coding Rubric of Swan and Smith (2001). The researchers ensured that the participants were not conscious of their writing being evaluated. This was done by randomly selecting the samples out of the more than eleven writing activities they were assigned during the semester. Over twelve weeks, the data produced a collection of 25,639 words. Writing samples which were quantitatively analyzed using the Coding Rubric for the nature of errors encountered in the samples, and evidence of trends in persistent errors was recorded. The Coding Rubric was shared with the participants when the researchers gave corrective feedback for the first writing sample. To ensure a high interrater reliability percentage, the first researcher examined a segment of students' writing. This identical segment was then checked by the second researcher independently. Last, a comparison was conducted to check the similarity in numbers and type of errors. This interrater reliability was pertained in which a degree of 89% was reached. Such a percentage is considered a strong reliability measure.

Data collection and analysis

The researchers employed an adapted version of the Coding Rubric designed by Swan and Smith (2001) to provide written feedback on each of the samples collected from the participants. This feedback was aimed at making learners conscious of their errors through a process referred to as" “retrospective self-correction” (Steve, 2021). This approach has proven effective in enhancing learners' awareness of fossilization, thereby bringing them closer understanding the true nature of their IL. In the early stages of the study, the researchers identified nine distinct error categories in the participants' writing samples, and these categories served as the benchmarks for categorization throughout the entire study period. Table 1 provides a summary of these initial findings.

Table 1 Frequency of errors

Results and discussion

This section provides the results regarding the three research questions. Research question one is answered by examining whether the academic writing produced by Saudi EFL undergraduates can be linked to the concept of fossilization, while research question two is answered to by showcasing specific instances of error fossilization identified in the participants' writing.

RQ1: can the errors in the academic writing of Saudi EFL undergraduates be attributed to the phenomenon of fossilization?

The first question in this study sought to determine whether errors in the academic writing of Saudi EFL undergraduates can be attributed to the fossilization phenomenon. Nine errors were identified in the writing samples. Verb agreement, verb usage, and idiomatic errors appeared to be fossilized, while the occurrence of other error types decreased over time. Specifically, plural marker errors reduced from 49 occurrences in phase 1 to 21 occurrences due to the implementation of a self-correction rubric. Similarly, negative markers, preposition errors, and word formation errors decreased from 11, 46, and 19 occurrences in phase 1–5, 17, and 2 occurrences, respectively. According to previous studies in error analysis (Han, 2004, 2009; Selinker & Han, 1996), idiomatic language usage, verb-related errors, and subject-verb agreement constitute the three specific types of errors identified as being prone to fossilization among the nine categorized errors observed in students' writing. Moreover, Le (2023) revealed that at the top 3 most common types are wrong verb forms, fragment sentences, and subject-verb agreement. A possible explanation for these results would relate to various contributing factors that led to the persistence of certain errors and the likelihood of their fossilization in the academic writing of Saudi EFL undergraduates.

To start with the first fossilized error, subject-verb agreement is a crucial part of English grammar and is essential for clear communication (Jäger et al., 2020). EFL students often struggle with matching the subject and verb correctly in their writing. Our findings align with those of Yusuf et al. (2021) and Ramzan et al. (2023), who found that omission errors were the most frequent type of error made by EFL students. For example, Yusuf et al. (2021) found that verb-agreement errors constituted 61.04% of all error.

First, differences in syntax between Standard Arabic and English. Participants may not have exhibited a change in error trends in English sentence formation due to avoidance of grammatical features perceived as burdensome, as Schachter (1974) suggested.

Moreover, inadequate teaching materials, outdated curriculum, insufficient teacher training, and demotivation could contribute to students' poor grasp of subject-verb agreement. Further, Mukattash (1986) believed that grammar is considered a complex area for second language learners, especially for Arabic speakers learning English. It is posited that this complexity is particularly evident in aspects and tenses. Emphasis is placed on factors related to the teaching process, methods or techniques used, the absence of an optimal learning environment, insufficient learner motivation, and shortcomings in the curriculum. An effective method for teaching subject-verb agreement is explicit instruction. This approach involves clearly explaining the rules and providing concrete examples to illustrate them. Further, insufficient or ineffective feedback from instructors can lead to the persistence of verb-agreement. Feedback needs to be timely, specific, and constructive.

