Testing the Natural Order Hypothesis on the Framework of the Competition Model

August 2007. Volume 2 Issue 1

Title
Testing the Natural Order Hypothesis on the Framework of The Competition Model

Authors
Forood Sepassi & Seyed Vahid Aryadoust

Shiraz Azad University, Iran

Bio Data:

Forood Sepassi, Ph.D., is an associate professor at Shiraz Azad University and is interested in interested in Psycholinguistics. He supervises MA theses and has published papers in Asian EFL Journal.
Seyed Vahid Aryadoust has an MA in TEFL from Shiraz Azad University. He has written seven books on different subjects such as Sociolinguistics, Pragmatics, IELTS, and general English. His areas of interest are psycholinguistics, second language acquisition, testing (especially teaching and studying the validity and reliability of IELTS, TOEFL iBT and PBT), Sociolinguistics, and Pragmatics.

PDF E-book version pps. 1- 130 view
SWF Version view


 

Abstract
The issue of natural order in acquiring some morphemes in English has been a controversial one in psycholinguistics and applied linguistics. This article seeks to test Krashen’s Natural Order Hypothesis employing the Competition Model among two age groups of pre-adolescent and adult age in an EFL setting. Research on this issue has been mostly conducted in ESL settings and the results have been generalized to EFL environments. The present study conducted with 60 Iranian EFL learners in two different age groups of 12 to 13 and 17 above. The participants were asked to say the sentences they heard and meanwhile apply changes wherever necessary. The results showed that, contrary to what Krashen has claimed neither plural –s nor third person –s appears to be acquired earlier.

Key words: natural language processing, Natural Order Hypothesis, Competition Model, Critical Age Hypothesis.

1. Introduction
A controversial issue in psycholinguistics has been the relationship between learner age and patterns of language acquisition. During the twentieth century, many notable educators have argued for “an early age start in second language learning”. On the other hand, some other researchers have found that older learners tend to syntactically outperform the younger ones (Krashen, 1982, pp. 166-168). These claims brought about the conclusion that the age differences among learners may play a significant role in learning a 2nd language. Hence, the present study aims at studying the behavior of two groups of EFL learners, to correspond to the Critical Age Theory of Lenneberg (1967).

1.1. Objectives
The hypothetical framework on which this study rests is attributed to studies of L1 and L2 processing. Nonetheless, Krashen (1988) states that there might be a difference in findings upon administrating the same or similar tests to EFL learners. In the realm of EFL, in reality, few studies have been conducted to shed light on the issue of naturally-ordered learning of grammatical morphemes among learners before and after their critical period. It is the aim of this study to employ the psycholinguistic ”Competition Model” as a practical tool to provide further insight on how the age of EFL learners affects the order in which they learn two morphemes, namely the third person singular (-s) and the plural (-s).

1.2. Significance of the study
The present study has both empirical and theoretical implications. The present study might be particularly useful in explicating the way EFL learners internally process the language, that they are learning, in this case English. In addition, in the light of the task invoked, we may gain more insight into the order of acquisition as an aid to syllabus designers, EFL teachers, and educators in developing appropriate syllabi and pertinent teaching techniques. This research study sought to examine the following issues:
1. The order through which the morpheme of third person (-s) and the morpheme of plural (-s) in young learners under 13 or in adult learners above 17 in an EFL context are acquired.
2. Finding out if the Natural Order Hypothesis is valid for EFL young learners who are below and/or above the age of puberty.
3. The way the competition among cues plays roles in distinguishing the morphemes needed to be used in their appropriate position.

2. Review of Literature

2.1. Theoretical Framework

In the second half of the twentieth century, research on first language acquisition and its application in classroom drew the attention of linguists of every persuasion as well as educators. Most standard textbooks and curricula are based on first language acquisition, so that the order in which grammatical morphemes are presented in an educational setting is based on data from first language acquisition.
But it may not be possible to generalize the findings from first language research to the second language. One must bear in mind that first language acquisition starts in childhood and 2nd language learning is mostly related to adulthood. The idea of putting 1st and 2nd language acquisition within the same framework, brought about a lot of new findings and at the same time created serious problems in the realm of linguistics and teaching. Brown (2000; p.51), for instance, mentions “…second language researchers and foreign language teachers began to recognize the mistakes in drawing direct global analogies between first and second language acquisitions”.

