Among these data, we observed fifty-nine if-conditionals in total. However, sixteen of them (27.1%) were fragmented only with the conditional marker if

CHAPTER FOUR

FINDINGS AND DISCUSSION

The participants’ spoken data altogether lasted for 89 minutes and 16 seconds.

Among these data, we observed fifty-nine if-conditionals in total. However, sixteen of them (27.1%) were fragmented only with the conditional marker if

. Such if-conditionals were considered incomplete and would not be of any concern in this study. Another type of if-conditionals excluded from the discussion was the idiomatic use of if-constructions (n=2), such as if necessary, if possible, if enough, etc. or those used in particular terms like the title of a movie, Catch Me If You Can. After eliminating the incomplete if-conditionals and the idiomatic expressions, we spotted forty-one applicable if-conditionals (69.5%). The specific distribution is listed as follows:

Table 4.1

The Number of If-conditionals in the Collected Data

Number

Complete 41 (69.5%)

*Incomplete 16 (27.1%)

*Idiomatic or other particular expressions 2 (3.4%)

Total 59 (100%)

Note. 1. “ * ” stands for those which were excluded from the analysis.

In this chapter, we will investigate how well the participant can construct the verbal forms of if-conditionals to express themselves in 4.1; in the meantime, since if-conditionals are a difficult construction for EFL learners to learn and to use, we will also inspect what strategies the participants adopt to avoid this construction. In 4.2,

23 There are some possible reasons. For example, the speaker might utter the conditional marker if first and then stop to think what s/he really intended to say. Or perhaps the speaker, since s/he was an EFL learner, was looking for proper words or trying to structure a correct sentence, so that s/he stopped to reorganize his/her thoughts as soon as s/he uttered if. Thus, there were some if-conditionals left incomplete.

we will move on to explore the discourse-pragmatic functions of if-conditionals in the spoken data to see how well the participants can use the construction to achieve communicative purposes. In 4.3, we will probe into possible reasons for the participants’ performance, such as the influence of L1 transfer and the input the participants received. Finally, 4.4 will summarize this chapter.

4.1 Form-construction of If-conditionals

In the study, the three conversation topics were selected based on the degree of hypotheticality. It was intended to examine if the participants were able to use correct and appropriate forms to express their attitude in a conversation. The results showed that the participants have difficulty constructing correct if-conditionals to express themselves.

Throughout the analysis, if any mistake was identified in either the protasis or the apodosis of an if-conditional, that if-conditional would be counted as one error.

The frequency of errors made by the participants is shown in Table 4.2.

Table 4.2

The Frequency of Errors

Total number Frequency of errors

Topic 1: Lottery 21 13 (61.9%)

Topic 2: Winter/Summer Vacation 3 1 (33.3%)

Topic 3: The 921 Earthquake 17 11 (64.7%)

Total 41 25 (60.9%)

As exhibited in Table 4.2, over half of the if-conditionals (60.9%) uttered by the

participants are incorrect/ungrammatical. Examined carefully, the number of

if-conditionals in Topic 1 reaches half (n=21) of the totality (n=41) and the frequency

of errors is 61.9%. Topic 2, unlike Topic 1, presents the fewest tokens of

if-conditionals (n=3) and only one error is spotted here. With respect to Topic 3, the

number of if-conditionals observed is smaller than that in Topic 1 by four tokens (n=17) but the participants make more mistakes here at the rate of 64.7%.

The striking differences in the number of if-conditionals and the frequency of errors among the three topics may be attributed to the intrinsic design of the experiment. Since the three conversation topics were selected according to the degree of hypotheticality, the complexity degree of the form-construction of if-conditionals would differ from one topic to another. As shown in Figure 4.1, the more hypothetical the meaning of an if-conditional was, the more complicated its structure would be.

The complication of forms further led to the higher frequency of errors.

Figure 4.1

Tendency of the Frequency of Errors in the Three Situated Conversation Topics

Topic 2, with lowest hypotheticality, was supposed to be the least complex one for the participants to construct if-conditionals. The results roughly conform to the assumption, for the number of if-conditionals is the lowest. The frequency of errors in Topic 1 (61.9%) generally corresponds with the average rate of errors (60.9%), suggesting the moderate level of complexity/difficulty to construct if-conditionals in Topic 1. Meanwhile, Topic 3, with the highest hypotheticality, displays the highest frequency of errors (64.7%). In the following sections, we will start to examine how and why the participants’ errors are produced specifically from the least complex

Topic 1 Lottery Topic 2

Winter/Summer Vacation

Topic 3 The 921 Earthquake

Less hypothetical Less complex forms Lower frequency of errors

More hypothetical

More complex forms

Higher frequency of errors

topic (i.e. Topic 2) to the most complex one (i.e. Topic 3).

4.1.1 Errors in Topic 2

Topic 2 Winter/Summer Vacation is less hypothetical in comparison with the other two topics. The if-conditionals used in Topic 2 are to make prediction (for the future). This kind of if-conditionals is the most prototypical and the forms are the easiest to construct, for there is no need to backshift the verb tense (Celce-Murcia &

Diane, 1999). Although there were only three if-conditionals observed in the spoken data of Topic 2, two of them were found correct. For example:

(1) Group 3, Topic 2

B3: Oh, okay ...Kaohsiung Kaohsiung …<M qijin M>. [You can ..]

A3: [Oh, I know.]

Æ B3: You can go there and .. take you to another place. Or if you visit me, I can take you to Tainan. There is so much ..

As indicated in (1), B3 predicted that she would take A3 to Tainan on the condition that A3 came to Kaohsiung to visit her. The condition set by B3 is actually quite probable; therefore, using the present tense in both the protasis and the apodosis can lower the hypotheticality.

The only error identified in the data of Topic 2 is exemplified below:

(2) Group 6, Topic 2

A6: Uhm. But I think .. I .. have to .. I have to have many experience with these special children .. in my .. in the future .. I .. I can ..

Æ because I have many experience, so I think .. if in the future I be a special teacher, I think I will .. teach them more.

B6: Do you want to choose another disabled children?

A6: Yes .. uhm .. I think ..

In example (2), the if-conditional was also used to make prediction for the future. It

can be seen that A6 was familiar with the prototypical form and function of if-conditionals even though the verb form “be” in the protasis of her utterance was left unchanged. It is speculated that since the if-clause A6 uttered was a prediction of her future career, she might unconsciously delete the future marker will in the if-clause and leave the bare form of the verb be intact. This mistake may not result from the lack of knowledge about the construction but from carelessness or slip of tongue.

Next, we will probe into the participants’ errors of if-conditionals in Topic 1.

