‘It’s Depend’, ‘Its Depend’, or ‘It Depends’? A portrait of ELF speaking variations

. The present study aims to contribute to the existing knowledge on individual variation in situations where English is used as a lingua franca (ELF). It explores the use of two ungrammatical constructions, ‘it's + V ø ’ (e.g., it’s depend ) and ‘its + V ø ’ (e.g., its depend ), in spoken conversations. These constructions involve an uninflected present-tense singular verb form with it’s or its as subjects, which diverge from the standard -s inflected third-person singular with the pronoun it (e.g., it depends ). This study examines the distribution of the variants of the ‘it's + V ø ’ and ‘its + V ø ’ constructions and the factors influencing the construction usage from an ELF environment. The data was from transcribed spoken texts of multilingual Asian English speakers in the Asian Corpus of English (ACE). The study identified 29 constructions from the idiolects of 20 individual speakers (4 males and 16 females) of various age groups, nationalities, and L1s. The analysis compared the individual usage of the ungrammatical constructions, contracted form it’s in grammatical sentences, and present-tense markers ( -s or zero) of singular verbs with singular subjects ( he, she, it, this, and that ). The findings suggest that the use of the ‘it's+ V ø ’ and ‘its + V ø ’ constructions of individual speakers is influenced by the priming of it’s , chunking of idiomatic it's , and variability of the English inflectional system. The speakers also exhibited interchangeability between the constructions and the present-tense markers of singular verbs with singular subjects.