2.1.1. The Critical Period Hypothesis (C.P.H.)
There is ample evidence that language acquisition is more a function of the age of the learner than any other factor. In relation to this idea, Lenneberg (1967) argued for the crucial effect of the age factor. He was of the opinion that the critical age period ranges from two to puberty. The cornerstone of his hypothesis is the concept of Brain Lateralization (Cerebral Dominance). Many have attempted to define the term “Lateralization” as precisely as possible. Among them, Richards and Schmidt (2002, p.68) mention “Lateralization [is] the development of control over different functions in different parts of the brain…Those parts of the brain that control language are usually in the left hemisphere”.
In Lenneberg’s view, due to the fact that language acquisition period terminates at puberty (i.e. as a result of lateralization), post-adolescent language acquisition is difficult. Hence, complete learning of a second language could not be possible after lateralization takes place. Mention should be made of the opponents of this hypothesis who are of the opinion that although there is a gradual decline in acquiring a second language, there cannot be any sharp breaks spotted to be called The Critical Age. That is why they have preferred to use another terminology, Sensitive Period. This latter idea, however, is more subject to dispute in comparison with the C.A.H.
The dominant standpoint, amongst a large amount of literature, is that there certainly exists a phenomenon by the name of lateralization. The argument is about the exact age when this process reaches completion. Krashen (1980), among many others, proposed a much earlier age, five. The same hypothesis was put forward by Scovel (1984, p. 57): “One must be careful to distinguish between ‘emergence’ of lateralization (at birth, but quite evident at five) and ‘completion’ (only evident at about puberty)”.

2.1.2. The Natural Order Hypothesis
The N.O.H. is defined as:
“The hypothesis that children acquiring their first language acquire linguistic forms, rules, and items in a similar order. For example, in English children acquire progressive -ing, plural -s, and active sentences before they acquire third person –s on verbs, or passive sentences” (Richards & Schmidt, 2002). This hypothesis was proposed by Krashen (1980). He believed that there was no noticeable difference, regarding the order of learning grammatical morphemes, between those whose exposure to second language takes place in the classroom and learners who are in an informal (out-of-classroom) context.
The first evidence for the Natural Order Hypothesis appears to come from ESL studies. “In 1974, Dulay and Burt published a study of what they called the order of acquisition of grammatical morphemes or ‘factors’ in English by five-to-eight-year-old children learning English as a second language” (McLaughlin, 1991, p. 89). In their research, Dulay and Burt utilized the Bilingual Syntax Measure, presenting cartoon pictures followed by questions. The idea was based on Brown (1973), which set forth an ‘invariant’ or common sequence “of acquisition for at least 14 factors, or function words in English that have a minor role in conveying sentence meaning” (McLaughlin, 1987). Chastain (1988, p. 74), trying to put the issue more into perspective, suggests that this does not “mean that all learners acquire language in the same order at the same time, but that similarities exist among learners and that they will learn some STRUCTURES early and some late”.
Table 1.1, from Krashen (1977) cited in Krashen (1987), illustrates an average order of second language acquisition.

Table 1.1. Average order of second language acquisition, in English

2.1.3. The Competition Model (C.M.)
Like other data-driven, connectionist models, the Competition Model allows statistical properties of the input to play a major role in determining the order of acquisition as well as the nature of the final state (MacWhinney & Bates, 1984) In point of fact, the existing language acquisition models have not yet taken seriously two essential facets of the acquisition process, both of which offer contribution to its variability: 1. cross-linguistic variation. 2. Individual differences amongst learners within a particular language.

The Competition Model relies on four cues – two grammatical, one prosodic, and one semantic: preverbal placement, agreement markers, stress, and animacy. These cues map the level of form to the level of function. They also compete with each other in order to grasp more space in the mind. In fact, the model holds “Whatever the speaker wants to communicate has to be achieved through these four…..So, the more a language uses information, the less it can rely on word order, the more emphasis it has on word-forms, the less on word order; and so on.” (Cook, 1991, p. 124)

2.1.3.1. Competition and Direct Mapping
Competition is the basic notion in most information-processing models in cognitive psychology. The C.M. computational model is similar to those occurring in other cognitive models of acquisition. The cues may unanimously boost the identification of the subject/agent, or one of them may undermine such a candidacy when in a state of conflict or competition with other cue(s). (Heilenman & MacDonald, 1993)

Direct mapping, on the other hand, models performance using only two levels of representation; namely, a function level (where all meanings and forms correspond) and a formal level (where all the surface forms or expressive devices available in the language are represented) (MacWhinney & Bates, 1984). The following instances help to illuminate the mechanism of mapping:

1. The teacher teaches the students.
In statement (1), three cues – preverbal placement, noun animacy, and verb agreement- in conjunction with each other boost the candidacy of the first noun phrase, the teacher, as the actor/ agent/ topic/ and subject.

2. The machine teaches the children.
Statement (2) presents two grammatical cues, preverbal placement and singular verb agreement, to indicate the first noun phrase as the actor/ agent/ topic/ and subject.
3. The books are compiled by the teachers.
Here; one observes that the functions may not necessarily co-exist. ”The book” is the patient placed in the preverbal position rather than the agent. Noun animacy and the preposition (by) collaborate to boost the candidacy of the teachers as agent/actor.

