In utterances, what needs a linguistic repair are those that exemplify disfluency or anything that obstructs fluency in any grammatical aspect. When not initiated explicitly, repairs are determined by making a pause, asking questions, or repeating the utterance. Failure to do repair may cause breakdown in conversation [7].

A repair sequence is the process of identifying speech errors and making an actual repair in the utterance [8]. This repair sequence is analyzed by identifying who pointed out the error and who does the repair. The types of repair are as follows: a) self-initiated self-repair (the trouble source speaker indicated the error in his own utterance and repaired it himself in the same turn), b) self-initiated other-repair (the trouble source speaker indicated the error in his own utterance and was repaired by the recipient in the latter’s turn), c) other-initiated self-repair (the recipient indicated the error of the trouble source speaker and was repaired by the speaker in the latter’s turn), and d) other-initiated other-repair (the recipient indicated the error of the trouble source speaker and repaired it himself in the same turn).

Furthermore, a repair is completed by using a pattern to address the error or trouble in the conversation. According to [9]. specific patterns of repair completion are as follows: a) replacement (substitution of a word by another term in the same sentence structure), b) modification (insertion of extra word or words to provide description or details), c) abandonment (complete abortion of the sentence and resuming with a new speech, and d) reorganization (reconstruction of the sentence structure using all or most of the words in the previous utterance).

Although various researches have been conducted on how repairs are made by interlocutors [9,10]. the analysis of repair in conversations between different cultures is left unexplored. To address this gap, this study investigated the types and patterns of repair found in the conversations between a Filipino and a Japanese speaker. As the participants have different nationalities and had shown difficulty in communicating in English, it is interesting to examine how speakers from different cultures with limited English proficiency identified errors and made repairs in their conversations. The research findings are expected to shed light on how individuals maintain verbal understanding and succeed in the communication process despite their differences.

MATERIALS AND METHODS

This qualitative research aims to examine the repairs in conversations between speakers from different cultures. Specifically, this study used conversation analysis in analyzing the research corpus. Conversation analysis is a study that aims to describe, analyze, and understand talk as a basic feature of social interaction [11]. It deals with spoken forms of natural (i.e., unscripted, non- elicited) communication and may involve audio and video recordings of talks [12].

Conversation analysis was used in this research by focusing on the repairs made by the interlocutors studied. A total of ten excerpts of conversations from the videos published in the YouTube channel of Pinoy Big Brother were transcribed and used as the research corpus, and non-English terms found in the conversations were translated. The video clips were selected based on the conversations which contained speech errors that called for repair strategies from the interlocutors. In particular, the participants chosen for this study were a Japanese named Fumiya Sankai and a Filipino named Yamyam Gucong whose tandem is called “FumiYam” in the reality show. These participants are adult housemates from Pinoy Big Brother Otso, a reality television series in the Philippines which aired from November 2018 to August 2019.

In this research, the participants, Yamyam and Fumiya, were labelled as speakers A and B, respectively. Yamyam is pure Filipino who was born and raised in the Philippines, while Fumiya was born and raised in Japan and has been living in the Philippines for two years at the time of the show’s airing. Aside from their cultural differences, both speakers had shown limited English proficiency, as evident in their conversations. Additionally, Yamyam’s native tongue is Cebuano while Fumiya can speak a few Tagalog and Bisaya terms based on what he learned while staying in the country.

RESULTS AND DISCUSSION

As this study involved the analysis of intercultural conversations, the following are the types and patterns of repair found in the conversations between a Filipino and a Japanese speaker (Table 1).

The self-initiated self-repair was the most frequent type of repair which shows that speakers are conscious and active participants who can identify and repair their own errors. On the other hand, the self- initiated other-repair was the least common type of repair used by the participants (Table 2).

For the patterns of repair, repetition was the most frequent, especially since the participants are not considered fluent in English. Hence, limited English proficiency may cause an interlocutor to repeat words as an indication of one’s ongoing sentence construction in a language that he is not that familiar with. Moreover, most of Quan [13] repair patterns, which are replacement, modification, and abandonment, except for reorganization, were found in the research corpus, while other patterns of repair, such as repetition, code-switching, demonstration, phonetic alteration, elaboration, syllabication and completion, emerged as other patterns used by the participants in making conversation repairs.

Based on the research corpus, 4 types and 10 patterns of repair transpired as the repair’s interlocutors make in the conversation. In total, there are 74 types and 77 patterns of repair in the Filipino-Japanese interaction. There is a higher number of patterns as there were three instances in which one type of repair used two patterns at the same time. The trouble source in each interaction was highlighted in this section for easier identification.