INTRODUCTION
Language variations among individual speakers have gained significant attention since the beginning of the new millennium (Griffiths & Soruç, 2021, p. 340).This increased focus is partly due to the availability of diverse data sets, which highlights the crucial role of individual differences in language description.The study of individual variation becomes particularly fascinating in the context of English as a Lingua Franca (ELF).In these situations, most individual languages are treated as secondary languages and are affected by the rapidly changing multilingual environment.As a result, the differences between the individual languages become more pronounced, and the description of the average becomes less informative.Vetchinnikova and Hiltunen (2020) argue the importance of observing individual variability in ELF usage.By observing ten individuals in online ELF environments, they found that the differences in the use of contracted it's and uncontracted forms it is between individual and communal levels are significant.Vetchinnikova and Hiltunen conclude that the communal use of ELF emerges from the preferences of individual ELF speakers.In other words, group languages, like the communal use of ELF, are just general ideas made from the natural differences in how individuals use language (Mauranen, 2018, p. 116).It is acceptable to rely on whatever seems to work in interaction, whether diverging from standard language or mixing languages (Ranta, 2018, p. 245).This way, natural and spontaneous norms arise to safeguard mutual intelligibility in ELF usage (Mauranen, 2012, pp. 6-8).
Research has revealed that individuals have unique speech habits as part of their lingua franca, which is apparent, for example, in their n-gram profiles (Wright, 2017), collocational preference (Mollin, 2009), lexicogrammatical patterns (Hall et al., 2017), preference for contracted and uncontracted forms (Vetchinnikova & Hiltunen, 2020), and grammar (Dąbrowska, 2012).These studies suggest that language is a system of interacting idiolects (The Five Graces Group, 2009), making variability a key feature of lingua franca usage (Osimk-Teasdale, 2018).Variability is in different forms of English, including those from the outer circle, such as Brunei English (Deterding & Salbrina, 2013), and from the expanding circle, such as Indonesian English (Endarto, 2020).These variations reflect individual usage of the varieties in the lingua franca context locally and internationally.Examples of observed idiolectal variation in spoken and written everyday conversations include the use of it's mean and its mean instead of the standard it means (Amnah, 2016;Deterding, 2010;English in Brunei, 2020;Hellokitten, 2021).
In contributing to current research on individual variations in ELF and language change in general, the present paper looks at undiscussed variants, which the author named the 'it's + Vø' and 'its + Vø' constructions.These constructions involve uninflected singular present-tense verb form with it's or its as subjects (e.g., it's depend or its depend, instead of it depends).Zero inflection for singular subjects in present-tense forms is neither a recent nor unpredictable linguistic phenomenon.Trudgill (1998) suggests that this simplification emerged as a contact feature due to non-native speakers in Norwich during the 16th century.However, no research has investigated how and why the singular morpheme -s is shifted to the subject it, making it pronounced as [ɪts] and written as either it's or its.When someone writes its instead of it's, it could suggest an attempt to economize while still conveying the standard meaning of it's ('it is' or 'it has').Thus, the rationale for discussing these constructions is the change of meaning in it's and its in the constructions.These constructions represent non-standard forms that could signal ongoing changes in the English language.Investigating the constructions can provide insights into how language evolves and new grammatical patterns emerge.Therefore, the current study aims to examine the construction usage and distribution.Additionally, it seeks to explore the factors influencing the use of the constructions in an ELF environment.
Previously, in Poulisse's (1999, p. 227) study, the use of it's look instead of it looks and it's belong instead of it belongs by L2 English speakers was regarded as nonhabitual and unintended slips of the tongue.Poulisse's (1999, pp. 82-89) data collection method consisted of an audio recording of 45 subjects of low, intermediate, and advanced English proficiency levels performing four controlled tasks which took 45 minutes per subject.This means that the recording session was performed once per subject and the use of it's look and it's belong was recorded during that one session.
However, a recent study indicated that the recurring use of it's look instead of it looks and it's mean instead of it means by an advanced L2-English speaker points towards habitual, not a slip of the tongue (Mohd Yusoff et al., 2019). Mohd Yusoff et al. (2019, pp. 165-166) employed a data collection method that involved conducting one-hour oral interview sessions with the same participant every week over 20 weeks.Their findings revealed that the participant employed the construction once during Week 1 of the interview and used it more than once in each recording session from Week 2 to Week 9.This pattern suggests that the subject's use of it's look and it's mean was attributable to speech habits.Analyzing the distinctions in methodology and findings between Poulisse's (1999) andMohd Yusoff et al.'s (2019) studies, it is conceivable that Poulisse's subjects' tongue slips may have been manifestations of speech habits.
The possibility of misunderstanding during spoken communication or raising mutual intelligibility should be considered when examining the use of the 'it's + Vø' and 'its + Vø' constructions.Thus, the present study also aims to explore speakers' preference for using it's (or its in informal writing without the apostrophe) and uninflected present-tense verb form for singular subjects.Bybee (2002) and Labov (1972) propose viewing contractions, such as it's, as instances of morphosyntactic or phonological reduction.This reduction is through cognitive processes, particularly frequency effects and chunking.Bybee and Scheibman (1999) argue that frequent word combinations become processing units through chunking, leading to changes in their constituent structure.These structural changes relate to shifts in meaning or pragmatic functions (Cheng et al., 2009).Vetchinnikova and Hiltunen (2020) further this understanding by asserting that chunking systematically affects individual speakers, influencing their preference for using contracted forms, such as it's over uncontracted forms like it is.Their data shows that chunking contributes to the formation of individual chunk repertoires that include both it's and it is, with significant variability observed across speakers.This variation highlights the personalized nature of language processing and usage, shaped by cognitive mechanisms like chunking.
Priming is another cognitive factor influencing language use (Pickering & Garrod, 2017, p. 173).Priming operates massively at a non-conscious or automatic level, manifesting as an unconscious tendency to repeat what one has 'It's Depend', 'Its Depend', or 'It Depends'? A Zayani Zainal Abidin Portrait of ELF Speaking Variations comprehended or produced.According to Pickering and Garrod (2017), priming plays a significant role in routinization, where initial impromptu expressions with specific meanings become conventionalized over time.While priming is studied over short timescale, such as within a single conversation, its effects can extend over longer periods, as evidenced by studies spanning weeks (Kaschak et al., 2011).Pickering and Garrod argue that priming can induce permanent changes across diverse groups, including adults, children, and native and non-native speakers, operating at different levels of linguistic representation.Even structures deemed ungrammatical can prime, resulting in heightened acceptability after exposure (Luka & Barsalou, 2005).Relevant corpus studies on priming effects include Mair (2017) and Barth and Kapatsinski (2017), who identified instances where the spoken occurrence of contracted form wanna ('want to') was primed by previous occurrences of wanna and gonna.Vetchinnikova and Hiltunen (2020, p. 226) also found similar results in their study, where the previous instance of it's increased the likelihood of using the contracted form.These studies collectively highlight the pervasive and nuanced nature of priming in shaping language use.
The objectives of this study are to investigate: 1) the correlation between the use of the 'it's + Vø' and 'its + Vø' constructions, and it's in idiomatic sentences (e.g., It's always raining, It's a nice house); 2) the relationship between the use of the 'it's + Vø' and 'its + Vø' constructions and the nonstandard use of uninflected present-tense singular verb form (Vø) with singular subjects (he, she, it, this, and that); and 3) the association between the use of the 'it's + Vø' and 'its + Vø' constructions and the standard use of -s inflected present-tense singular verb form with singular subjects (he, she, it, this, and that).