2.1.3.2. Cue Validity
Psychological mechanisms in human beings attach information value or validity to cues. In other words, validity is an objective property of certain cues. Cue validity, in essence, is ”the major predictive construct in the Competition Model. The theory of cue validity is composed of two components: cue availability and cue reliability” (MacWhinney and Bates, 1984)

2.1.3.3. Cue Availability
This concept exercises a significant role in this study it is defined as how often a piece of information is offered during a decision making process. It is presented in terms of numbers in McDonald’s (1984) scheme as the following proportion:
“Cases wherein cue is available the total number of cases in a task domain to take an example, the availability of accusative case (ra) in Persian is very high. The same case is available in some other languages, such as German, but not as frequent as that in Persian. Needless to say, in modern English, accusative case has no availability.” (Sepassi, 2002)

2.1.3.4. Cue Reliability
Another component in the Competition Model seeks to clarify how often the cue leads to a correct conclusion when it is used. This concept can be numerically expressed as the following fraction:
The cases in which a cue leads to the correct conclusion over the number of cases in which it is available; for instance, preverbal position is a highly reliable cue in English; , for, it is always assigned to the agent of a transitive action. Nevertheless, it is not reliable at all in Persian since OV and SOV syntactical constructions are possible.

2.1.3.5. Cue Strength
Unlike cue validity, cue strength is a subjective property of the individual’s knowledge. Being quintessentially a connectionist notion, ”it refers to the probability or weight that the organism attaches to a given piece of information relative to some goal (MacWhinney, 1978, p. 8). ”In psycholinguistics, a link between two levels is assumed to make an instantiation out of it. On the one hand, the surface form is considered and, on the other hand, an underlying function. The link between these two is given a weight or strength.”

2.1.3.6. Ongoing Updating
This theory holds that the parser engages in an ongoing updating of assignments of noun to roles. For example, when parsing a sentence such as ”the dogs are chasing the cat”, the assignment of “dogs” as the agent is first promoted by its appearance as the initial noun. Then the fact that ”are chasing” agrees with ”dogs” in number further supports its assignment. Finally, when the singular noun ”cat” appears post-verbally, its binding to the object case role further supports the candidacy of ”dogs” as the agents. Thus, at each point in sentence processing the mapping from the lexical item ”dogs” to the agent role is updated. In this particular case, each updating increases the strength of this assignment (MacWhinney, 1978). In other words, the concept of competition is founded on the assumption that as the parser moves through the sentence, cues are used to boost or undermine the candidacy of each noun for the agency role. This process is also referred to as ”ongoing updating” (MacWhinney, 1989; cited in Sepassi, 2002).

2.1.3.7. Perceivability
Perceivability refers ”to the extent to which a listener encounters difficulty in trying to detect a cue in sentence processing” (MacWhinney, 1989, p.179). “If a subject is asked to correct a grammatically ill-formed sentence like: “The drivers drive the cars”, she might invoke [at least] two ways the former of which would be:

The drivers drive the cars
And the latter is:

The driver drives the cars.
In correction number 1, the parser might have encountered difficulty in attempting to properly detect the third person –s cue in sentence processing. Therefore, this correction lacks the third person –s perceivability. A similar analogy can be made about the second way of correcting the sentence.” (MacWhinney, Personal correspondence, February 5, 2006)

2.2. Empirical Literature
Besides Krashen, other researchers have attempted to substantiate that there is an order of acquiring various grammatical morphemes in English as a first language. Brown (2000), for instance, “demonstrates that on obligatory occasions certain morphemes, such as -ing and plural -s, tend to be acquired relatively early, while others, such as the third person singular -s on verb in the present tense or the possessive ‘s marker tend to be acquired late. Brown’s longitudinal (study) findings were confirmed cross-sectionally by De Villiers and De Villiers (1973). The discovery was extended to child second language acquisition by Dulay and Burt (1972; 1974; 1976) in several cross sectional studies….”
Bailey, Madden, and Krashen (1974) employing a SLOPE test, found that adult 2nd language learners displayed a ”natural order” for eight grammatical morphemes. The study showed that there was no difference in rank order between Spanish Speakers and non-Spanish Speakers. (SLOPE stands for Second Language Oral Production Exam, a test which estimates general oral ability of candidates by having them produce specific English grammatical structures. Then, diagnostic scores are produced on these structures (Moussavi, 1999). This test was designed by Ann Fathman (1976), who employed it to collect evidence that the previous findings by Krashen and other researchers mentioned above were valid.
In another study, Roger Anderson (1978) made use of compositions in Puerto Rico. He, too, obtained some similar results authenticating the natural order reported by Krashen and others. Soon after this study, Krashen, Butler, Birnbaum, and Robertson (1978) employed a composition task to add to the body of evidence corroborating the order of acquiring the mentioned morphemes. They asked subjects, all of whom were ESL students, to write compositions under two conditions. First, they were expected to write ”fast”; they were required to write as much as possible about a given topic in a short time. After completing this task, they were encouraged to go over their work carefully and edit it. After collecting data and interpreting them, the team of researchers found a natural order under both conditions.
The result was interpreted as showing that students were concerned with communication when writing rather than with form; apparently, the ”focus on form” condition for monitor use… is more crucial for bringing out the monitor than is the ”time” condition…this is not to say they did not edit – it does imply that when they do edit, they do not use their conscious knowledge to any great extent when communicating (Krashen, 1988).

