Abstract: This study tries to validate three kinds of discourse completion tasks, which are mainly used in order to elicit sociolinguistic data. The discourse completion tasks under investigation will be open written discourse completion task, dramatic written discourse completion task and discourse role play task. Yet as, discourse completion tasks have not yet undergone the rigorous processes of validation-which can be the result of the presence of a large number of pragmatic situations and speech acts-this study does not intend to correlate one newly developed test with an established one, instead it intends to see whether the three common kinds of discourse completion tasks converge in their final results or not. The results of this study disfavor the use of open written discourse completion tasks.
Vahid Parvaresh and Mansoor Tavakoli, 2009. Discourse Completion Tasks as Elicitation Tools: How Convergent Are They?. The Social Sciences, 4: 366-373.
Initially used by Blum-Kulka (1982) for the sake of investigating speech acts, Discourse Completion Tasks (hereafter DCT) have been extensively used in studies of pragmatic knowledge. A DCT is a data gathering device specifically designed in order to elicit responses to problematic, contextually-specific prompts (Zuskin, 1993). DCTs have become so vital tools in eliciting language to the extent that different formats of them have also been utilized in studies of metaphorical conceptualizations as well (Parvaresh, 2008).
Alavi and Dini (2008) contend that in spite of the large amount of research on pragmatic knowledge, the assessment issues in this area have not been vigorously investigated. They believe that a careful attention to research on pragmatic assessment may lead to developing appropriate means for testing pragmatic competence. And due to fundamental role of pragmatic knowledge in improving second language learners overall language proficiency and making them communicatively competent language users, the necessity of developing appropriate pragmatic tests, which examine the learners' application and understanding of such sociocultural conventions of the target language in various contexts seems to be beyond dispute.
And it also goes without saying that pragmatic tools like DCT are not only used for investigating second Language (L2) pragmatic knowledge, but also for the pragmatic studies of the first Language (L1) since the more the issues of convergence and divergence between languages are concerned, the better will be the ability to prepare the language materials accordingly (Blum-Kulka et al., 1989).
Being in agreement with the above-mentioned discovery, the purpose of this study is to validate three kinds of DCTs, which are used for eliciting sociolinguistic data more often than the other forms of the DCT in both L1 and L2. In other words, the purpose of this study is to investigate whether or not these three forms of DCT converge in their final results.
Generally, six types of DCTs can be distinguished, which claim that they can tap into the pragmatic knowledge of the test takers. They include:
Written Discourse Completion Task (WDCT), which requires students to read a written description of a situation and then asks them to write what they would say in that situation. This form can be a purely open one (OWDCT) or one made based, for example, on a piece of dramatic research, in which some parts have been removed (DWDCT)
Multiple-choice Discourse Completion Task (MDCT), which requires students to read a written description of a situation and select what would be best to say in that situation
Oral Discourse Completion Task (ODCT), which requires students to listen to a description of a situation and to say aloud what they would say in that situation
Discourse Role-Play Task (DRPT), which provides a description of a situation and asks students to play a particular role with another person in that situation
Discourse Self-Assessment Task (DSAT), which provides a written description of a situation and asks the students to rate their own ability to perform the necessary speech act in that situation
Role-Play Self-Assessment (RPSA), which combines the DRPT with the DSAT by requiring students to rate their own pragmatics performance in a previously performed role-play that has been recorded on a video recorder (Blum-Kulka, 1982; Brown, 2001)
The three kinds of DCT, which will be the focus of this study are: OWDCT, DWDCT and DRPT. Such DCTs will be used in order to investigate the realization of the speech act of politeness among male speakers of Persian.
A look at validity: According to Bachman (1990), validity refers to the degree that the evidence supports inferences that are made from the tests. In this way, he argues that it is not the scores, which are validated but the inferences derived from the scores of a test. For this reason, he considers validity as the agreement among tests to measure the same trait through maximally different methods.
Weir (2005) also contends that the key measurement problem is determining the extent to which, the sample of language use we obtain from a test adequately characterizes the overall potential of the individuals. Messick (1988, 1989) also believes that the validity of a test resides in both the evidence for and the potential and actual consequences of the test interpretation and use. In other words, as argued by Cronbach (1988), validity is a never-ending process. It requires, the constant process of theory building and testing.
Construct validity is the most important aspect of validity, which is specifically used when testers have no definite criterion measure of the quality with, which they are concerned and also, when they have to rely on indirect measures. Generally speaking, in construct validation of a test three different but related processes can be utilized (Fulcher and Davidson, 2007; McNamara and Roever, 2006).
Group difference: A valid test should adequately distinguish between different groups of individuals. For example, a test of attitude should distinguish between the attitudes of church-goers and non-church-goers.