There are, however, other possible explanation for the challenges in grammar acquisition that could be attributed to cognitive development. As learners progress through different stages of cognitive development, their ability to comprehend and internalize complex grammatical structures may vary. Factors such as cognitive maturity, working memory capacity, and processing speed could influence how effectively learners grasp and apply grammatical rules.

RQ2: What categories of errors are found in the writing of Saudi undergraduate EFL learners?

Six categories of errors are found in the Saudi EFL learners' writings. They are article errors, idiomatic errors, verb and subject-verb agreement, plural marker, negative marker, and proposition. These six types will be defined with examples from the students' writings.

Article errors

Students misused the definite article "a" as in this example, " I miss a second chance for taking a exam…". Using the indefinite article ‘a’ is problematic for the Arabic speaker since Arabic does not have an equivalent article. Arabic follows nunation, i.e., the addition of ‘un’ to the end of the noun; however, it indicates the indefiniteness of the noun. Consequently, the process of overcorrection is frequent where English articles are to be applied by Arabic users. In this study, retrospective self-correction proved effective in helping the participants check the fossilized errors with wrong use of articles dropping to 34 occurrences in the fourth sample from 138 in the first sample. Group discussions taken up later revealed that most learners are not conscious of the differences in article use between English and Arabic and tend to apply the Arabic grammar rule in English articles. It was interesting to note that the participants did not falter in the use of the definite article, which also exists in Arabic, and conceded during the group discussion sessions that had they been shown the differences in article use between English and Arabic as soon as they were introduced to English grammar, they may not have applied the specific generalized L1 grammar rule in L2. According to Bakri (2023), the influence of Arabic interference and English grammar rules on the students' article errors, emphasizing the significant role of L1 interference in shaping the acquisition of English articles among ESL learners.

Idiomatic errors

The analysis of samples in this study highlights the challenge faced by Arabic learners of English in using idiomatic expressions. They produced, "In the other hand…" instead of on the other hand. Their tendency to translate materials back and forth between the target language (English) and their mother tongue (Arabic) leads to awkward and incorrect usage of idioms in their writing. This phenomenon indicates a lack of proficiency in understanding and applying idiomatic language in English, resulting in fossilized error trends persisting throughout the study samples.

One possible reason for the occurrence of idiomatic language errors lies in the difficulty Arabic learners face in acquiring English prepositions. For example, some learners may directly translate English prepositions like 'in' or 'on' into Arabic equivalents, such as 'ﻓﻲ'. This can lead to confusion when selecting the appropriate preposition in idiomatic expressions like "on the other hand," contributing to errors in their usage.

Furthermore, research by Mutar (2016) supports the finding that Arabic learners continue to struggle with using English idioms, suggesting that this is an ongoing challenge. Every language possesses its own set of idiomatic expressions and attempting to apply them without a deep understanding of their cultural and linguistic context can prove difficult for learners who are not yet proficient in the target language.

Moreover, previous studies by Albelihi(2022), Barcroft (2007), and Salem (2007) have identified various types of errors made by Arabic EFL learners, including grammatical, lexical, and word combination errors. Salem (2007) specifically categorized errors made by advanced Arab EFL learners as either grammatical or lexical. Understanding the distinction between these types of errors can help learners develop linguistic awareness, while error analysis can enable teachers to provide targeted feedback to improve students' language competence.

In light of these findings, it is imperative for English teachers to increase students' awareness of these errors and provide effective solutions to prevent them from internalizing such mistakes. Hamzah (2012) suggests that teachers play a crucial role in guiding learners to overcome these challenges and develop proficiency in using idiomatic language appropriately.