3. Methodology
3.1. Participants
This study was conducted with 60 intermediate pre-adolescent learners of 12 to 13 and 17 to 20 age groups all of whom had been learning English in Dibaagaran Language Institute (DLL), Shiraz, Iran. The reasons for selecting these two groups of students is that first of all they are of similar socio-economic background as well as the fact that the younger group had not yet found the opportunity to analyze syntactical structures in English as the adults had. Apart from the other practical hints mentioned on employing the C.M., it is deemed crucial to make use of the two different age groups with the above-mentioned characteristics in addition to strictly controlling some intervening factors, such as the task and behaviors elicited (MacWhinney, personal correspondence, February 14, 2006).
At the time of the interview, the subjects who had just started their intermediate-level course had been put through a program exposing them to 190 hours of formal instruction in English L2. Furthermore, the participants’ grade point average for their previous term fell in the range of 85 to 95 out of 100. In teaching the subjects, a homogeneous methodology and syllabi had been used. Then, the sample was classified into two distinct age groups:

12-13-year­-olds, to accord with Lenneberg’s Critical Age Hypothesis which considers this two-year span as the point where the individual’s language acquisition ability hits a trough;
17-20- year-olds, to take into consideration the gradual loss of control over the motor skills in the adult learners.

3. 2. Stimuli
A list of 24 audio-taped sentences was prepared. In order to prepare the sample sentences, a vocabulary pool containing some animate nouns, inanimate nouns, and transitive verbs was made. Every sentence contained an animate agent/subject, a transitive verb, and an inanimate object/patient following the pattern of Np1, Vp, Np2. Three sentences were formed for each one of the eight patterns (see Table 3.1), totaling 24 sentences. In order to control every intervening variable, the following measures were taken:

The first noun phrases (Np1) were selected to be animate.
All verbs (Vps) were chosen to be transitive ones so that they would be followed by a direct object.
All the second noun phrases (Np2) juxtaposed after the verbs were arranged to be inanimate.
Some of the stimulus sentences are as:
The salesperson sells the houses.
The waiter cleans the tables.
The robber steals the bicycles.
One more important point was the fact that all words, verbs and nouns, in the preliminary vocabulary pool were chosen pertinent to and within the proficiency level of the candidates (see appendix C). A complete list of all 24 sentences is provided in appendix B. The 24 sentences in connection with the possibility of having plural (-s) in subject and object positions as well as the possibility of including or excluding grammar violation had the following structure:

Table 3.1. Possible patterns used to make sample sentences

 

Subject (Np1)

Verb (Vp)

Object (Np2)

*1

Plural -s

Third person -s

Plural –s

*2

Plural -s

Third person -s

O (no plural –s on Np2)

 3

Plural -s

O (no third person
-s on the verb)

O (no plural –s on Np2)

 4

Plural -s

O (no third person
-s on the verb)

Plural –s

 5

O (no plural –s on Np1)

Third person -s

Plural –s

*6

O (no plural –s on Np1 )

O (no third person
-s on the verb)

Plural –s

 7

O (no plural –s on Np1)

Third person -s

O (no plural –s on Np2)

*8

O (no plural –s on Np1)

O (no third person
-s on the verb)

O (no plural –s on Np2)

[Ungrammatical patterns are indicated with *]
Most of these sentences had some competitions embedded in them. They were to be played for the candidates and the candidates were to repeat them and, in the meantime, correct the grammatical violation they might run into. So, it was deemed necessary that the subjects be told that they would encounter grammar violations which they had to correct orally (MacWhinney, personal correspondence, January 26, 2006). It is also useful to mention that every sentence was likely to be modified in eight ways by the participants; four of these are held to be correct responses. These correct responses are as follows:
[SOS], for example The drivers drive the cars.
[SOO], for example The boys eat the apple.
[OSS], for example The doctor sees the books.
[OSO], for example The pilot flies the plane.

To elaborate, we can conclude that in correction types of SOS and SOO, cue validity and cue availability of plural –s are in higher than those of third person –s. However, in OSS OSO corrections, cue validity and cue availability of third person –s are higher because the subjects have paid attention and used third person –s to correct the sentences.
The utterances of the participants were tape-recorded. The tape was reviewed thoroughly later in order to analyze the data.

3.3. Administration procedure
The task was conducted over the course of one week, from April 14 to April 21, 2006. Each single administration lasted for about 5 minutes. Owing to the fact that the researcher, himself, was the only interviewer, in order to minimize the fatigue attributed to protracting sessions, in every sitting no more than 6 subjects were interviewed. The participants were seated, one at a time, before the interviewer across a table. As mentioned before, two tape recorders were made use of, one to play the set of pre-recorded sentences and the other to record the participant’s responses. One potential problem was the anxiety that might have arisen in the subjects. In order to alleviate this problem, the subjects were assured the task would not have any effect on their class grades.
After this, the participants were supplied with instructions as follows:

They were told that the sentences might contain grammatical violations which had to be corrected, or might be free of them.
In order not to activate their monitor, no wrong production by the subjects was corrected. They, hence, were assured that the interviewer would not interfere with their responding process. This, in turn, mitigated their affective condition in that they were not disturbed by the pressure of receiving negative feedbacks from the interviewer.
The participants were supposed to withhold their replies until the completion of the sentence that was being played.
After playing each sentence, a five-second interval for the onset of production of each sentence was incorporated into fulfilling the task. They were allotted 10 seconds to commence, continue, and terminate every sentence, then. For each participant, production of all the responses took about 5 minutes.