4.1.2 Errors in Topic 1

The degree of hypotheticality in Topic 1 Lottery lies in the intermediate zone of the scale of hypotheticality. The if-conditionals in Topic 1 thus yield dual interpretations. They can be interpreted as highly hypothetical as long as the speaker thinks s/he has little chance to win the lottery through his/her life. On the other, the if-conditionals can also be considered less hypothetical as predictions, like those in Topic 2. Due to the double interpretations on the degree of hypotheticality, the difficulty/complexity in using correct if-conditional constructions in Topic 1 is magnified; hence, the frequency of errors (61.9%) is only slightly lower than that (64.7%) in Topic 3 by 2.8%. This suggests that to construct correct if-conditionals in Topic 1 is no less difficult than in Topic 3.

Most of the errors in Topic 1 were found to result from the dual interpretations of

lottery-winning. For the convenience and consistency in the identification of errors,

we would resort to the context of the participants’ conversations. We observed that the

participants preferred to think lottery-winning as something with possibility. This

interpretation made the if-conditionals in Topic 1 act more like predictions for the

future as those in Topic 2. The “impossible” reading would thus contradict with the

overall context and should be counted as error. Table 4.3 illustrates specific types of

error in Topic 1.

Table 4.3

Types of Error in Topic 1 Protasis Apodosis Number Types of error

(if-clause + main clause)

Specific number

X O 7 simple past + future 7

O X 3 simple present + simple present 3

(1) present perfect + simple present 1 X X 3

(2) simple past + simple present 2

Total 13

Note. 1. “O” stands for “correct” while “X” for “incorrect.”

As shown in Table 4.3, the participants tended to have simple past-tense verbs in the if-clause but verbs with a future marker will in the main clause (7 out of 13). Let’s examine example (3) carefully.

(3) Group 2, Topic 1

Æ A2: So if you won the lottery, what will you do?

B2: I will want to travel all around the world .. yeah .. because I want to … I want to know the other cultures … it’s very …

A2: Interesting?

B2: Attractive.

In (3), A2 seemed to think winning the lottery was scarcely possible because she used the past tense in the protasis. However, she shifted the verb tense to the future in the apodosis, which implies her attitude toward the possibility of lottery-winning is positive

. Apparently, the inconsistency of verb tense in the protasis and the apodosis obscures not only the realizability (i.e. probability) of the propositions but also the speaker’s intended meaning and attitude.

24 A2’s use of future tense in the apodosis can be viewed in another way. That is, the future tense may be used to denote the event time of the proposition in addition to the speaker’s attitude.

Likewise, the second type of error with correct protasis but incorrect apodosis suggests that some participants were not aware of the importance of maintaining the consistency of the speaker’s perspective in if-conditionals. For example:

(4) Group 3, Topic 1

A3: Yeah, yeah .. too bad. What will you do when you win the lottery?

B3: I will quit my school and .. and .. move to … uhm .. maybe Taitung.

A3: Why? Why?

B3: Because there .. the scenery there is very beautiful and I can live there for a long time .. for my life time and .. I don’t have to worry Æ about my … uhm … economics ..<@@> no. So if I win the first

prize, I don’t worry my life .. for a long time .. for my life time.

How about you?

A3: I will .. I will take a trip to … uhm .. everywhere.

Through the discourse of (4), A3 and B3 considered lottery-winning quite possible, for A3 asked What will you do when you win the lottery? rather than What will you do if you win the lottery? The when-clause usually implies higher possibility and realizability of the proposition than the if-clause (Akatsuka, 1985; Schwenter, 1998).

B3’s response corresponded with this interpretation. She used an if-conditional to make a prediction about her life after winning the lottery. However, she chose the factual simple present tense

in the apodosis. Supposedly, the proposition in the apodosis represents the effect/result when the condition in the if-clause is realized. In this case, the effect I don’t worry my life can only be carried out when the condition if I win the first prize is realized, which is not now but sometime in the future. It is obvious that the perspectives in the protasis (future and predictive) and the apodosis (present and factual) contradict with each other.

25 The simple present tense is usually used to express “facts” or “habits” (Givon, 1993; Celce-Murcia

& Diane, 1999).

More importantly, the fact that B3 in (4) failed to use the auxiliary will in the if-conditional indicates that the participants were unaware of the function of the modal auxiliary will in making possible predictions for the future. Because the apodosis of if-conditionals displays a possible outcome on condition that the hypothetical situation in the protasis is realized, modal auxiliaries like will or would are used to signal the conditioned degrees of the hypothetical mood. Without the modal auxiliaries, the meaning of if-conditionals can not be properly derived. Hence, errors like example (4) can also be seen as evidences of the participants’ poor understanding about the essentiality of modal auxiliaries in if-conditionals.

Furthermore, some participants made errors in the verb tense/aspect of both the protasis and apodosis of if-conditionals. Consider example (5):

(5) Group 2, Topic 1

B2: uhm .. <@trust me … yeah@>. Uhm .. because I think there are many poor people in … about .. Africa .. yeah .. so I think they many want some money to … to develop their country, yeah.

A2: Wow, [you are a nice girl, I think.]

Æ B2: [And what do you want to] do if you have .. won the lottery?

A2: First I want to buy a box and all the money will … <@@> .. and put it under my bed .. and <@ I think I will count the money every day. @>

In this example, B2 expressed that she would donate the lottery money to those in

need. And then she was curious about what A2 would do after lottery-winning by

means of an if-conditional what do you want to do [will you do] if you have .. won

[win] the lottery? Evidently, the verb forms in both the protasis and the apodosis are

wrong and inconsistent in the speaker’s perspective/epistemic stance. All these

illustrate that Chinese EFL learners are confronted with great obstacle in the

form-construction of if-conditionals. Especially as speakers whose L1 is

morphosyntactically different from L2, the learners have not fully realized the significance of proper verb tense/aspect in if-conditionals.

Generally speaking, the errors identified in Topic 1 suggest that the participants were perplexed about how to construct the structures of if-conditionals to express their opinions about lottery-winning appropriately. Now we will move on to discuss the errors in Topic 3.

4.1.3 Errors in Topic 3

In contrast with the data in the other two topics, most of the if-conditionals in Topic 3 the 921 Earthquake are supposed to be highly hypothetical in order to highlight the effect of counterfactuality. The verb tense in an if-conditional should be backshifted to past perfect to signify the distance of the event time from the speech time and thus to increase its hypotheticality (Comrie, 1986; Werth, 1997). All the eleven errors marked in Topic 3 resulted from the lack of backshifting. More specific descriptions of the error types are listed in Table 4.4.

Table 4.4

Types of Error in Topic 3

Protasis Apodosis Number Types of error (if-clause + main clause)

Specific number (1) simple present + questions 1

X O 3 (2) simple past + questions

(3) future + future

1 1

O X 0 0

(1) simple past + future 4 (2) simple present + simple present 3 X X 8

(3) simple present + simple past 1

Total 11

Note. 1. “O” stands for “correct” while “X” for “incorrect.”