Data
The data used in this study consists of transcribed spoken conversations sourced from the Asian Corpus of English (ACE, 2024).ACE, a 1-million-word corpus, captures naturally occurring interactions among ELF speakers, predominantly with Asian language backgrounds.All ACE participants demonstrated a high proficiency in English.
ACE has employed data collection teams in diverse locations for six years across Asia, including Brunei, China, Hong Kong, Japan, Malaysia, the Philippines, Singapore, Taiwan, and Vietnam.It is essential to clarify that although the teams were present in these countries, their objective was not to collect data on local English varieties.Instead, their focus was gathering authentic instances of ELF usage by multilingual speakers in each location (Kirkpatrick, 2016, p. 226).ACE enriches the dataset by providing metadata on speakers, and details, such as age, gender, nationality, first language (L1), education, and occupation, available after each transcription.

Data extraction and preparation
The data collection approach for this study involves internally searching for sentences containing the patterns 'it's + X + Vø' and 'its + X + Vø', where X is an adverb that may be present or absent in the constructions, and Vø represents the uninflected present-tense singular verb form.The initial process began with inserting the keywords it's and its into the ACE online interface.Subsequently, the corpus concordance unveiled instances of these constructions and pinpointed the specific filenames containing these instances.Figure 1 shows a sample of the concordance of the keyword it's in ACE.

Figure 1. Concordance for 'it's' in ACE
Identifying the occurrences of 'it's + X + Vø' and 'its + X + Vø' demanded a comprehensive understanding of how these constructions functioned within their respective contexts.The construction usage in ACE is exemplified in ( 1)-( 3).The markup in the examples extracted for the present study (e.g., speaker identification 'S1', 'S2', and 'S3', utterance pause '(.)') is part of ACE annotation.
(1) S1: (.) the <7> the </7> that's your decision it's not depend on the contract at all (in file 'MS_LE_con_11') (2) S3: but it's belong to the: <creak> oh ah </creak> because i live in the newer bukit panjang side (in file 'SG_ED_con_5') (3) S2: so when its come to this problem i think this gonna be less surreal to them than the kids from the rich family (in file 'PH_ED_sed_recounting lesson') In the context of the conversations, it becomes evident that it's not depend in Example (1) corresponds to the standard it does not depend, it's belong in Example (2) corresponds to it belongs, and its come in Example (3) corresponds to it comes.However, it is uncertain why its in its come does not have the apostrophe as it's in Examples ( 1) and ( 2).Because the apostrophe is not pronounced in speech, it could suggest that the file transcriber for 'PH_ED_sed_recounting lesson' was economical.This suggestion will be discussed in detail in the results and discussion section.
Given that the primary focus of this study is to scrutinize the distribution of the use of it's and its with the uninflected present-tense singular verb form, adverbs present in these constructions will not be factored into the data analysis.Consequently, instances like it's not depend in Example (1) will be analyzed as it's depend, hence the 'it's + Vø' and 'its + Vø' constructions.Instances such as it's seem in Example (4) will be excluded from the findings due to potential phonological ambiguity.'It's Depend', 'Its Depend', or 'It Depends'? A Zayani Zainal Abidin Portrait of ELF Speaking Variations 6 (4) S2: i'm <8>not too</8> sure it's seem that i ha-have not come across this p-position lecturers here (in file 'HK_ED_int_helpdesk2') After identifying the instances of 'it's + Vø' and 'its + Vø' constructions and the filenames containing the examples from the concordance, the next step was to extract the text from selected filenames.The initial process is viewing the filename at https://corpus.eduhk.hk/ace/view/#/browse.Then, the stages include copying the text, pasting it into a text file or Word, and renaming these new documents according to the filenames.
The new text files or Word documents are to search the speakers who used the 'it's + Vø' and 'its + Vø' constructions.After identifying the users of the constructions, their age group, gender, nationality, and L1s were extracted from the metadata provided at the end of each file (see Figure 2).The calculation of words spoken of the selected speakers was done by manually removing the markup.The frequency of speakers of construction usage, age group, gender, nationality, L1, and word count were all extracted into an Excel sheet.