Correlation: There should be a correlation between two tests, which claim that they measure a single trait.
Changing over occasion: If the test prompt will change, so will the test results. For example, if in a test the test taker is given the prompt tell the truth compared with make yourself look good, his/her performance on the test should change accordingly.
The validation process of this study lies, in fact, in the second dimension, that is the dimension of correlation establishment; but since, as it was argued in the above, discourse completion tasks have not yet undergone the rigorous processes of validation, which can be the result of the presence of a large number of pragmatic situations and speech acts-this study does not intend to correlate one newly developed test with an established one, but it intends to see whether, three common kinds of DCTs converge in their final results or not. In other words, this study will attempt to investigate whether or not the three widely used kinds of DCTs (OWDCT, DWDCT and DRPT) will yield similar results.
A look at politeness: Politeness is a term used in sociolinguistics characterizing linguistic features mediating norms of social behavior in relation to the notions such as rapport, courtesy, deference and distance.
Politeness phenomena are especially, involved when and where one of the interlocutors wants to disagree with what the other interlocutor has just said or mentioned. When we disagree with what a person is saying or doing, we might in one way or another, threaten that persons face. The sociologist Erving Goffman (1967) introduced the term for the first time and since, then it has had an enormous impact on the scholars working in the fields of pragmatics and sociolinguistics. He defined face as the positive social value a person effectively claims by the line others assume he has taken during a particular contact.
One of the greatest impacts of Goffmans face was on Brown and Levinson (1987). They arrived at the conclusion that some basic aspects of politeness are universal and as a result, they argued that face is invested in the individual and can be lost and therefore, it must be constantly attended to in the course of interaction. They distinguished between two universal aspects of face, namely positive and negative ones. They defined the former as the want of every member that his wants be desirable to at least some others and the latter as the want of every competent adult member that his actions be unimpeded by others. The researchers believed that each utterance might impose a threat to the other interlocutors face. In this regard, they contended that the assessment of the seriousness of Face Threatening Acts (FTAs) involves the following factors in many and perhaps all cultures, the social Distance (D) of the speaker and hearer, the relative Power (P) of the speaker and hearer and the absolute Ranking (R) of impositions in the particular culture.
But although, it seems that politeness is a universal concept, at least its realization will vary from culture to culture (Edstrom, 2004; Ladegaard, 2004). As a result, the use of different politeness strategies in different languages and cultures has been the topic of intensive research in recent years. In fact, intercultural pragmatics has investigated how one culture uses one strategy rather than the other in order to save face in one specific situation (Felix-Brasdefer, 2006).
One of the most widely used systems of politeness has been proposed by Scollon and Scollon (2001). Such a system includes the following determining facets, deference, solidarity and hierarchical.
According to the authors, in deference politeness system, although none of the interlocutors exerts power over the other (-Power), the relationship is distant (+Distance). In a solidarity politeness system, still none of the interlocutors exerts power over the other (-Power); but, unlike in a deference politeness system, the relationship between the interlocutors is intimate or close (-Distance). And finally, in the hierarchical politeness system, although one of the interlocutors has power over the other (+Power), the relationship might be either close (-Distance) or distant (+Distance).
This study will try to focus on what happens in an eastern and Islamic when boys want to disagree with other interlocutors with special emphasis on how different assessment tools can investigate this phenomenon accurately. The systems under investigation will be solidarity and deference. The reason for choosing such systems is that in Iran-which is an eastern and Islamic culture-boys seem not to have the same freedom when they interact with interlocutors who are from the opposite sex. This might be the result of the fact that in a society like Iran women and men spend most of their lives interactionally and spatially segregated.
MATERIALS AND METHODS
Participants: One hundred and eighty male native Persian speakers took part in this study. Age of the participants ranged from 19-25. The university students were chosen in this study in order to increase the comparability level of the study with other studies in other cultures because most of the politeness studies have targeted university students. The participants were either chosen from the University of Isfahan located in the center of the Islamic Republic of Iran or from two of Isfahans language institutes.
Procedure: A Dramatic Written Completion Task (DWDCT) comprising of 3 tasks was developed based on a Persian short story written by Jalal Ale Ahmad (Appendix A). In those tasks, the specific characteristics of the context would force the participants to disagree with what the other interlocutor had just said or requested.
The original story was about a boy who wanted to urge his friend to climb a staircase, which would lead to the roof of a mosque. But because the aim of the study was to investigate the effects of both solidarity and deference on male speakers, two versions of that DCT were developed. In the first version, the name of the addressee remained constant (a boy name) and in the second version the name of that person was changed into a girl name. Of the participants of the study, 76 were randomly assigned to one of the versions of the DWDCT. Half of them received the version in which, they had to disagree with a male speaker and the rest had to disagree with a female one. The Table 1 summarizes the situations in the two developed versions of the DCT.