The structures of if-conditionals in Topic 3 are the most complicated among the

three topics and thus have caused the most trouble to the participants. It is evident from Table 4.4 that most of the errors, eight out of eleven, were located both in the protasis and the apodosis. As suggested, the participants were unable to construct correct forms to highlight the counterfactuality of if-conditionals. Without proper verb-backshifting, they either used the simple present tense or the simple past.

One of the typical errors is as follows:

(6) Group 4, Topic 3

A4: This bad experience. So .. how about your view?

B4: Uhm .. the place, the time. Uhm .. I think … it is because .. in Taipei .. we have … better rescue system, so I .. would feel more ..

safe .. safety .. safe. So … so … so I think … people in <M zhongnanbu M>, they will get bigger hurt. And .. about the time … uhm … I think that’s because .. most .. most people was Æ sleeping. So they can’t .. they have no time to escape. So .. if .. if

it was in the morning, maybe the … the … number of …<@> … uhm .. how to say …(6.2) … we will … (3.1 ) oh .. I don’t know … A4: I think I can catch your meaning. [@@]

B4: [@@] The number … will .. not

so much.

A4: Okay.

In example (6), B4 was recalling the 921 earthquake. Her utterance … about the time ..

I think that’s because .. most people was [were] sleeping shows that B4 was aware

that she was talking about a past event. Yet, through her talk, we can observe evidence

to suggest her unskilled use of English verb tense/aspect. She showed even less

confidence in the construction of if-conditionals. She struggled with many pauses but

failed to construct the correct verbal forms (i.e. the past perfect) in the protasis if it

was [had happened] in the morning to increase counterfactuality (because of the

irreversibility of the past). Even worse, she shifted to the future in the apodosis The

number … will .. not [would have not been] so much. The non-application of the past

perfect and the contradiction between the meanings in the protasis and the apodosis reveals not only the participant’s deficiency in the knowledge of form-and-function mappings but also their inability to reach an agreement on the epistemic stance.

Meanwhile, it was found that most participants did not realize the purpose of verb-backshifting when describing a rejected past situation in if-conditionals.

Example (7) can verify this observation.

(7) Group 1, Topic 3

Æ B1: There are many people die. But … I think is the time. If it happen at morning or afternoon, many people can .. can … go out but that in midnight, many people….

A1: don’t feel and .. and they die. Uhm … okay .. I hope one day will not happen the earthquake.

In this scratch of conversation, B1 and A1 were discussing an alternative outcome if the event time was different. Apparently, B1 did not backshift the verb tense in the if-conditional. This does not mean that she failed to recognize the 921 earthquake as a past event because the pronoun it tells us she was not referring to other earthquakes but the 921 earthquake. This if-conditional could not have been used to make prediction for the future. Possible explanation for B1’s error lies in the complexity of the forms of counterfactual conditionals. She would rather choose the verb tense on which she had more control – the present tense – than backshift it. In addition, it is likely that she was unaware of the function of backshifting verb tense/aspect in English if-conditionals. Without the backshift of verbal forms, the counterfactuality of the propositions in the protasis and the apodosis cannot be projected.

Based on the discussion so far, we found that English counterfactual conditionals

caused more difficulty to the learners than open conditionals (i.e. predictions for the

future) because there were a high proportion of errors identified in both the protasis

and apodosis of the if-conditionals in Topic 3. However, the participants seemed to encounter little obstacle in the comprehension of a counterfactual conditional through their conversations. Let’s take example (6) for explanation again. In this case, B4 was apparently having a language breakdown with long pauses. While B4 was struggling to construct the if-conditional, A4 interrupted, saying I think I can catch your meaning.

This reveals that A4 had already understood what B4 intended to say by resorting to the overall discourse context and the content of their conversation. Example (6) substantiates that the participants had little difficulty decoding a counterfactual interpretation even without an explicit counterfactual construction. However, their competence to construct correct forms of counterfactuals definitely leaves a lot of room for improvement. It is also strongly suggested that the participants’ ability to perceive if-conditionals should serve as the foundation to recognize the correlation between structure/form and meaning/function.

4.1.4 Tendency of the Error Types

After examining the errors in each topic, we will go a step further to investigate the general tendency of the error types. Of the greatest interest is whether there is commonality among the various errors discussed in the previous sections. Broadly speaking, a large number of the errors in the participants’ spoken data resulted from their lack of sense of epistemic stance; that is, they failed to maintain the consistency of the speaker’s attitude or perspective through an if-conditional sentence. Only a few errors were originated from other reasons such as unfamiliarity with the basic syntactic structure of English. In what follows, we will start from the exploration of the major tendency of the error types – epistemic disagreement – and end with minor ones.

Usually, the choice of different verb tenses implies the speaker’s subjective

judgment on the event described in an if-conditional sentence (Fillmore, 1990). The semantic/pragmatic feature of epistemic stance determines the syntactic structures of if-conditionals. Nevertheless, the participants who received the traditional grammar instruction did not fully understand that the matching requirement between the protasis and apodosis of an if-conditional is subject to restrictions on the matching of epistemic stance. In their interlanguage, errors that have arisen from the insensitivity to the agreement of epistemic stance were commonly identified, as Table 4.5 shows:

Table 4.5

The Frequency of Errors due to Epistemic Disagreement

Topic Total number

of errors

Errors due to epistemic disagreement

Topic 1 Lottery 13 12 (92.3%)

Topic 2 Winter/Summer Vacation 1 0 (0%)

Topic 3 The 921 Earthquake 11 5 (45.5%)

Total 25 17 (68.0%)

Over two-thirds (68.0%) of the errors are labeled because of the incompatibility of epistemic stance between the protasis and the apodosis. It is worth noticing that 92.3% of the errors in Topic 1 result from the inconsistency of epistemic stance. The rate (45.5%) of epistemic disagreement in Topic 3 is twice less than that in Topic 1.

Reasons for that discrepancy will be scrutinized later. Now let’s consider an error of epistemic inconsistency in (8).

(8) Group 6, Topic 3

B6: I think .. uhm .. because I live in Taoyuan, and … uhm .. because Taipei .. is very huge city .. <@@> .. live a lot of people. So .. I Æ think if .. the earthquake … uhm .. happened ..

A6: Happen in Taipei?

B6: Uhm .. will … will damage .. very hardly.

Æ A6: Uhm .. I can’t agree with you anymore. Uhm .. I think if the

earthquake happened in Taipei, I think .. uhm .. if .. many people

will die and .. even Taiwan .. will disappear.