In this study, the trouble source refers to the error made by the speaker which in any grammatical form creates disfluency or misunderstanding in the conversation. A repair is initiated by identifying or indicating the trouble source and is completed by giving a certain correction to the trouble source (Table 3).

(1)         Self-initiated self-repair

The self-initiated self-repair is the most common type of repair in the research corpus. This indicates that a speaker can consciously identify his own error and fix it using a certain pattern.

As shown, the speaker had difficulty in expressing something, so he repeated a part of the word while he was thinking of the complete word. Noticeably, the speaker also identified that his statement lacked details, so he inserted the word “tomorrow” to specify the time. In the latter part, the speaker had once again showed difficulty in communicating, so he resorted to repeating the same words while constructing a sentence. Finally, the speaker, in the last part, recognized the need to code switch from English to Japanese in order to express his thanks in a language that is more familiar to the Japanese recipient.

(2)         Other-initiated self-repair

In some cases, the recipient also indicates the trouble source in the speaker’s utterance and allows the speaker himself to repair his own error. In this type of repair, the recipient is either uncertain or unknowledgeable of completing the repair, which leaves him to only identify the trouble but leaves the completion of the repair to the trouble source speaker. This is the second most common type of repair in the research corpus.

Speaker B identified the trouble by asking questions to clarify the meaning of the Tagalog utterance. As speaker B raised the lack of understanding on the speaker’s utterance, speaker A himself made the repair by giving examples to explain what the trouble source word means.

(3)         Other-initiated other-repair

Other-initiated other-repair is a type of repair done when the recipient identifies the error and knows how to repair such error. In this repair, the speaker is completely unaware of the trouble in his utterance which calls for the recipient to initiate and complete the repair.

Speaker A: Nakahinga ka nang maluwag.

[I felt quite at ease.] Speaker B: Nakahinga ako nang malog!

Speaker A: Maluwag... [At ease.]    Other-initiated other-repair Phonetic alteration

Speaker A uttered the Tagalog statement spoken by speaker B. However, the trouble source speaker neither noticed or made repair in his own utterance, so speaker A corrected this error. The identification of error and the repair itself were made by the speaker who is more knowledgeable of the language as it was commonly spoken in his country, hence, he has more expertise in making repairs in such language.

(4)         Self-initiated other-repair

The self-initiated other-repair is done when the participant identifies the trouble in his speech but experiences difficulty in completing the repair, which in turn allows the recipient to complete the repair of the speaker’s trouble. This is the least common type of repair with only two occurrences in the research corpus.

Speaker A:        Mostly, uh, mostly, in the, uh, in the, in our hometown, ‘yun dun! [yes, there!]

Hirap naman neto, teka muna. [This is so difficult, wait.]

Speaker B:       Mahirap. [It’s difficult.]         Self-initiated other-repair    Abandonment

The trouble source speaker himself acknowledged the trouble, where he showed his difficulty in expressing his thoughts in the target language. He wanted to say something but because of his difficulty, speaker B completed the repair by starting a new utterance and abandoning the discussion of such topic. Hence, in these cases, the trouble was identified by speaker A but repaired in completion by speaker B.

Furthermore, patterns of repair were also analyzed in the research corpus. Repair patterns are specific strategies that the participants use in order to repair a certain error in the conversation. The following table is presented to summarize the common patterns of repair found in the research corpus (Table 4).

(1)         Repetition

Repair was made by repeating words or syllables to continue the conversation. Although repetition was not mentioned in [14]. it appeared to be the most frequent pattern in the research corpus.

• You… you… you know how to speak Bisaya? (Self-initiated self-repair)
• But how we… how we answer it?
• Mostly, uh, mostly, in the, uh, in the, in our hometown…

As shown in the first two statements, the speaker either repeats the first word of the sentence or certain words in the middle of the sentence to continue the utterance. This repetition was intermediated with pauses that indicate that the speaker was still processing the information. While in some cases, like in the last example, the words were repeated and intermediated with the filler uh to fill in the gaps of his utterance.

(2)         Code-switching

Code-switching, or the alternate use of two or more languages, was another common pattern used by the participants. It is a common strategy and repair pattern for bilingual and multilingual speakers. Just like repetition, it was not mentioned in [15]. but found out to be the second most common.

• Uh, clean… In Tagalog,
• Mostly… in our hometown, ‘yun dun! [yes, there!]

The speaker expressed his difficulty in speaking in English so he repaired his own utterance by switching to Tagalog. The speaker conveyed what he meant by using his own language since he knows that the Japanese recipient understands some Tagalog and Bisaya words.