Figure 2. Metadata of speakers in file 'HK_ED_con_conference briefing2'
In total, 20 individuals used the 'it's + Vø' and 'its + Vø' constructions in ACE.These speakers are identified by the initials of their gender, nationality, and participant number.For example, FChi1 denotes the first female Chinese speaker.The metadata of the speakers who employed these constructions is in Table 1.Three additional datasets were essential for analyzing the factors influencing the 20 speakers' use of the 'it's + Vø' and 'its + Vø' constructions.The first dataset was the speakers' use of it's in idiomatic structures, where it serves as the subject, be as the copula, and the clause as the complement (e.g., It's just a different way of drawing things, It's really cute).The second dataset was the present-tense marker -s of singular verbs for singular subjects he, she, it, this, and that.The third dataset was the zero present-tense markers of singular verbs for singular subjects he, she, it, this, and that.These datasets can be searched and identified from the new text files or Word documents, and subsequently extracted into the Excel sheet with manual coding.The collected data was classified into priming and chunking.
For the priming category, the total occurrences of idiomatic it's structures and present-tense zero markers for each speaker were in two parts.The first part was the total occurrences of idiomatic it's structures and present-tense zero marker before the initial occurrence of the 'it's + Vø' or 'its + Vø' constructions.This assessment aligns with the methodologies employed by Barth and Kapatsinski (2017) and Mair (2017) on priming effects mentioned earlier.The second part was the total occurrences of idiomatic it's structures and present-tense zero markers after the initial occurrence of the constructions.This two-part classification enables comparing priming proportions for the 'it's + Vø' and 'its + Vø' constructions.The extraction of these classifications into the Excel sheet was done with manual coding.'It's Depend', 'Its Depend', or 'It Depends'? A Zayani Zainal Abidin Portrait of ELF Speaking Variations For the chunking category, it's was regarded as a part of a chunk and operationalized as a semi-fixed, three-word n-gram (e.g., it's okay) [1] .The analysis for chunking will concentrate on calculating the use of the most prevalent threegrams involving idiomatic it's by each speaker.For example, a speaker might use contracted forms it's like or it's not multiple times in their conversation instead of uncontracted forms it is like or it is not.The contracted forms were classified as three-word chunks.The extraction of the chunks into the Excel sheet was done with manual coding.Then, the frequency of these chunks was calculated accordingly.This approach aims to discern patterns related to the use of it's within specific linguistic chunks and assess its recurrence in each speaker's speech.______________ [1] The string it's is treated here as consisting of two words.