Subject-verb agreement (SVO)

The subject-verb agreement was a major casualty in the samples examined in this study. These examples are found in their writings, "Their father like to drive them to school"; "we goed to the supermarket near the coast that the car break down". For various reasons, but primarily due to the different syntax in Standard Arabic, the participants did not exhibit a change in error trends in sentence formation in English. This was a surprising finding and can certainly not be attributed to MT transference as Arabic has a more complex and varied word order with colloquial Arabic varieties preferring SVO and Standard Arabic more commonly following VSO. Discussions with the participants shed light on poor exposure to authentic English text as the reason for their depressed performance in English syntax. Further, at least nine of the participants pointed out that the books that comprised their study materials were unduly ‘easy’ and ‘unchallenging’, which caused them to lose interest in experimenting with language use.

Arabic does not have all forms of the continuous tense and in some forms, it only adds a time modifier to indicate the time of the action. This can explain the apparent fossilization of this error in the four samples examined in this study over time. When the researchers had follow-up discussions with the participants, he learned that they were most comfortable with generalizing the rule for past tense formation in English and most of the times, since their teachers (sharing the same MT) fully understood the intended meaning, they seldom pointed out the error as one. Thus, the communicative need being fulfilled, the researchers concluded, the error got fossilized over time. For the other two fossilized errors occurrence (verb & verb agreement), is attributed to the different syntax in Standard Arabic and English: the participants did not exhibit a change in error trends in sentence formation in English. This can be related to what Schachter (1974) affirmed: language learners avoid certain grammatical features of the language that they consider onerous. A lack of sufficient teaching materials, outdated curriculum, unsatisfactory teacher training, and demotivation, among other considerations, are associated with students' poor grasp of subject-verb agreement is another possible reason that led to fossilization. However, with a small sample size, caution must be applied, as the findings might not be consistent with different samples in distinct contexts (i.e., samples from different departments and universities).

The results of this study align partially with previous study's findings. They match with Hasbún (2007) who reported that while some errors occur more frequently at different levels in an unpredictable manner, mistakes involving subject omission, subject-verb agreement, and negative forms are more prevalent among beginners. Also, errors concerning the use of articles and prepositions, and incorrect verb forms, were the most common categories across all levels. They also agree with Khatter (2019) that identified that interlingual and intralingual transfer as the primary sources of the most common errors.

Plural marker

Data in this study showed remarkable improvement in the plural marker errors in the samples, with a drop from 49 to 21 occurrences. An example is "Their ancestors were sheeps-herd." Participants opined the English plural were rather ‘easy’ for them (except for the irregular forms) as their MT follows more complicated plural formation rules with the addition of the dual case. Just as in dual in Arabic, the typical rule for plural implies that you can change a noun from singular to plural form, simply by adding a suffix at the end. For the masculine words, you add ـﻮن or ـﯿﻦ (“Uon” or “Een”), while for the feminine, you simply add ـﺎت (“aat”), removing the “Taa marbouta” ة. This is called “sound masculine plurals” (“jam’ mozakkar saalim”) and “sound feminine plurals” ﺳﺎﻟﻢ ﻣﺆﻧّﺚ ﺟﻤﻊ (“jam’ mo’annath saalim”). Thus, even recording their individual errors helped the participants work on their English plurals and the errors in the early samples were not attributed to fossilization.

Negative marker

Negation can be a challenge even for some proficient L2 users of English. It is not surprising that in all the 128 writing samples, only 11 occurrences of errors were found in the earliest samples, and this number came down to 5 in the last sample. One of the examples in which students misused negation is "I do not speak no mind with them". Discussion with the participants revealed that they found the double negative usage rather fancy as it is what they often came across in the western media, such as movies they are fond of watching. Additionally, the teachers do not usually mark these errors on their writing as they ‘understand’ their limited facility with the language. This error can be surmised from this that it is not a result of fossilization, but choice, as these young learners, considerably want to ‘sound’ like native speakers whose language they come across on the media. These results are in agreement with those obtained from Sawalmeh (2013) who identifies ten common errors made by Saudi Arabic speakers in English essays, including double negatives. This rather intriguing finding could be due to cultural norms and values may also play a role in shaping language use. In some cultures, linguistic features such as double negatives may be considered acceptable or even preferred for certain rhetorical or expressive purposes.