• That’s why [I] thank you, thank Arigatou guzaimasu! [Thank you very much!]
• Ipakita sa mundo? Show it to the world…

While in some cases, the speaker intentionally switched from English to Japanese as shown in statement c and Tagalog to English in statement d not because these languages were more familiar to the speaker but because they were to the recipient. As both speakers know some words in each other’s languages, they communicated those target languages which are more familiar to the recipient in order to ensure understanding in the conversation.

(3)         Replacement

Replacement is a pattern that makes use of more accurate words as a substitute to the previously uttered terms. This is used to clarify the meaning of the message.

• Say in Bisaya, please tell me. Teach
• Fumiya, I will show you, I will share [to] you my love life…

In the statements shown, the previously mentioned verbs were replaced by other verbs that the speaker thought were more accurate. In particular, the word “tell” was replaced by “teach” while “show” was replaced by “share” to accurately convey what the speaker meant.

(4)         Demonstration

Demonstration refers to the use of sounds or gestures in order to substitute words in explaining one’s statement. This pattern shows that nonverbal cues can be useful alternatives to words.

• No, don’t call me kuya [big brother] ‘cause if you call me kuya, my knee… aray [ouch]… [holds knee while acting out pain]

The speaker wanted to say that being called a “kuya” or older brother makes him feel old, as implied by having a weak knee. In order to explain what he meant, the speaker held his knee and acted out pain while saying “aray”, a Filipino expression for pain, in order to convey what was difficult for him to express in words.

Speaker A: I’m so relieved.

Recipient: Huh? What is “relieved?”

• Like [sighs].
• Relieved is [sighs]... I think it’s release because [sighs], release your
• No… Kanina, very very pompompom [imitates heartbeat sound], diba? [A while ago,it was very nerve-wracking, right?]… and then [sighs] now.

In another example, speaker A had difficulty explaining what is “relieved,” so he instead sighed as an example to explain the word. Speaker B, however, mistook the word as release, as shown by the release of one’s breath. Speaker A corrected him by recalling an earlier event and imitating the sound of a nervous heartbeat while repeatedly tapping his chest to emphasize intensity, and ended the demonstration with a sigh to show what being “relieved” means.

(5)         Modification

Modification, as one of Zhang’s patterns (1998), is done by inserting a modifier in some part of the utterance to specify the noun, adjective, verb, or other adverb. This is done in order to provide more details in one’s utterance.

• Mostly, uh, mostly, in the, uh, in the, in our hometown…
• We… the… our relationship…

Since using the determiner “the” lacked details about his being part of his hometown and a certain relationship, the speaker inserted “our” to modify and specify whose hometown and relationship he was referring to.

(6)         Phonetic Alteration

As observed in the research corpus, phonetic alteration is done by altering certain phonetic segments or sounds in one’s speech to correct the pronunciation of words.

• Bahara na. Bahala [Leave it be.]
• Kring… Very kring… Clean. Clean. Ma-li-nis

In these statements, the Japanese speaker initially uttered the /r/ sound in the words shown whose correct pronunciation was with the /l/ sound as in bahala, clean, and malinis. All of these were corrected by repeating the word but with the correct pronunciation—pronouncing the sound /l/ instead of /r/.

Speaker B: [sings the PBB theme song] Pinoy, ikaw ay Pinoy, pinatita sa mundo…

Speaker A: Pinatita?

• … ipakita…

Speaker B: Nakagig…

• Nakaginhawa… [I felt at ease.]

While in some utterances, the local Filipino words “pinatita” was corrected as “ipakita” and “nakagig” as “nakaginhawa” to correct certain sounds, such as /k/ instead of /t/ and /n/ instead of /g/, respectively.

(7)         Elaboration

This is a repair pattern used as a speaker engages in lengthy repairs to explain the meaning or provide examples about the trouble source utterance. This is used when a speaker relies on words to give details about the misunderstood or unknown idea.

Speaker A: I’m so relieved

Speaker B: Huh? What is “relieved”?

(a)          For example, the problem is five, and then you got minus two.

The speaker explained the meaning of “relieved” by providing an example. He explained that having less problems would result to the feeling of being “relieved.”

Speaker A: Baka umutot… [I might fart]…

Speaker B: What? [laughs] Why?

(b)         Kuya [Big Brother] said that I am... strongly…If, if I’m holding my, holding my laughing, maybe I’m, I’m changing of… wind of change.