Within such a short scratch of conversation, two if-conditionals uttered by different participants were identified to contain the same errors. In the first if-conditional of example (8), B6 was stating her opinion on an alternative situation of the 921 earthquake, saying if the earthquake … uhm .. happened [had happened] .. in Taipei, [it]… will .. will damage [would have damaged] .. very hardly. She was trying to visualize an alternative past with a different outcome. She should have used the past perfect to show her “negative

” conviction or disbelief to that condition (Fillmore, 1990) and, most importantly, to highlight its counterfactuality. However, she used the past tense in the protasis and changed to the future tense in the apodosis, which implies a “neutral” epistemic stance of the speaker, allowing the realizability of the proposition in the apodosis. The speakers in this example were not only unable to use the correct morphosyntactic structure but also inconsistent in the epistemic stance between the protasis and the apodosis

.

In addition to the seventeen errors arising from the lack of sense of epistemic agreement, the rest (n=8) of the errors could be classified into three miscellaneous sources – disorientation of time frame, unfamiliarity with the basic syntactic form and structure of English and strategy of avoidance – as Table 4.6 reveals. More interestingly, six out of the eight errors were identified in Topic 3. This result can account for the discrepancy between the error rates of Topic 1 and Topic 3 in Table 4.5. In Topic 3, we found that the participants tended to use open conditionals with

26 Fillmore (1990) classifies if-conditionals into three levels of epistemic stance: positive, neutral and negative, which can be notionally correspondent to Akatsuka’s (1985) surprise, uncertainty and negative conviction.

27 Also, example (8) indicates that EFL learners’ inability to construct correct forms of if-conditionals may result from their general unfamiliarity with English tense/aspect system. Throughout the talk in (8), both interlocutors used the present or future tense to describe a past event where the past tense should be required. This shows that most learners have a poor command of the overall tense/aspect system, so that to make correct choice of verb forms in if-conditionals becomes an even tougher job to them.

present tense or even root forms of verbs, which is incorrect and incompatible to the discourse context. This shows that the participants are unable to maintain the epistemic agreement not only within an if-conditional as (8) but also between an if-conditional and its discourse context. Also it is reconfirmed that the high hypotheticality level of if-conditionals has multiplied the complexity and difficulty of form-construction to the participants, so that they made all kinds of errors in Topic 3.

Table 4.6

Errors from Other Sources

Possible sources Number Examples

Disorientation of

time frame 3

(1) If it happen at morning or afternoon, many people can go out but that in midnight, many people don’t feel and they die. (Group 1, Topic 3)

(2) We can imagine how damage will be if it happen in Taipei. (Group 2, Topic 3)

(3) If I run, it will be very strange. (Group 8, Topic 3)

Unfamiliarity with the basic syntactic form and structure of English

3

(1) If I win the lottery … have so much money, maybe [I will] do my first traveling … do my first abroad travel. (Group 4, Topic 1)

(2) If in the future I be [am/become] a special teacher, I think I will teach them more. (Group 6, Topic 2) (3) If there will be [is] another big earthquake, I ..

maybe I will runaway this time. (Group 8, Topic 3)

Strategy of

avoidance 2

(1) If the earthquake happen in Taipei very seriously, what about your opinion? (Group 6, Topic 3)

(2) What do you think if the earthquake was happened in Taipei? (Group 5, Topic 3)

Among the three other sources, the first one that catches our attention is that the

participants were disorientated about the time frame of the whole discourse. They

were unable to keep the verb tense/aspect in an if-conditional to be correspondent

with the overall time frame of the discourse. For instance:

(9) Group 8, Topic 3

B8: Our neighborhood all .. all run .. went out and .. we all together ..

we were all .. in the street and .. talk about the earthquake.

Æ A8: I didn’t run because I thought .. there are .. no other people run. If I run, it will …

B8: It will be very strange.

As can be observed in (9), the overall time frame was located in the past. Their use of the past-tense verbs such as went out, I didn’t run … or I thought … verifies that the participants were aware of the past time frame of the whole discourse. Nevertheless, that recognition was disorientated in the if-conditional sentence. They uttered the if-conditional If I run, it will be very strange without any verb tense backshifting. In fact, they should have produced If I had run, it would have been very strange to ensure that the time frame of the if-conditional was compatible to that of the whole discourse.

The second possible source of the participants’ errors lies in their unfamiliarity with the basic syntactic form and structure of English, which is SV(O). It is evident in example (10) where A4 dropped the subject of the apodosis in the if-conditional.

(10) Group 4, Topic 1

Æ A4: Save it. But now I think .. if I .. I win the lottery .. have so much money, maybe do my first traveling .. do my first abroad travel.

B4: Traveling. You never .. go abroad? I have never .. too. But I .. I don’t think … <@@> it’s the time for me to .. go.

A4 unconsciously deleted the subject in the apodosis, uttering …if I win the lottery

[and] have so much money, maybe [I will] do my first traveling … . She adopted a

salient feature of Chinese structure, which is that a noun phrase usually does not need

to be specified if it can be understood from context (Li & Thompson, 1982; Huang,

1984; Tao, 1996). She seemed to presuppose that the subject of the apodosis could be

comprehended through the context; hence, she substituted the pronoun I with a zero anaphora. This strategy is commonly exercised in Chinese but is not allowed in English.

The last source of errors is considered a strategy that the participants used to avoid speaking in the conversation as example (11) demonstrates.

(11) Group 6, Topic 3

Æ A6: If .. if the .. if the earthquake happen in Taipei very seriously, what about your opinion?

B6: I think .. uhm .. because I live in Taoyuan, and … uhm .. because Taipei .. is very huge city .. <@@> .. live a lot of people.

Usually in an if-conditional, the protasis and the apodosis maintain a relation of

“enablement” (Sweetser, 1990: 115). The realization of the condition in the protasis enables the occurrence of the result in the apodosis. This tendency is violated in the if-conditional sentence in (11), though. A6 replaced the result clause (i.e. the apodosis) with a question: what about your opinion? Logically speaking, there is no

“enablement” relationship between the protasis and the apodosis. A6 may plan to avoid her turn of speaking by asking her partner’s opinion. Raising a question can be a common strategy that L2 learners will adopt when they do not know what to say or how to say it.

In 4.1.4, we have indicated that most errors in the participants’ production of

if-conditionals were originated from the epistemic disagreement between the protasis

and the apodosis; only some arose from the reduction of English tense/aspect system

and syntactic structure or avoidance strategy. Next, we are going to further investigate

what kind of strategies the participants adopted to avoid producing if-conditionals.

4.1.5 Substitutes for If-conditional Constructions

Since most participants are unaware of the mappings between conditional forms and functions and have a poor control of the form-construction of if-conditionals, they tend to look for replacements to avoid using this construction in communication.

From the perspective of learning strategies, language learners will adopt the

“avoidance strategies

” to avoid any syntactic structures with which they are not familiar enough (Brown, 2000: 128). In the spoken data, it is revealed that the participants had many ways to avoid such perplexing construction as if-conditionals.

Some would directly omit the structure that they felt uncertain as example (12) shows.