Based on this DWDCT, an OWDCT was also developed (Appendix B). This OWDCT also appeared in two versions. In the first version, the participants had to disagree with a male interlocutor and in the second version with a female one. This OWDCT was taken by another group of 76 participants following the same procedure as the one used for the DWDCT. Finally, 14 cohorts of male participants were engaged in a role play based on the three scenarios of the story (Appendix C). Half of them were asked to play the role of a person interacting with a male interlocutor and half of them the role of a person interacting with a female one.
Coding and data analysis: The unit of analysis was decided to be the utterance or sequences of utterances the respondents produced. These utterances were, then, analyzed based on Leech (1983) classification of different illocutionary functions (Table 2).
Two DCT versions of the study
Leech (1983) classification of illocutionary functions
According to this classification, illocutionary goals are divided into four categories of competitives, convivials, collaboratives and conflictives based on how they relate to the social goal of establishing and maintaining comity. Being in agreement with this kind of classification, the first (competitives) and the last category (conflictives) were regarded as impolite with the latter being the rudest and the second (convivials) and the third category (collaboratives) were regarded as polite in nature.
RESULTS AND DISCUSSION
Analysis of the dramatic DCT: In the current study, several expressions were used in order to disagree with what the other interlocutor had previously said. They ranged from highly impolite, rude responses to more polite ones.
Table 3 shows, in task number 1 highly impolite utterances like cursing and reprimanding accounted for more than half of the overall responses where, the respondents had to disagree with another male speaker, whereas such responses accounted for only 13% of the whole responses where, the respondents had to disagree with a female interlocutor. In other words, in Boy-Boy situation the dominant pattern was a reprimand like dobaare zeddehaal zadi (you have become a wet blanket) sometimes followed by a curse like aakhe ahmagh in moaqzzene pire (are you nuts? The muezzin is old); but in Boy-Girl situation the dominant pattern was a claim like vali man miram baalaa azash (but I will go up) followed by an invitation like age biyaay khoobehaa (If you come, that will be fun).
The analysis of task number 2 has been presented in Table 4. Table 4 shows, in the Boy-Boy situation cursing utterances like ah, tarsoo (Oh, you cowardly) or to ke inghadr tarsoo naboodi (you used to be more courageous) were dominant where as in Boy-Girl situation it was the pattern of claim or promise like age shekastani bood taahaalaa shekaste bood (If it was to break, it should have been broken by now) or ghol midam nashkane (I promise it wont break), which turned out to be dominant.
The results of the analysis of the final task (Table 5) revealed that in the Boy-Boy situation the combination of a curse plus a reprimand like maskhare natars (you idiot, he cannot do the damn thing) accounted for the majority of the responses (80.70%). In the other situation, it was the pattern of claim and complaint, which was dominant (89.85%). In other words, in this situation, the participants used claims like oon hich kaari nemitoone bokone (he cant do anything) preceded or followed by complaints like cheraa mitarsi in ghadr (why are you so much afraid of him).
Illocutionary goals used in task 1
Illocutionary goals used in task 2
Analysis of the open DCT: Although, the analysis of the Boy-Boy situations in the open DCT revealed similar results as the ones found in the Boy-Boy situations of the dramatic DCT, in Boy-Girl situations the responses were found to be inconsistent with the Boy-Girl situations of the dramatic DCT. In other words, in the open DCT the participants revealed quite different patterns when they had to disagree with an interlocutor from the opposite sex.
In the first Boy-Boy situation of the open DCT (Table 6) the dominant pattern was, like the one in Boy-Boy situation of the DCT, a reprimand followed by a curse (88.33%). In the Boy-Girl situation, however, the dominant pattern was not a claim followed by an invitation but a curse sometimes followed or preceded by a reprimand or blame (72.40%).
As shown in Table 7, in the second Boy-Boy situation of the open DCT, the pattern of cursing was repeated as the one in the Boy-Boy situation of the dramatic DCT (85.89%); but, in the Boy-Girl situation, the pattern of claim and promise of the dramatic DCT was superseded by the conflictives like cursing and reprimanding (56.92%).
Illocutionary goals used in task 3
Illocutionary goals used in task 1
Illocutionary goals used in task 2
In the final task of the open DCT (Table 8), the pattern of cursing and reprimanding in the Boy-Boy situation remained dominant as predicted (85.71%); but, in the Boy-Girl situation although, the presence of the pattern of claim and complaint could easily be noticed, the pattern of curse and reprimand increased drastically (54.32%).