Variation between individuals
Table 2 shows 'it's + Vø' and 'its + Vø' constructions usage by the 20 individual speakers.A total of 29 variants of the constructions were found in ACE.These variants used 12 uninflected present-tense singular verbs (belong, come, depend, guarantee, have, help, make, mean, keep, look, take, and want).Notably, recurring instances include verbs belong (three occurrences), depend (seven occurrences), and mean (10 occurrences).
It is crucial to highlight that this study does not dismiss instances where speakers use these constructions only once.The rationale behind this approach stems from the uncertainty surrounding whether the singular use in a speaker's conversation during a single meeting is a mere slip of the tongue or indicative of speech habit.As previously mentioned, Mohd Yusoff et al. (2019, pp. 165-166) presented evidence suggesting that the chronic use of phrases like it's look instead of it looks and it's mean instead of it means initially occurred as a single instance in the first week of a one-hour interview.However, the phrase usage recurred from the second until the ninth week.It suggests that the speakers' one-time use of the 'it's + Vø' and 'its + Vø' constructions, as outlined in Table 2, could be attributed to speech habits or slips of the tongue.In contrast, the repeated use of these constructions by FBur1, FBur2, MMal1, MMal2, FTha3, and FVie5 in their conversations is deemed more likely due to speech habits.
As stated in the introduction section, the author suggested that the transcriber for the file 'PH_ED_sed_recounting lesson' was economical for excluding the apostrophe in its in its come.The collected data has identified that speakers FBur2 and FVie3 were from the same file 'PH_ED_sed_recounting_lesson'.For FBur2, there were two instances where it's was transcribed as its in its such and its depend.There were 15 it's usage by FBur2, including the two cases of it's depend.This makes the total count of it's for FBur2 17.For FVie3, there was one instance where it's was transcribed as its in its come.There were 10 it's usage by FVie3.Therefore, the total count of it's for FVie3 was 11.It confirmed that the transcriber for the file missed adding the apostrophe for its depend by FBur2 and its come by FVie3.
A statistic measure, logDice, was conducted to determine the attraction or strength between the two words in each variant.LogDice is only based on the frequency of the node (since there should have been an apostrophe for its depend by FBur2 and its come by FVie3, the node for all variants is thus it's), the collocate (belong, come, depend, guarantee, have, help, make, mean, keep, look, take, and want), and the frequency of the co-occurrence of the node and collocate.High logDice scores identify words that are strongly attracted to each other.Low scores identify words with low attraction that are together only by chance.The median logDice score is 11.97.Scores above this median indicate strongly associated collocations of the variants produced by FBur1, FBur2, MCam1, FChi1, MMal1, FTha1, FTha2, MTha1, FVie1, FVie2, and FVie4.
Subsequent sections will delve into the remaining data sets, exploring factors that may influence the use of these variants in greater detail.The results and discussion will be based on the order of the three research objectives.

Results and Discussion Related to Objective 1
This section discusses the correlation between the use of the 'it's + Vø' and 'its + Vø' constructions and it's.The following findings demonstrate the effect of it's usage on the 'it's + Vø' and 'its + Vø' constructions.
Vetchinnikova and Hiltunen (2020, p. 219) report that previous instance of the contracted form increases the likelihood of using it again.The priming effect is also 'It's Depend', 'Its Depend', or 'It Depends'? A Zayani Zainal Abidin Portrait of ELF Speaking Variations reported by Barth and Kapatsinski (2017) and Mair (2017) who identified instances where the occurrence of wanna ('want to') was primed by previous occurrences of wanna and gonna.Even in this case, one of the main factors causing the use of the 'it's + Vø' and 'its + Vø' constructions is certainly the priming of it's.
In the current study, all 20 speakers used it's in their conversations.Three speakers, FTha2, FVie1, and FVie4, only used it's once which was in their 'it's + Vø' construction usage.Figure 3 illustrates the proportions of it's occurring before and after the first occurrence of the 'it's + Vø' and 'its + Vø' constructions.This observation implies a potential priming effect of it's on the use of the constructions.