• Bad Bad air.
• Oh, Ah! Carbon dioxide!

On the explanation of “utot” or fart, one of the repair patterns used was elaboration.  The speaker explained that it was “bad air” and provided a scientific description that would elaborate what “utot” is.

(8)         Abandonment

Abandonment was used by leaving the trouble or error and restarting with a new s utterance to keep the conversation going. This is a pattern used by speakers who think that pursuing the problematic utterance is unnecessary, thus resorting to a new topic or utterance.

Speaker A: Mostly, uh, mostly, in the, uh, in the, in our hometown, ‘yun dun! [yes, there!]  Hirap naman neto, teka muna. [This is so difficult, wait.]

• [It’s difficult.]
• Mahirap, pero kaya. [It’s difficult, but manageable.]

The speaker expressed that he had difficulty in explaining something about his hometown and ended with such admission that the recipient also recognized. In this case, the message that the speaker wanted to convey was not expressed at all and both participants continued the conversation by talking about a new topic, which is what “kaya” means.

(9)         Syllabication

Syllabication is done by separating a word into syllables in order to emphasize its correct pronunciation. This pattern is identified as an interlocutor slowly and clearly pronounces every syllable of the word for clarification.

Speaker B: Maringis.

(a)          Ma-li-nis

Speaker A: Ma-li-nis. Malinis! [Clean!]

• Good! In Bisaya, lim-pyo.

In the statements above, the interlocutor who did the correction or repair was the recipient. This repair is done by repeating the word and slowly pronouncing its correct syllables to demonstrate to the Japanese recipient how these local Filipino words are correctly pronounced or syllabicated.

(10)    Completion

Completion is done by supplying words to complete the utterance. This pattern is used to help the statement become complete and clearer.

Speaker A: ‘Cause we [are] already eating, um…

(a)          Tikoy!

In his utterance, the speaker found it difficult to continue his sentence as indicated by the filler “um”, thus having the trouble been identified by the speaker himself. The recipient immediately made the repair by completing his sentence and naming the food that they had been eating. By adding one word, the sentence of the trouble source speaker had been completed and repaired.

Speaker A: It means…

Speaker B: It means?

(b)         He agree[s]…

The speaker paused after uttering the first part of his sentence. In response, the recipient initiated the repair by repeating the utterance with a rising intonation to indicate that he was asking about the continuation of the incomplete sentence. The trouble source speaker himself made the final repair by adding “he agrees” to complete the utterance.

CONCLUSION

By being unscripted in nature, a spoken conversation is typically flawed. Interestingly, the troubles that arise in these conversations and the repairs that are done to address such troubles show that participants are active processors of information who recognize errors and repair them upon identification. Such repair strategies are important in order to keep the conversation going. All of the four types of repair-self-initiated self-repair, other-initiated self- repair, other-initiated other-repair, self-initiated other-repair-discussed by [16]. were used by the participants. The self-initiated self-repair as the most common type exemplifies that speakers are capable of identifying and fixing their own errors.

Additionally, Schegloff [17] only cited four patterns of repair, this paper revealed ten types of repair used most specifically in intercultural communication where both speakers have limited proficiency in English. These patterns are namely repetition, code-switching, replacement, demonstration, modification, phonetic alteration, elaboration, abandonment, syllabication, and completion, as arranged from the most common to the least one. More types were employed as greater language barriers were experienced by the speakers from different cultures compared to those who speak the same language. Moreover, although the two most common patterns, repetition and code-switching, were arguably considered as deficits in one’s language use, various authors support the idea that these two are effective forms of repair as repetition strategically delays the ongoing sentence construction of the speaker and provides him time to plan for the next words to say [17-19]. while code-switching shows the linguistic competence of bilinguals and multilinguals who use more than one language to make repairs and explain the meanings of words [20,21]. All of the discussed repair patterns were either used simultaneously or consecutively in making repairs in the conversation. Most of the repair patterns are used with the others to repair different errors at the same time. Repairs are also made by both interlocutors, either the trouble source speaker or the recipient, which shows that these participants have manifested unity in addressing communication problems and maintaining the conversation [22,23].

Furthermore, this study recommends that trouble source or errors should be classified in specific categories in relation to the patterns of repair that are applicable to each type of error. To elaborate the connection of errors and repairs, it would be interesting to explore the trouble source that causes such repair. Lastly, a study on the conversation repairs used by other nationalities conversing not only in dyads but also in groups is recommended as well to further investigate how repairs are done when various cultures are involved.

• 
  Table 1
• 
  Table 2
• 
  Table 3
• 
  Table 4