(12) Group 7, Topic 3

A7: Uhm .. what do you think about the earthquake?

B7: You mean .. the place or the time?

A7: That’s okay.

B7: I think it’s much better .. it happens in the midnight than .. in the Æ morning because in the morning everybody go to work. And if ..

uhm … the building .. it may have a huge … A7: A serious condition.

B7: Oh, yeah.

B7 in this case did not utter the protasis of the if-conditional at all after a short period of thinking. It is very likely that she had a hard time constructing the complicated structure of the counterfactual conditional: If it had happened in the morning, the building would have had a serious damage. Thus, she decided to omit the protasis and let the overall discourse context help A7 to understand her talk.

28 According to Brown (2000), avoidance is a common strategy that language learners would use in communication. It can be divided into a few categories. First, “syntactic avoidance” is to avoid producing unfamiliar language structures. Second, “lexical avoidance” is to avoid lexical items that the learners have not yet acquired. Third, “phonological avoidance” is to avoid using words with difficult articulation to the learners. Last but not least, “topic avoidance” is to avoid talking about some subjects that the learners have problems to express themselves clearly. No matter which kind of avoidance strategies the learners adopt, they aim to lower the chance of making mistakes and to avoid misunderstanding in communication.

Additionally, some would retain the superficial sequence of the protasis and apodosis of an if-conditional without any conditional marker as demonstrated in example (13).

(13) Group 5, Topic 3

Æ B5: Oh, that’s great! I will screaming or .. uhm … I think the earthquake happen in Taipei .. will .. cause … uhm .. serious damage. So it’s fine to happen .. in the …

A5: Mid ..

B5: Yeah .. in mid-Taiwan. Because it’s almost mountains there and ..

not many people.

B5 was supposed to utter an if-conditional like if the earthquake had happened in Taipei, it would have caused serious damage. Sensing the complexity of the form (e.g.

the change of verb tense/aspect), she simplified the if-conditional by juxtaposing the two clauses in a condition-consequence sequence to avoid making mistakes. In the participants’ L1 Chinese, juxtaposing two sentences in adjacency is enough for a conditional interpretation even if there is no conditional marker. As a matter of fact, B5 was using the prototypical topic-comment construction of Chinese. She has unconsciously transferred this pattern of language use in Chinese

.

Unlike the prior two cases, some would paraphrase an if-conditional with the comparative.

(14) Group 3, Topic 3

B3: If is … in afternoon or in the morning, there were many people go .. go to work and .. there are many people in the tall building so

29 Example (13) may be explained in another way. B5’s utterance of the conditional talk may look similar to other Chinese conditionals, such as dehua-construction. Dehua-construction has been observed to undergo a process of grammaticalization from conditionals to a source for topic marker (Wang, 1996; Chiu & Chen, 2005). Similarly, the English conditional marker if is often associated with topic marker (Haiman, 1978). In (13), B5 might transfer the Chinese dehua-construction rather than use the English conditional marker if to topicalize the conditional proposition. However, this explanation seems to lack enough evidence in the data. We can only be certain that there are traces of L1 transfer in cases like (13) but cannot be definite which L1 language feature is actually transferred.

I think there were .. there were more damage.

Æ A3: Really? But I think .. uhm …at day is better … I think .. it happen in a day maybe the … the number of death will be lower .. much lower .. fewer.

B3: I don’t know, maybe.

As illustrated in example (14), A3 challenged B3’s view about the event time of the 921 earthquake. She seemed to feel intimidated in the utterance of if-conditionals. She chose the comparative, with which she might feel more confident, to replace the if-conditional construction. Furthermore, she produced a seemingly conditional construction, which only retains the sequence of the protasis and apodosis like that in example (13). We speculate that A3 may also transfer the Chinese pattern of language use by juxtaposing two clauses in a cause-effect sequence to make a conditional talk.

What is more interesting is that some of them found a lexical substitute to signal a conditional talk in the conversation. They would use the word maybe, which indicates options or possibilities, to replace the if-conditional construction. Let’s have a closer look in example (15).

(15) Group 6, Topic 2 A6: Sell books?

B6: No, I don’t sell books. Uhm .. like choose right books to … <M yunsong M> <@@>

A6: Oh.

B6: And very tired. Uhm .. one experience I spent .. fifteen hours one day.

A6: Wow!

B6: So I think so tired for me. So this summer vacation I don’t want to choose .. like this part-time job. I think I will took some camps.

A6: Oh. I think that .. uhm .. in special education, we seems to have ..

Æ <M tangbaobao M> camp. Maybe we can go to this camp. I think we will learn many things .. to .. get together with these special children. Do you want to go with me?

B6: Yes. <@@>

In example (15), B6 was recalling her taking a tiring part-time job during the winter vacation. Hearing B6’s complaint, A6 suggested that they join a camp together in the summer vacation, stating an alternative future Maybe we can go to this camp. [I think]

we will learn many things … . She actually used two single sentences instead of an if-conditional construction. The former resembled the protasis while the latter displays the outcome under that condition.

For EFL learners like our participants, to comprehend a word like maybe is much easier and less complex than to process a more cognitive-demanding syntactic structure such as if-conditionals because the word maybe possesses a more transparent connection between meaning and form. The if-conditional construction, nevertheless, involves the operation of many language mechanisms such as the verb tense/aspect, modal auxiliaries, negation or word order, etc. Since using a single word maybe could also realize a conditional talk, the participants like example (15) may consider it an easier way than the if-conditional construction. This illustrates that even if the participants have not fully acquired if-conditionals, they are still able to search for the most appropriate word in their mental lexicon to substitute the construction in order to make a conditional talk.

So far, we have probed into a few strategies the participants adopted to avoid

if-conditional construction, such as by simplifying if-conditionals or by substituting

with other words or structures. The avoidance can also be evidenced by the hesitation

in the utterance of this construction. All these strategies indirectly reflect the difficulty

of if-conditionals for EFL learners to learn and to use, suggesting the deficiency in

their overall language competence. In the next part, we will have a short interim

summary of 4.1.

4.1.6 Interim summary

From the analysis of the participants’ spoken data, they made numerous mistakes with respect to the verbal forms of if-conditionals as expected. It has been indicated that the degree of hypotheticality of each topic does exert effect on the participants’

use of if-conditionals. The participants’ poor understanding about the interrelation of verb tense/aspect and the hypotheticality of conditions/the speaker’s epistemic stance is the major cause of their errors. Three minor sources of the participants’ errors were also observed – the disorientation of time frame, unfamiliarity of English basic syntactic form and structure and the strategy of avoidance. Moreover, the participants adopted many strategies to avoid if-conditionals either by simplifying the syntactic structure of an if-conditional or by substituting with other structures or lexicon. All these imply the participants’ deficiency in general language competence.