4. Results and Discussion
4.1. Plural -s on Np1 and third person -s on Vp in production
After obtaining the results from both age-groups, the raw data were tabulated for analysis. Recall that Krashen (1980) reports that the ESL learners of the age before lateralization acquire the plural (-s) before the third person (-s). In Competition Model terms, this means that Krashen assumes that the cue validity and reliability as well as cue strength of plural –s are more than those of third person –s and hence the subjects have mastered them earlier. Many test results obtained from various studies substantiate the claim. The same analogy has been made about the ESL learners who have just passed their lateralization period. Moreover, Krashen (1980) asserts that regardless of the type of learning, whether in an educational setting or in a community, the same order of acquisition is always observed.
It appeared from the analysis of the results of the present study that the distribution of the data was not normal. Therefore, we employed a Mann-Whitney test in order to compare the obtained mean instances of the four-hitherto mentioned correction types in group 1 (adult learners) and those in group 2 (pre-adolescent learners), independently. The following graphs show the non-normal distribution of the data gathered. Also, the P-P Plot graphs corroborate the fact that the distribution of the data was not normal (see Appendix A).
Graphs 4.1 and 4.2 provide an overview of the overall distribution of the two correction forms that confirm the attention the subjects paid to employing plural –s on Np1 and third person –s on Vp, respectively. It is clear from the graphs (histograms) that the distributions are not normal. Hence, it was decided to utilize the non-parametric tests, i.e. Mann-Whitney and Wilcoxon, which are useful for situations where the data are not normally distributed.

Graph 4.1 The distribution of the sum of the two correction types showing the attention paid to plural -s on Np1; that is [SOO] and [SOS].

After tabulating the raw data, a Mann-Whitney test revealed that the sum of corrections of types [SOO] and [SOS] used by the subjects to reproduce the sample sentences was dominant in group 1, i.e. the obtained P-value of 0.12 was not statistically significant. They, therefore, do not pay enough attention to plural –s, which is evidence to claim they have not mastered this structure yet. So, the immediate conclusion is that the performances in this group did not show a significantly high cue reliability and validity for the plural marker on the subject.

In addition, none of the other behaviors, i.e. the correction types [OSO] and [OSS], could be in line with the Critical Age Hypothesis’ claim. The mean reports in Table 4.1. demonstrated high standard deviations in both groups. A comparison of this study’s learners and Krashen’s ESL learners revealed that although within both age groups the ESL learners in an English-speaking environment followed the order of learning of the two morphemes, as claimed by Krashen (1988) and others, EFL learners did not display the same behavior. The tables below show how many [SOS], [SOO], [OSS], and [OSO] structures have been produced by the subjects. These structures are considered grammatically correct reproductions.

Table 4.1.Instances of the Four Correct Reproductions in Group(s) 1 & 2

Group

C1(SOO)

C2 (SOS)

C3 (OSS)

C4 (OSO)

Mean
N
Std. Deviation

2.0333
30
1.58622

4.8667
30
2.16131

1.0667
30
1.08066

6.4667
30
3.12645

Mean
N
Std. Deviation

1.0667
30
1.36289

4.5000
30
1.88917

1.1000
30
.80301

7.7000
30
3.43561

Total    Mean
N
Std. Deviation

1.5500
60
1.54509

4.6833
60
2.02101

1.0833
60
.94406

7.0833
60
3.31556

  p-value

0.12

 0.6

0.76

 0.19

Then, a Wilcoxon test was used to compare the sum of mean instances of the structures illustrating the tendency of the participants towards employing either plural –s on Np1 or third person –s on Vp. In other words, the sum of [SOO] and [SOS] (labeled C1 & C2) was compared with that of [OSO] and [OSS] (labeled C3 & C4) in order to find out if third person –s would outweigh the production of plural –s in Np1 or vice versa. The summary of this comparison is given in Tables 4.2 and 4.3.