Figure 3. Proportions of 'it's'
To examine the influence of priming associated with it's on the use of 'it's + Vø' and 'its + Vø' constructions by the 20 speakers, a one-tailed dependent t-test was conducted.The results of the test indicate that the likelihood of using 'it's + Vø' and 'its + Vø' constructions (M = -0.65,SD = 0.34) increases following the prior occurrence of it's, t(19) = -1.90,p = .036.Vetchinnikova and Hiltunen (2020, p. 219) also report that individuals who frequently use contractions tend to do so even in the absence of a priming context.This suggests an entrenched habit of using contractions.Supporting this observation, the present study finds that all 20 speakers preferred the contracted form it's over the uncontracted it is.Table 3 provides the raw frequency of it's and it is usage by the 20 speakers.This finding aligns with the concept of routinization due to priming, as highlighted by Pickering and Garrod (2017) Normalized Frequency Speakers Before After According to Vetchinnikova and Hiltunen (2020, p. 221), chunking is similar to priming.They report that chunking seems to increase the likelihood of the contracted form it's across individual speakers.This holds at least for those who prefer to contract in general.In this case, the other main factor causing the use of the 'it's + Vø' and 'its + Vø' constructions is the chunking of it's.
The present study analyzed 228 instances of idiomatic it's structures produced by 17 speakers (excluding FTha2,FVie1,and FVie4).The examination of all idiomatic it's structures in the data revealed that the syntactic structures belonged only to the copular category.There were no idiomatic it's structures in the progressive, such as it's going, found in the data.Due to the limited data, a cautious approach to non-compositional processing was adopted.This entailed considering a sequence of words as a chunk if it constituted a fixed three-gram and occurred at least twice in each conversation.
12 out of the 17 speakers produced three-grams involving it's which occurred at least twice in each conversation.Table 4 showcases the most frequent three-gram chunks of idiomatic it's structures for each speaker.This finding highlights the diverse repertoires of personal chunks developed by individuals.It also emphasizes that chunking systematically influences the variation in three-grams involving it's, mirroring the variability observed in the variants across individuals in Table 2.This observation suggests that the chunking effect of idiomatic it's contributes to the 'It 's Depend', 'Its Depend', or 'It Depends'? A Zayani Zainal Abidin Portrait of ELF Speaking Variations speakers' use of the 'it's + Vø' and 'its + Vø' constructions.Furthermore, it is possible to tentatively hypothesize that the 'it's + Vø' and 'its + Vø' constructions usage may be associated with the copular category, which is conducive to chunk formation.

Results and Discussion Related to Objective 2
This section discusses the relationship between the use of 'it's + Vø' and 'its + Vø' constructions and the nonstandard use of uninflected present-tense singular verb form (Vø) with singular subjects (he, she, it, this, and that).The following findings demonstrate the effect of zero inflection usage on the 'it's + Vø' and 'its + Vø' constructions.
Zero inflection for singular subjects in present-tense forms is neither a recent nor unpredictable linguistic phenomenon.Trudgill (1998) suggests that this simplification emerged as a contact feature due to non-native speakers in Norwich during the 16th century.This phenomenon is particularly prominent in varieties that have experienced extensive linguistic contact.These varieties include highcontact L1 varieties, such as Aboriginal English and Urban African American Vernacular English, and indigenized L2 varieties such as Hong Kong English, Malaysian English, and Pure Fiji English (Kortmann et al., 2020).ELF, being a contact language, commonly exhibits zero marking of the third-person singular (Breiteneder, 2009).Even in this study, there are instances zero inflection for singular subjects in present-tense forms.
In the current study, 13 out of 20 speakers used zero marking in present-tense singular verbs with singular subjects (he, she, it, this, and that) in their conversations.Figure 4 illustrates the proportions of zero inflection occurring before and after the first instances of the 'it's + Vø' and 'its + Vø' constructions.It is observed that only six out of the 13 speakers used the zero marking before the first instances of the 'it's + Vø' and 'its + Vø' constructions.