To sum up, Hypothesis I – namely, EFL learners are expected to have difficulty constructing correct forms of if-conditionals, especially counterfactuals, in spontaneous communication due to the morphosyntactic differences between Chinese (L1) and English (L2) – is verified since the participants obviously had a hard time constructing the correct forms of if-conditionals to convey their epistemic stance in real communication. The findings suggest the essentiality of a more function-oriented way of teaching and learning if-conditionals in order to reinforce EFL learners’

understanding and manipulation of this construction. In the subsequent section, the attention of the study will be drawn to investigate how well the participants can use if-conditionals to achieve communicative goals.

4.2 Discourse-pragmatic Functions of If-conditionals

Based on the previous literature, the participants were expected to have little

problem in performing the discourse-pragmatic functions of if-conditionals because

their L1 and L2 share many similarities in this aspect (Wang, 1996). Indeed, we observed three major types of discourse-pragmatic functions of English if-conditionals – assuming, contrasting, and exploring of options – throughout the participants’ interlanguage. Details are tabulated in Table 4.7. As shown, the most frequently-used discourse function of if-conditionals is “assuming,” which occupies around forty percent of the totality (41.5%) in the participants’ spoken data. The second frequently-observed function is “contrasting,” which is nearly ten percent less (31.7%) than “assuming.” The third function found is “exploring of options” at the rate of 26.8%.

Table 4.7

Discourse-pragmatic Functions of If-conditionals in the Collected Data

Functions Number Rate of occurrences

Assuming 17 41.5%

Contrasting 13 31.7%

Exploring of options* 11 26.8%

Total 41 100%

Note. 1. *The function of “exploring of options” includes “exemplification” in Ford and Thompson’s (1986) term.

All of these functions represent the relationships that conditional sentences bear

to the preceding discourse. Since the communicative values of if-conditionals cannot

be interpreted without account of the overall discourse, it becomes a necessity to

examine the if-conditionals with consideration of the context provided by the three

situated conversation topics. Chart 4.1 thus summarizes how the discourse functions

of the if-conditionals were dispersed in each topic.

Chart 4.1

Distribution of the Discourse-pragmatic Functions in Each Topic

0 2 4 6 8 10 12 14

Topic 1 Topic 2 Topic 3 Topics

Nu m ber

Assuming Contrasting

Exploring of options

Chart 4.1 uncovers that the function of “assuming” (n=12) outnumbers others in Topic 1. As for Topic 3, the number (n=8) of if-conditionals as “contrasting” is the largest.

In Topic 2, the if-conditionals are indicated to perform only two discourse-pragmatic functions – “assuming” and “exploring of options.” The forthcoming sub-sections will discuss the types and distribution of the discourse-pragmatic functions case by case with examples from the participants’ interlanguage.

4.2.1 If-conditionals as assuming

Most of the if-conditionals in the participants’ interlanguage performed the function of “assuming;” that is, they are used to “repeat[s] an earlier claim,” which namely serves as a framework for the development of the succeeding discourse (Ford and Thompson 1986: 356). This function outnumbers the rest of the functions at the rate of 41.5% as shown in Table 4.7. What is worth noticing here is that nearly three-fourths (n=12) of them appear in Topic 1 as Table 4.8 displays. The distribution is likely to be related to the intrinsic topic design, for in Topic 1, there was lower possibility for topic shift. The participants had fewer subject matters to explore than in Topics 2 and 3. They mostly talked about what they would do if they won the lottery.

Thus, most if-conditionals in Topic 1 were found to perform the function of assuming

to encode the major topic of the conversation.

Table 4.8

Distribution of ‘Assuming’ in Each Topic

Topics Assuming

Topic 1 Lottery 12 out of 21

Topic 2 Winter/Summer Vacation 2 out of 3

Topic 3 The 921 Earthquake 3 out of 17

Total 17 out of 41

The function of “assuming” can be illustrated in the following example.

(16) Group 4, Topic 1

B4: Oh I think it’s a .. very special present because .. it can be a big present and .. or a small present.

A4: So the present may be .. its meaning may be not the money .. may be its surprise … I think. Someday maybe it’s a big surprise for you, so … [@@]

Æ B4: [@@] Oh, uhm .. if I have someday. Oh, I think .. uhm … I will donate … donated to some .. charity or .. save it.

In example (16), B4 and A4 were talking about receiving a lottery ticket as a gift. B4 seemed to believe that the value of that present highly depended on luck as she said it can be a big present and .. or a small present. A4 encouraged B4 to look on the bright side by saying Someday maybe it’s a big surprise for you. By means of an if-conditional if I have someday, B4 repeated A4’s claim as a foundation for the following talk to express her reaction if that was really the case. As demonstrated in this example, if-conditionals can be used to resume the prior claim as the basis for the development of the following discourse.

In fact, if-conditionals can be used as a strategy of making one’s turn relevant to

the prior context (Ford, 1997).

(17) Group 3, Topic 2

B3: How about you? You can go .. come to Kaohsiung to … A3: To visit you?

B3: Yeah yeah.

A3: Kaohsiung .. Kaohsiung … uhm … do they .. do there … B3: What?

Æ A3: have something interesting in Kaohsiung if we .. if I .. come here?

B3: Oh, okay ...Kaohsiung Kaohsiung …<M qijin M>. [You can ..]

A3: [Oh, I know.]

Æ B3: You can go there and .. take you to another place. Or if you visit me, I can take you to Tainan. There is so much ..

At the start of example (17), B3 was suggesting that A3 come to Kaohsiung to visit her in the winter vacation. In this way, the idea of taking a trip to Kaohsiung was established on the participants’ mind and became the major subject under discussion.

To make sure her turn was relevant, A3 used an if-conditional to perform the function of assuming. She asked about some famous sightseeing spots in Kaohsiung based on the proposition of the protasis if I come here. Similarly, B3 uttered another if-conditional if you visit me as a framing mechanism to maintain the connection of her talk with the previous context in order to further highlight her desire or expectation toward that condition (McCarthy, 1991: 20). For the purpose of showing relevance of their contribution, A3 and B3 both re-emphasized the topic by means of if-conditionals when taking their turn in the conversation.

Meanwhile, we found that if-conditionals were also used as a framing

mechanism for a wide range of discourse. Some participants tended to encode the

topic in if-conditionals in the very beginning of conversation. Their partners could

carry on the talk with no need to restate the if-conditional, namely the topic, in their

turns because the two parties had reached a tacit understanding about the focus of

their conversation. Let’s look at example (18) closely.

(18) Group 5, Topic 1

A5: Oh, hi! Have you ever won the lottery?

B5: Uhm .. no.

Æ A5: If one day I win the lottery, I will buy a big house for my family ..

for my cousin .. for my grandmother .. everybody live together. And I will buy .. cars for my father because .. he wants a new car. And ..

we will take trip .. to everywhere because my father and mother haven’t .. go out for a long time. And we didn’t .. never <@@> ..

go to foreign countries in my life .. so I will take a trip around the world. And you?