In Table 4.2., the mean instances of 6.9 and 5.667 for the age groups 1 and 2 respectively showed that the subjects did not attend to plural –s on Np1. Similarly, the means 7.5333 and 8.8 for groups 1 and 2 proved the lack of grasping third person morpheme -s on the verbs. The conclusion is that one can claim that both age groups had neither a strong command of plural -s nor of third person -s morphemes.
Table 4.2 Mean and Standard Deviations of the subjects’ performances

Group

Sum12 (C1&C2)

Sum34 (C3&C4)

1          Mean
                   N
         Std. Deviation

6.9000
30
3.38710

7.5333
30
2.88556

2          Mean
                   N
     Std. Deviation

5.5667
30
2.0

8.8000
30
3.39777

Total          Mean
                   N
        Std. Deviation   

6.2333
60
2.87223

8.1667
60
3.18985

P-value

0.177

             0.103

Finally, Table 4.3 provides two p-values, 0.177 and 0.103, which are not statistically significant. A direct conclusion is that, there is no systematic inclination towards employing either the plural -s on Np1 or the third person -s on Vp.

Table 4.3 The Results of Mann-Whitney, Wilcoxon Tests, and P-values (Inferential statistics)

Test & Obtained Values

Sum12

Sum34

Mann-Whitney

359.500

340.500

Wilcoxon

824.500

805.500

P-value

0.177

0.103

4.2. Plural -s on Np2
In order to find out the significance of the data, Chi-square and the p-value of these percentiles were calculated. The Chi-square of 5.212 and the p-value of .2 appeared not to be statistically significant. Hence, in correct reproductions containing plural –s on Np2, there was no significant difference between the two age group learners. Therefore, one might claim that they had not attached a higher significance to employing plural –s on Np2 and this, in addition to the previously mentioned performance of the participants, provides evidence that they have not mastered the structures, yet. The results are demonstrated in Table 4.4.
Table 4.4 Pearson Chi-Square Value, relevant to Plural -s on Np2

 

Value

df

P-Value

Pearson Chi-Square

5.212b

1

.022

The total of correct performances and incorrect ones containing plural –s on Np2 is summarized in Table 4.5. The table indicates no considerable difference in the percentage of Np2 responses which contain the plural –s. This difference again indicated that grasping -s as a plural-maker morpheme on Np2 in Group 1 (adult learners) and in Group 2 (young learners) is similarly non-significant. Table 4.5 provides a Chi-Square value of 9.192 and a p-value of .2.

Table 4.5 The total of correct performances and incorrect ones containing plural –s on Np2

Chi-Square value        P-value               Df
              9.192         0.2                1

And finally, to check that there are no differences due to the different sets of words used to exemplify the eight patterns, an ANOVA was run. The results are shown in the following table, table 4.6.

Table 4.6 ANOVA run to make sure there are no differences due to the different sets of words used

   

Sum of Squares

df

Mean Square

F

Sig.

sum12

Between Groups

44.844

5

8.969

1.096

.373

 

Within Groups

441.889

54

8.183

   
 

Total

486.733

59

     

sum34

Between Groups

80.358

5

16.072

1.669

.158

 

Within Groups

519.975

54

9.629

   
 

Total

600.333

59

     

The table above demonstrates there is no significant result observed.

5. Conclusions and Educational Implications
5.1. Testing the Natural Order Hypothesis and the C.M.
The Natural Order Hypothesis, proposed by Krashen (1980), holds that regardless of whether the learners’ setting or their exposure to the SL has been in an informal context, all individuals will show the same pattern in learning the basic grammatical morphemes. The explanation that the C.M. may provide is to learn any special structure in advance of others means that in the competition among cues, the structure which was preferred to the others has outweighed them. In this regard, to claim learning plural –s always precedes the third person –s means that the plural –s enjoys a higher cue availability. That is to say, the following fraction will be a considerable one:

Cases wherein the cue is available
______________________________________ = Cue availability
The total number of cases in a task domain

In this study, the statistics did not demonstrate any significant importance attached to either of the morphemes. That is, no order of acquisition, which is manifested through the attention paid by the learners to the morphemes in the stimulus sentences, was observed. The cue availability of both plural –s and third person –s as well as their reliability were found too low and they were considered not to be mastered by the EFL learners in both age groups. As a result, one may claim, based on the results of this study, that the hypothesis postulated by Krashen (1980) regarding the order of acquisition may not be true with regard to EFL learners. That is to say, neither of the two groups of candidates showed a tendency to prefer plural –s in Nps or third person –s in VP. It appears that even age did not work as a factor in acquiring plural and third person –s as the cue reliability of the performances of the two age groups are not statistically significant.

The reason may be attributed to the fact that Krashen and others mentioned in this study had collected data in the settings where English was spoken as the first language and learned as the second one. They, then, generalized the results to the EFL settings. However, the learners’ exposure to English in these two settings is very different. ESL learners have certainly a better opportunity to ‘pick up’ the language and may follow the order claimed by Krashen. However, the EFL learners do not have this opportunity and their exposure to English is far less than that of the former group. Thus, it is not appropriate to over-generalize the results of the studies carried out in ESL contexts to EFL contexts.
Meanwhile, one must highlight the fact that in both groups the comparison between two kinds of performance carried out by the candidates indicated that the structure [SOO] in comparison with [SOS] and the structure [OSO] in comparison with [OSS] in each group showed lower strength. The P-value obtained for groups 1 and 2 in both cases is 0.2 which is evidence that the cue validity, cue availability, and cue reliability of the structures containing plural –s on their Np2 are not significant. Further, the subjects displayed poor perception in that their attempts to properly spot the third person –s or the plural –s cues in sentence processing failed.