Figure 4. Proportions of uninflected present-tense verbs with singular subjects
To determine whether the priming effect of zero inflection of present-tense singular verb form with singular subjects (he, she, it, this, and that) contributes to the use of the 'it's + Vø' and 'its + Vø' constructions by the 20 speakers, a one-tailed dependent t-test was conducted.The results of the test indicate that the prior use of zero inflection does not increase the likelihood of using the 'it's + Vø' and 'its + Vø' constructions (M = 0.45, SD = 0.36), t(19) = 1.25, p = .113.

Results and Discussion Related to Objective 3
This section explores the relationship between the use of 'it's + Vø' and 'its + Vø' constructions and the standard use of the -s inflected present-tense singular verb form with singular subjects (he, she, it, this, and that) The use of the -s inflection for third-person singular verbs varies widely across many non-standard varieties of English (Trudgill, 1990).Calle-Martín and Romero-Barranco (2017, p. 82) note that not all speakers of a given variety consistently adopt the inflected form.Even in ELF interactions, the use of third-person presenttense markers showcases the diversity in preferences among English speakers (Jenkins et al., 2011).
The present study finds that among the 13 speakers who used zero marking in their conversations, seven demonstrated interchangeabilities between the -s and zero inflection in present-tense singular verb forms.Figure 5 illustrates the interchangeability demonstrated by speakers FBur2,FMal1,FTha3,FVie3,FVie5,FVie6,and FVie7. For instance,FVie5 (in file 'VN_LE_con_culture') showcased this interchangeability by using the 'it's + Vø' construction and employing both -s and zero inflection of the present-tense singular verb forms with singular subjects (she and it), as demonstrated in examples ( 5)-( 7) respectively.
(5) it's mean that now okay i i just got a bachelor degree okay i'm (.) working as a high school teacher (6) <6> yeah she doesn't have time </6> yeah yeah (7) <3> yeah </3> in vietnam yeah it become popular now

Figure 5. Variation of present-tense singular -s and zero inflection with singular subjects
The use of the 'it's + Vø' construction and present-tense -s and zero inflection can extend to involve the same verb within the same conversation.FChi2, who exclusively used zero inflection throughout her conversation (in file 'MS_ED_con_6'), demonstrated the interchangeable use of the 'it's + Vø' construction and zero inflection with the same verb have, as exemplified in Examples ( 8) and (9).A similar pattern also emerged with the verb mean in the conversation of FTha3 (in file 'MS_LE_con_6').FTha3 employed both -s and zero inflection with the verb mean, as illustrated in Examples ( 10) and ( 11).These findings highlight the speakers' flexibility and interchangeability in using these structures within the same discourse.This offers intriguing insights into the dynamic nature of individual language use.(8) it's only have the like english to french and the other language (FChi2) (9) very very beautiful place and it's: summer but it have the snow in the mountains (FChi2) (10) one hour and a half so it's mean in-include going to the toilet (FTha3) (11) because <L1th>haa {five}</L1th> it mean er <L1th>haa {five}</L1th> thai in thai means <2>five</2> (FTha3) Conversely, FVie2, who exclusively employed -s inflection with singular subjects throughout her entire conversation (in file 'MS_PB_con_1'), exhibited the interchangeability of using the 'it's + Vø' construction and -s marking with the same verb take, as exemplified in Examples ( 12) and ( 13).This interchangeability is similarly evident with the verb look used by FVie7 (in file 'VN_LE_con_pho restaurant') as illustrated in Examples ( 14) and ( 15).FVie7 produced 10 instances of -s marking and one of zero inflection.These findings showcase the potential of these structures to be used within the same discourse, even when a speaker predominantly uses a specific form of inflection.
(12) <6>it's</6> take time and then <7>they don't feel don't feel comfortable</7> with waiting (FVie2) (13) oh my goodness <5>it takes time</5> (FVie2) (14) oh the same the same the (.) it big it's look quite big (FVie7) (15) do you know why she looks so young?(FVie7) The interchangeability observed from these findings highlights that, despite the use of nonstandard zero inflection and the 'it's + Vø' and 'its + Vø' constructions, the speakers were aware of the standard form.This suggests that the use of these constructions is not random but rather influenced systematically by the priming of it's, chunking of idiomatic it's, and variability of the present-tense inflection marker.Despite the use of nonstandard structures of zero inflection and the constructions, the speakers maintained effective communication flow.This interaction between standard/grammatical and non-standard/ungrammatical forms within an individual's linguistic framework is driven by various dynamic forces, including the complex interplay of the speaker's cognitive processes, prior language experiences, and social motivations (Vetchinnikova, 2017).