B5: I will .. bring my family go to .. foreign countries. Uhm … uhm ..

maybe I won’t .. I won’t …I won’t study anymore because .. there are a lot of money. <@@> And I am a lazy person. So I don’t want to study anymore.

A5: Me too.

Example (18) was the very beginning of the conversation between A5 and B5. A5 initiated the conversation by asking whether B5 had ever won the lottery. Receiving minimal response, A5 used an if-conditional to keep the relevance of her talk and most importantly, to pinpoint the topic for the upcoming conversation. While A5 was elaborating her opinions, B5 had already borne in mind the topic if one day I win the lottery; hence, it was easy for her to keep the turn relevant even without reiterating the topic. As illustrated in this case, the if-conditional produced by A5 differs slightly from those in the previous examples (i.e. (16) and (17)) because it establishes a framework for a much broader range of discourse. This shows that if-conditional construction can function not only as a micro framing device between lines but also a macro one across turns. If the topic encoded in if-conditionals is commonly agreed between interlocutors, the conversation can be proceeded naturally and smoothly even without any reassertion.

The three instances discussed so far indicate the first major function of the

if-conditionals – assuming – in the participants’ interlanguage. We have observed little problem encountered by the participants when they were using if-conditionals to encode the topic either for a limited or a much wider range of discourse. They can make use of this construction to highlight the main concern and to keep their turn as relevant as possible. Even without an explicit utterance of if-conditionals, they can still capture the focus of the previous discourse and continue to make a conditional talk as B5 in example (18). All of these show their natural sense of being interactive and cooperative in real communication.

4.2.2 If-conditionals as contrasting

The second highest proportion of the if-conditionals in the participants’ spoken data was to contrast (31.7%). If-conditionals were observed to be used to emphasize a wide disparity from the previous discourse. Namely, they can present an assertion that is the opposite of, or is an option competing with, some assertion or set of assertions from the prior discourse (Ford, 1993). From Table 4.9, this discourse-pragmatic function is seen to appear more frequently in Topic 3 (n=8).

Table 4.9

Distribution of ‘Contrasting’ in Each Topic

Topics Contrasting

Topic 1 Lottery 5 out of 21

Topic 2 Winter/Summer Vacation 0 out of 3

Topic 3 The 921 Earthquake 8 out of 17

Total 13 out of 41

Generally speaking, contrasting includes the if-conditionals that “offer[s] a

contrast to something which has gone before” (Ford and Thompson 1986: 357). We

have also noticed that the hypotheticality and the speaker’s epistemic stance in this

category are often so intensified that the if-conditionals have, in fact, become

counterfactuals. Let’s consider example (19).

(19) Group 3, Topic 3

B3: I think it is earthquake and .. the earthquake was so serious so ..

the Taichung and the neighborhood … uhm … the neighborhood was damaged seriously.

A3: Oh .. uh .. in Kaohsiung so so … but .. the earthquake … play in ..

at midnight. I think it’s not the very good time to happen this thing.

B3: I think it is good time.

A3: Uh?? Why?

B3: If [it happen …]

A3: [People is all fall asleep.]

Æ B3: If is … in afternoon or in the morning, there were many people go .. go to work and .. there are many people in the tall building so I think there were .. there were more damage.

In example (19), A3 and B3 were in disagreement when discussing about the event time of the 921 earthquake. A3 thought that midnight was a bad time for the occurrence of the earthquake. However, B3 challenged that claim by using a counterfactual conditional (the form was not exactly correct, though) to imagine a much more disastrous circumstance if the earthquake had taken place in the morning or afternoon rather than in the midnight.

From example (19), it can be seen that the participants realized how to use

if-conditionals to present a contrastive argument, except that they used the wrong

morphosyntactic representations. The intensification of the hypotheticality and the

speaker’s epistemic stance should be realized by a counterfactual past perfect to

depict a rejected past scenario to contrast what has happened and to highlight the

speaker’s “desirability or undesirability” about the proposition in the counterfactual

(Ford & Thompson, 1986; Akatsuka & Strauss, 2000). What’s more, most of the

if-conditionals (n=8) in Topic 3 of the participants’ interlanguage were inspected to

perform the function of contrasting. This fact can probably be associated with the

irreversibility of the event the 921 Earthquake and the relatedness of otherwise situations, lending itself to highly hypothetical counterfactuals.

What deserves more attention here is that if-conditionals can serve as a politeness device in communication (Brown & Levinson, 1987; Thomas, 1995).

Lakoff (1973: 293-298) has pointed out three guidelines for politeness from the speaker’s perspective: (1) don’t impose; (2) give options and (3) make the listener feel good – be friendly. In example (19), A3 and B3 held distinct attitude toward the issue of when would have been a much better time for the 921 earthquake to take place. A3 thought the midnight was not a good time by saying I think it’s not the [a] very good time to happen this thing [for the earthquake to occur]. B3 disagreed to this view by stating I think it is [was a] good time. To further elaborate her opinion, she imagined an alternative situation with an if-conditional. The hypotheticality gives the listener A3 an option to think in another way. In other words, the optionality of if-conditionals enables B3 to decrease the assertibility of her statement, and thus to soften her disagreement so as to avoid direct confrontation with A3.

Koike (1989) also ties politeness to the distance of temporal or person deictic center. The temporal deixis usually refers to the speaker’s “present” moment of speaking. Distancing the verb tense from the temporal deictic center by using counterfactual if-conditionals (e.g. example (19)), for instance, will convey a sense of

“tentativeness” on the part of the speaker. This sense enables the speaker “to avoid being dogmatic, categorical, or rude” (Bull, 1968: 104) so that the hearer will not feel challenged or offended. Generally speaking, we can “minimize the speaker’s egocentric role in the utterance” for greater degree of politeness by distancing the verb tense/aspect (Koike, 1989: 192).

According to the above investigation, the participants’ capability of using the

if-conditional construction to make contrast conforms to the observation of previous

research (Ford & Thompson, 1986; Ford, 1993, 1997). Also, some if-conditionals in the participants’ interlanguage have been indicated to function as a strategy of politeness. Though the grammatical structures in their production are not well-developed, the real language use of if-conditionals has already been naturally performed.

4.2.3 If-conditionals as exploring of options

Another major discourse function of if-conditionals noted in the participants’

interlanguage was to explore options or possibilities (26.8%). This function refers to the if-conditionals that have no direct relationship with the preceding discourse but

“opens up new possibilities whose consequences are to be explored” (Ford and Thompson 1986: 358). The distribution of the discourse function in each topic is presented in Table 4.10. Six out of eleven appear in Topic 3, four in Topic 1 and one in Topic 2.