In the present writers’ opinion, there can be two justifications for the subjects’ behavior. One is that their performance proves no more significant attentiveness to the plural –s on the first and last Np heard in the stream of input than their inattentiveness to plural –s on both first and last Np.

The other justification concerns the competition between cues. Namely, in the competition between plural –s, third person –s, and Np1, the winner which turns out to be dominant is animacy. On the other hand, the animacy cue seems to be a more available, valid, and reliable cue in its competition against third person –s and plural –s. It can even prepare grounds for conducting further research studies. However, we should remember one important point. A huge unpredicted pattern in the data may have rather interesting implications for the Competition Model. The singular verb (-s marked) is likely to be used – and ONLY likely to be used – when BOTH the subject and object are singular, and the plural verb (0 marked) is similarly likely to be used only when both subject and object are plural. This is not a pattern of either language (Farsi which is the first language of the participants and English, their second language), but it suggests that the free word order and agreement morphology of Farsi may be deeply affecting the participants’ response to English agreement. Even explaining it in the C.M. is novel, because the explanation may require transference of the bi-directional subject-verb agreement cue from Farsi to correct subject-verb agreement in English plus a non-existent cue of noun-after-verb agreement in English.

Also, in the light of the opposed findings of the present study and those of Krashen’s (1980) study, the following explanations may be offered. First, as mentioned above, Krashen’s studies were completely conducted in an ESL context whereas the candidates participating in the present study were all learning their English in an EFL context. This fact also shows that the anticipation that he made a perfunctory attempt in conducting studies in EFL contexts. Meanwhile, he strongly claimed the possibility of generalizing his findings about the ESL learners to EFL learners.

Second, the subjects of the present study had all been exposed to L2 in a planned setting, using a special English material. Clearly, the more an individual is exposed to a special structure, the higher the possibility to acquire it. However, Krashen’s (1980) conjecture is deemed inappropriate in that his claim does not elucidate what EFL learners he has meant and with what background. In addition, it is also necessary to indicate the hours of the EFL learner’s exposure to English. It goes without saying that ESL learners are ‘immersed’ in the context of the second language they are learning. But this condition is not met when turning to the EFL learners. Hence, the analogy made by Krashen (1980) does not seem to be logical.

Furthermore, one must bear in mind that residing in an L2 environment always facilitates learning the target language. And due to the haphazard nature of L2 input in a natural environment, it is impossible to claim that the subjects of the studies of the same nature had the advantage of the same load of input as the ESL learners do. To recapitulate briefly, it appears very difficult to generalize the findings in a second language acquisition context to a FL (foreign language) setting.

Finally, to explain the observed behavior of the participants, there may be two possible explanations: the failure to show a difference in the rate acquisition of the plural and third singular endings might be because that none of the students have really learned either morpheme, yet. Or else, that novel paradigm used by the researchers might not be sensitive enough to pick up the amount of learning of each that has taken place. Thus, there are future plans to look for further evidence on both of those points.

5.2. Educational Implications
Talking of the pedagogical overtones of the findings, we can point out two important considerations drawn upon the results attained in this study. A major effort needs to be undertaken in re-designing those syllabi that rest heavily on the theories that resemble Krashen’s. Further, perhaps both teachers and designers need to lay more emphasis on a grammar which is more morpheme-based. This routine could be more practical in EFL environments in replacing a bundle of materials used in an EFL context, but originally designed for the ESL environments.

However, the considerable difference in producing the plural –s in Np2 might be an indicator that it is about time that the language schools paid more attention to the age of learners. More often than not, due to extrinsic considerations in abundance, EFL learners of different ages are enrolled in the same class, but more attention must be paid to the learners and their preferences according to their age. It is hoped that the linguists take steps in order to clarify the fuzzy borders between EFL and ESL and, on the whole, second language acquisition (SLA) and learning a foreign language (FLA).

APPENDIX A

Detrended Normal P-P Plot of the sum of the corrections [SOO] and [SOS] or Sum12, demonstrating the non-normal distribution of the sum of these two correction forms, necessitating employing a Wilcoxon test

Detrended Normal P-P Plot of the sum of the corrections [OSS] and [OSO] or Sum12, demonstrating the non-normal distribution of the sum of these two correction forms, necessitating employing a Wilcoxon test.

APPENDIX B
Stimulus sentences used in the interview with the subjects of the study
1. The drivers drives the taxis.
2. The students studies the book.
3. The chefs cook the food.
4. The engineers repair the cars.
5. The salesperson sells the house.
6. The officer wears the uniform.
7. The dog watch the yard.
8. The animal eat the orange.
9. The dogs eat the apples.
10. The horses drinks the water.
11. The doctors read the book.
12. The professors write the dictionaries.
13. The waiter cleans the tables.
14. The donkey eats the apple.
15. The girl buy the book.
16. The boy hit the door.
17. The nurses serves the hospitals.
18. The players kicks the ball.
19. The boys smoke the cigars.
20. The musicians love the pianos.
21. The camel gets the stone.
22. The teacher ride the motorcycle.
23. The customer buys the orange.
24. The girl steals the bicycles.