Analysis of the role play DCT: In the role play DCT, the patterns found were very much similar to the ones found in the DWDCT. Table 9 reveals, what happened when the interlocutors were engaged in role-playing task one. As the Table 9 shows, impolite utterances like cursing and reprimanding dominated where, the respondents had to disagree with another male speaker (76.92%) where as such responses accounted for only one fourth of the whole responses where, the respondents had to disagree with a supposedly female interlocutor. In other words, in Boy-Boy situation the dominant pattern was a reprimand but in Boy-Girl situation the dominant pattern was claiming (50%).
Illocutionary goals used in task 3
Illocutionary goals used in task 1
Illocutionary goals used in task 2
In task number two, again the patterns found were very much similar to the ones in the DWDCT. In fine analysis, in the Boy-Boy situation cursing utterances were found to be dominant (80%) where,as in Boy-Girl claims (66.66%) constituted the majority of the responses (Table 10).
The analysis of the final task (Table 11) revealed that again in the Boy-Boy situation it was the combination of a curse followed or preceded by a reprimand, which accounted for the majority of the responses (78.57%). In the Boy-Girl situation, on the other hand, it was the pattern of complaint (a subcategory of non-intrusive acts), which turned out to be dominant (72.72%).
Illocutionary goals used in task 3
This validation study tried to investigate, whether or not the three widely used kinds of DCT, namely OWDCT, DWDCT and DRPT are, in fact, valid or not. Among different validation procedures the procedure of convergence and diverge analysis was utilized. Two forms of the DCT, namely DWDCT and DRPT converged in their final results indicating that in the Iranian culture solidarity is usually overridden by considerations of deference when and where, the addressee is of the opposite sex as that of the speaker. In this way, it was found that boys employ conflictives, which have the most impolite intention, mostly when and where, the addressee is of the same sex. Such pattern was not found in the OWDCT. For this reason, this study argues against the use of OWDCT at least for the investigation of the politeness strategies. This study does not, however, try to convey the idea that the two validated forms of DCT are without any limitations. According to Eslami-Rasekh (2005), data collected through a DCT may not be exactly as rich and complex as natural data. Billmyer and Varghese (2000) mention that in some validation studies discrepancies between DCT and spoken data have also been observed. Golato (2003) considers a DCT as an off-line task in which, the respondent has enough time for conscious concentration and introspection. This seems to be totally different from an authentic, on-line task, in which language is produced spontaneously with no conscious focusing on the linguistic output. Yuan (2001) argues that oral and written DCTs share similar drawbacks in eliciting compliments and compliment responses. Yuan is of the opinion that DCTs are more appropriate to the study of what people think they would say than to the study of what people actually do say in a given speech setting.
Appendix A: Please complete the missing parts. You should disagree with your close friend.
Asghar (Maryam)* held my hand and dragged me to the courtyard and once I saw the minarets I hankered for them. After we walked for a while, I asked him (her) Why are the minarets incomplete? He (she) said I dont know, I think when Moayyer died they were left incomplete. People say his sons were clumsy. I asked Who is Moayyer? He (she) said I dont know. I should ask my dad.
I wanna climb one of them.
He (she) said
It is impossible. Even the muezzin cannot climb it.
I said .
As we reached the roof of the mosque the children saw us and started to shout. We rushed into the staircase of the minaret. Asghar (Maryam) was in front of me. The staircase was circular, round and narrow, so we couldnt go fast. We could still hear the children shouting. Asghar (Maryam) said I think the staircase is going to break.
I said .
Asghar (Maryam) went one step higher to the extent that children could see his (her) head. He (she) said It is too cold here. Lets get down. I said Wait. Guess how much the top of the minaret is higher than us.
Asghar again (Maryam) said
Its too cold. Lets get down.
I wish the minarets had been completed.
Asghar (Maryam) said
Oh! The principal is staring at us.
I want to climb higher
He (she) said
Are you an idiot? The wind will make you fall and the principal will punish us.
I said .
Note: Asghar is a boy name where as Maryam is a girl name.
Appendix B: Please read the following descriptions of situations and then write what you would SAY in each situation.
You want to climb a minaret, but your very close friend (male/female) insists that this is impossible since even the muezzin cannot do so
You and your very close friend (male/female) are climbing a narrow staircase but your friend keeps telling you that the staircase is going to break
While, you and your friend (male/female) are on the top of the minaret, your friend tells you that the harsh wind is going to make both of you fall on the ground of the school and the school principal will punish you both
Appendix C: Please read the following descriptions of situations and then play the roles of the person in that situation.
You want to climb a minaret but you very close friend (male/female) insists that this is impossible
You and your very close friend (male/female) are climbing a narrow staircase but your friend tells you that it is going to break
|•||While, you and your very close friend (male/female) are on the top of the minaret, your friend tells you that the harsh wind is going to make both of you fall on the ground of the school and the school principal will punish you both|