Table 4.10

Distribution of ‘Exploring of Options’ in Each Topic

Topics Exploring of options

Topic 1 Lottery 4 out of 21

Topic 2 Winter/Summer Vacation 1 out of 3

Topic 3 The 921 Earthquake 6 out of 17

Total 11 out of 41

Now let’s take a scratch of the participants’ conversation as an instance to illustrate how they use if-conditionals to explore options.

(20) Group 1, Topic 3

B1: uhm .. for me, I sleep. I don’t be shot … the earthquake didn’t wake .. wake me up.

A1: That’s nice. That is not a good experience. So if I .. I hope I am

sleeping and feel nothing.

B1: But for my father and mother … that night … uhm … they are in Æ their bed. … And .. they talked about if the earthquake

made them die, they were so scared if .. because we don’t in the same room. I and .. my sister and I live together, they are in another room and they don’t have a chance to wake me up. So ..

they are very scared. Uhm … and the next morning, my mom tell me the earthquake and I don’t believe it. I said why I …

In example (20), B1 and A1 were talking about their terrible experience of the 921 earthquake. B1 recalled that her parents were worrying about the children’s safety on that very night. She retold her parents’ feeling, saying …they talked about if the earthquake made them die, they were so scared [if]…because we don’t [weren’t] in the same room. Examined carefully, the if-conditional embedded in the main clause they talked about … presents an alternative of the 921 earthquake from the perspective of B1’s parents rather than B1 herself. In other words, B1’s parents were picturing a possible situation (i.e. they might have been dead) after the earthquake happened.

In the same vein, we observed that if-conditionals can be used to explore options for the future. In the following example, A1 and B1 continued to talk about their experience after the 921 earthquake.

(21) Group 1, Topic 3

A1: Uhm .. the school is … uhm … anything will … like a … the wall … change the topic. Uhm … how about … do you .. do your school give the lesson to .. how to save your life in an earthquake?

B1: No .. just discuss the experience at that night .. but I don’t … I .. I do not join their discussion because I have … I didn’t discussion this topic because I never .. I didn’t wake up, so I don’t feel … A1: uhm … since then, my house is always be food, water and ..

Æ just like trying .. like this .. to prepare if one day earthquake

happen again, we can save our time .. our life.

After the earthquake, A1’s school gave the students a lesson on how to save one’s life in an earthquake. In that lesson, she learned that to prepare some food or water beforehand could save her life in an earthquake. Closely inspecting the structure A1 used, we found that she made a possible prediction for the future through an if-conditional if one day [an] earthquake happen[s] again. In light of what she learned at school, she thought if the possibility explored in the if-conditional was realized, her family would know how to survive. If-conditionals like example (20) and (21) do not possess as strong a connection with the prior discourse as the other two functions – assuming and contrasting – but open up a new option from the pool of possibilities.

In the participants’ spoken data, we also observed that if-conditionals can be accompanied with phrases like for example to explore options. This kind of if-conditionals is usually classified into the discourse function of “exemplification” in Ford and Thompson’s study (1986). According to Ford and Thompson, the if-conditionals that function as exemplification serve to “present a special instance of the generalization expressed in the previous discourse (p. 358). However, the definition may not be applied to the case in the participants’ use of if-conditionals. As can be seen in the next talk, B1 was using an if-conditional to explain her preferred habit of lottery-purchasing.

(22) Group 1, Topic 1

B1: My ... I will .. I have a habit to buy lottery. If one day…

A1: Tuesday and Friday?

Æ B1: No … for example, if today is my birthday, I will buy a lottery to .. record the day.

A1: Then the number will choose your birthday number, right?

B1: Right!

Note that in example (22), B1 meant to explain her habit of lottery-buying to A1. At first, it seemed that B1 was still organizing her thoughts. Meanwhile, A1 was eager to make a guess about what B1 tried to say but failed. To clear A1’s confusion, B1 illustrated her favored lottery-purchasing habit with a specific example in an if-conditional sentence. That is to say, B1 hypothesized a situation where she would want to buy lottery in remembrance. Yet, the value of the example encoded in the if-conditional does not correspond with Ford and Thompson’s idea about

“exemplification” even though the if-conditional is preceded by the phrase for example. The reason lies in the lack of a more general claim in the prior discourse. If B1 had made a general statement like I will buy lottery on special occasions for remembrance first, this conditional sentence would have functioned more like

“exemplification.” We thus interpret example (22) more like “exploring of options.”

On the strength of example (22), we decide that the two discourse functions,

“exploring of options” and “exemplification,” overlap notionally to some extent.

Using if-conditionals to exemplify can be broadly related to option-exploring. No matter an if-conditional is used to concretize an abstract concept or to narrow down the scope of a topic, it is exploring a possibility.

Among the eleven if-conditionals that were recognized to explore options/possibilities in the participants’ interlanguage, most of them were used to make probable prediction for the future as example (21). This consolidates that we should not deny the participants’ capability of setting up a conditional/hypothetical scenario for further exploration in communication even though their if-conditionals are often filled with grammatical errors.

4.2.4 Comparison between the interlanguage and the target language

After the analysis of the discourse-pragmatic functions of the if-conditionals in

the participants’ interlanguage, what is highly concerned in the study is whether the EFL learners’ use of if-conditionals resembles that of English native speakers. We will take Ford and Thompson’s study (1986) as a sample for comparison

. Detailed comparison of the two studies

is exhibited in Chart 4.2.

Chart 4.2

Comparison of the Use of If-conditionals in the Study and Ford & Thompson’s (1986)

0.00%

10.00%

20.00%

30.00%

40.00%

50.00%

The study (EFL learners)

Ford and Thompson's study

(speakers of English)

Percentage Assuming

Contrasting

Exploring of options Polite directives

From Chart 4.2, it can be seen that the EFL learners’ use of if-conditionals did not entirely agree with the speakers of English in Ford and Thompson’s study (1986). In the present study, “assuming” is the most frequently-observed function. Particularly, the percentage (41.5%) of “assuming” in the EFL learners’ data is ten to twenty percent higher than the other two functions (“contrasting”: 31.7%; “exploring of options”: 26.8%). Differently, Ford and Thompson’s finding shows that “exploring of options” (45%) is the most frequently-used function, which is nearly twice higher than “assuming” (25%) and contrasting” (21%). We speculate that the disparity

30 The data in Ford and Thompson’s study (1986) may not be totally compatible to the data in this study because Ford and Thompson include some written/formal language, which is in contrast with the spoken data of the present study. Yet, the comparison is still made in order to show a general picture of how EFL learners’ language competence reflects the real language use of English native speakers.

31 Since “exemplification” is merged into a subtype of “exploring of options” in this study, we will integrate the percentage of “exemplification” with that of “exploring of options” in Ford and Thompson’s study (1986) for further comparison.