APPENDIX C
The vocabulary pool used to make the stimulus sentences.

Part A, Animate Nouns to be used in Np1:
Salesperson, waiter, girl, officer, donkey, camel, dog, lawyer, teacher, animal, boy, customer, driver, dog, nurse, student, horse, player, pilot, doctor, chef, father, mother, engineer, professor, musician, steward, flight attendant, soldier.

Part B, Inanimate Nouns to be used in Np2:
Taxi, apple, hospital, book, water, ball, cigar, ocean, sea, food, river, car, dictionary, piano, house, table, bicycle, uniform, apple, stone, yard, book, motorcycle, wood, door, orange, tomato, phone, telephone, bag, pen, pencil, rope, milk,

Part C, Verbs to be used in Vp:
Sell, clean, steal, wear, eat, get, watch, read, ride, cut, hit, buy, drive, serve, study, drink, kick, smoke, visit, cook, love, repair, write, love, see, touch, punch, smell.

References

Andersen, R.W. (1978). Implicational model for second language research. Language Learning, 28, 221-282.

Bailey, N., Madden, C., & Krashen, S.D. (1974). IS there a natural sequence in adult language learning? Language Learning, 21, 235-243.

Birdsong, D. (1992). Ultimate attainment in second language acquisition. Language, 68, 706-55.

Brown, D.H. (1973). Affective variables in second language acquisition. Language Learning 23, 231-244.

Brown, H.D. (2000). Principles of language learning and teaching. San Francisco: Addison Wesley Longman.

Brown, R. (1973). A first language. Cambridge MA: Harvard University Press.

Chastain, K. (1988). Developing second language skills. Orlando, Fl.: Harcourt Brace Jovanovich.

Cook, V. (1991). Second language learning and language teaching. London: Edward Arnold.

De Villiers, J. & de Villiers, P. (1973). A cross-sectional study of the acquisition of grammatical morphemes in child speech. Journal of Psycholinguistic Research, 2, 267-278.

Dulay, H. & Burt, M. (1972). Goofing: An indicator of children’s second language learning strategies. Language Learning, 22, 235-252.

Dulay, H. & Burt, M. (1974). Natural sequence in child language acquisition. Language Learning 24, 37-53.

Dulay, H. & Burt, M. (1976). Creative construction in second language learning and teaching. Language Learning, Special Issue Number 4, 65-79.

Fathman, A. (1976). Variables affecting the successful learning of English as a second language. TESOL Quarterly, 10, 433-441.

Heilenman, L.K. & McDonald, J.L. (1993). Processing strategies in L2 learners of French: The role of transfer. Language Learning, 43(4), 507-557.

Krashen, S.D. (1980). The theoretical and practical relevance of simple codes in second language acquisition. In Scarcella, R. & Krashen, S.D. (Eds.), Research in second language acquisition (pp. 7-18). Rowley, Ma.: Newburry House.

Krashen, S.D. (1982). Principles and practice in second language learning and acquisition. Oxford: Pergamon.

Krashen, S.D. (1987). Principles and practice in second language acquisition. California: Prentice-Hall International (UK).

Krashen, S.D. (1988). Second language acquisition and second language learning. California: Prentice- Hall International (UK).

Lenneberg, E. (1967). Biological foundations of language. New York: Wiley.

MacWhinney, B. (1978). The acquisition of morphophonology. Monographs of the Society for Research in Child Development, 43 (serial No.174).

MacWhinney, B. & Bates, E. (1984). Cue validity and sentence interpretation in English, German, and Italian. Journal of Verbal Learning and Verbal Behavior, 23, 127-150

MacWhinney, B. (1989). The Competition Model. In B. Macwhinney (Ed.), Mechanisms of language acquisition. Hillsdale, NJ: Lawrence Erlbaum Associates.

McDonald, J.L. (1984). Semantic and syntactic processing cues used by first and second language learners of English, Dutch, and German. Unpublished doctoral dissertation, Carnegie-Mellon University, Pittsburgh PA.

McLaughlin, B. (1987). Theories of second language learning. Chicago III: The University of Chicago Press.

Moussavi, S.A. (1999). A dictionary of testing. Tehran: Rahnama publication.

Richards, J.C. & Schmidt, R. (2002). Longman dictionary of language teaching & applied linguistics. Essex: Pearson Education Limited.

Scovel, T. (1982). Questions concerning the application of neurolinguistic research to second language learning/teaching. TESOL Quarterly, 16, 323-331.

Sepassi, F. (2002). Intelligence-related constrains on information processing to serve as a platform for the enhancement of the descriptive adequacy of the Competition Model. Unpublished doctoral dissertation. Shiraz University, Iran.

 

Tags: , , ,

Category: 2007