Preposition

As per the British Council, English prepositions can be a problem for learners as the language has several prepositions with more than one function. For instance, ‘at’ can be used in as many as 18 places and others with lesser but equally challenging functions. Secondly, the choice of the preposition in English is sometimes a matter of phrasal use, i.e., some prepositions need to be learned as a phrase rather than in isolation. For instance, "The responsibility to work hard is in us" is found in their writings. Another is the expressions ‘angry with’ and ‘angry at’ are both correct but have different connotative meanings that may well have to be learned rather than acquired. In the samples examined in this study, prepositions exhibited somewhat depressed usage in the writing of EFL learners. However, with some margin notes by the researchers on the usage of the most common English prepositions (of, on, at, for, with, in, about), the output showed some improvement, with the error occurrence coming down to 17 from 46 in the first sample. Discussion with the participants showed that their awareness of the multiple uses of English prepositions is limited as they are ‘glutted’ with these as words with fixed functions with drill-like exercises in preposition use and limited opportunities for authentic language exposure. These possible sources of error could have affected by the semantic relations of these prepositions. For example, Alhammad’s (2023) findings indicate that participants encounter difficulties in learning 'on' and 'at'. This is due to the limited correspondence between the meanings of 'on' and its Arabic equivalent 'ʕala'. Additionally, Arabic lacks an equivalent preposition for 'at', further contributing to the learners' struggle in understanding and using these English prepositions.

Conclusion

In conclusion, this study delves into the phenomenon of error fossilization in the academic writing of Saudi EFL undergraduates, addressing two key research questions. Firstly, we examined whether errors in their academic writing can be linked to fossilization, revealing that certain errors, such as idiomatic language usage and subject-verb agreement, appear to be fossilized, while others decrease over time. Secondly, specific instances of error fossilization are showcased, highlighting challenges such as wrong articles, idiomatic language, plural markers, negative markers, and prepositions. These errors are attributed to various factors, including linguistic differences between Arabic and English, limited exposure to authentic language materials, and cultural influences.

In addressing these findings, it becomes evident that effective language learning strategies are essential for mitigating error fossilization. Teachers play a crucial role in raising awareness of common errors, providing targeted feedback, and creating opportunities for authentic language exposure. Moreover, addressing cultural norms and values that may influence language use can help learners navigate linguistic challenges more effectively.

Overall, this study provides valuable insights into the complexities of error fossilization in the academic writing of Saudi EFL undergraduates. By understanding the nature of these errors and their underlying causes, educators can develop more tailored approaches to language instruction, ultimately enhancing the English proficiency of learners in similar contexts.

Recommendations

This study's findings have several important implications for future practice. Speaking is a skill learned in the teacher’s presence; fossilization indicators are likely to manifest in speaking much earlier than writing. The onus to check this lies in the teacher, who needs to devise constructive strategies to give feedback. Adequate opportunities should also be created for learners to recall corrected language. Overloaded, indiscriminately planned EFL syllabus must be replaced with realistic, well-planned mini-language syllabi, which prepare learners for quality input rather than quantity inputs in English. Moreover, the study highlights the need to carefully examine the quality of the learning outcomes as exhibited in the Saudi learners’ language output. This will help identify the areas that need greater attention from a pedagogical point of view. As is evident from the study data, these learners’ weakness lies in formulation and articulation of ideas, a shortcoming that can be effectively tackled by training them to ‘think’ in English rather than translate from Arabic. Finally, the responsibility for learning should be shared between teachers and learners, with learners taking accountability for their progress.

Limitations

A number of limitations need to be noted regarding the present study. The most important limitation lies in the fact that studying error fossilization is more meaningful with cross-sectional data than using a longitudinal method. In other words, examining long-term writing data of fewer participants from amongst the sample would add to the value of the findings. This would be a fruitful area for further work. An additional uncontrolled factor is the possibility of variation in participants' language proficiency. Lastly, the study should be repeated using individual interviews that are valuable in identifying trends in language fossilization.