CONCLUSION
One of the potential factors influencing language use and change is variation among individuals.Scholars generally agree that social, cognitive, and linguistic factors interact to produce linguistic patterns.These factors should be apparent in each individual.Despite all language users having similar cognitive abilities such as chunking, using an analogy, categorizing, and generalizing to new situations, the linguistic patterns they produce can differ considerably based on the input they 'It's Depend', 'Its Depend', or 'It Depends'? A Zayani Zainal Abidin Portrait of ELF Speaking Variations receive.This becomes particularly intriguing in ELF environments where people have greater language exposure diversity than in monolingual settings.
The present study delved into the usage and distribution of two ungrammatical constructions, namely 'it's + Vø' and 'its + Vø', in spoken ELF conversations.It examined the effect of priming and chunking on the variations observed in these constructions across 20 individuals.The main analysis involved a comprehensive examination of each speaker's usage of the 'it's + Vø' and 'its + Vø' constructions, the contracted form it's, and present-tense markers (-s or zero) of singular verbs with singular subjects (he, she, it, this, and that).The most noteworthy findings in the study suggest that the use of the 'it's + Vø' and 'its + Vø' constructions by individual speakers are prompted by the priming effect of it's (p = .036)and chunking effect of idiomatic it's, and the variability in the inflectional system in individual Englishes.
A limitation of this study is the uncertainty regarding whether the isolated instances of the constructions produced by the speakers result from speech errors or tongue slips rather than established speech habits.As aforementioned, this study considered instances of the 'it's + Vø' and 'its + Vø' constructions, even when used only once.This decision was influenced by findings from Mohd Yusoff et al. (2019, p. 165-166), suggesting that recurrent use of expressions like it's look instead of it looks and it's mean instead of it means initially manifested as a single occurrence in the first week of a one-hour interview but later recurred from the second until the ninth week.This implies that the speakers' one-time usage of the 'it's + Vø' and 'its + Vø' constructions, as depicted in Table 2, could potentially be linked to speech habits.Another limitation lies in the lack of variation in the syntactic structure of idiomatic it's as the chunking effect of it's, which was exclusively from the copular category.To enhance the depth of the study, the researcher could have incorporated spoken ELF data from an additional corpus, allowing for a comparison of the variants of the constructions and the influence of different syntactic structures on the chunking of idiomatic it's (e.g., extraposition, progressive).This could serve as a potential direction for future research, alongside the possibility of a longitudinal study to investigate the evolving distribution patterns of the 'it's + Vø' and 'its + Vø' constructions over time.
Additionally, according to Kaschak and Glenberg (2004) and Luka and Barsalou (2005), there is a possibility that exposure to ungrammatical structures, such as the 'it's + Vø' and 'its + Vø' constructions, can lead to increased acceptability of these structures.Since the present study does not extend to investigating how the 20 individual speakers were exposed to these constructions, a qualitative study could explore this aspect, providing further insights for future research.

Table 3 .
Speakers' use of 'it's' and 'it is'

Table 4 .
Most frequent three-grams involving 'it's' illustrate how the use of the standard -s inflection can still lead to the occurrence of the 'it's + Vø' and 'its + Vø' constructions.