Some Swedish students´learning of subject-verb agreement in English

(1)

Malmö högskola

Lärarutbildningen

Kultur-språk-medier

Examensarbete

15 högskolepoäng

Some Swedish students’ learning of

subject-verb agreement in English

Några svenska elevers beskrivning av hur de uppfattar lärandet av

engelskans subjekt- och verbkongruens

Mona Lindelöf

Lärarexamen 270 poäng Handledare: Ange handledare Huvudämne: Engelska

Höstterminen: 2007

Examinator: Bo Lundahl Handledare: Stefan Early

(2)

(3)

Abstract

Persons with Swedish as their first language often find it hard to learn subject-verb agreement when studying English. In Swedish this grammatical difficulty does not exist and it is

therefore hard to introduce to learners that have Swedish as their native language.

This investigation is based on the texts of 28 ninth graders of whom four were interviewed. My interest was in finding out how the students reflect on their own written work with a focus on subject-verb agreement with a particular focus on the third person singular s.

My study shows that the four interviewed students claim that they never reflect on grammar in their spontaneous writing and that they never consciously try to apply rules that they have studied in school through being offered grammatical explanations. Instead they make their grammatical choices intuitively, using their procedural knowledge.

Key words: English as a foreign language; subject-verb agreement; declarative knowledge; procedural knowledge; learner awareness

(4)

(5)

1 Introduction

It is a well-known fact that persons with Swedish as their first language have difficulties in using subject-verb agreement when learning English. In Swedish this grammatical structure does not exist and it is therefore hard to learn for learners who have Swedish as their native language. The structure as a grammatical rule might be easy to understand, but using the structure in spontaneous production is a different matter. In English, the third person singular agrees with the s form of the verb, and the verb “be” has its own form for each pronoun. At first I thought that subject-verb agreement in the English grammatical structure would be easy for students to learn and use – compared to Spanish where you inflect the verb for every subject and not only to the third person singular, but after correcting tests and texts written by ninth graders I was surprised. I realized that most of them were confused and did not know when to attach the inflected morpheme -s to the verb, which made it difficult for them to choose between “do” or “does”, or “work” or “works” in both cloze sentences and self-produced texts. Even though some of them had studied this intensively during weeks, most of them still had great difficulty in producing the correct verb form. This made me wonder if explicit teaching of grammar had any effect on students in their use of the target language and it made me curious to know how the students experienced grammar teaching of this rule.

What does it mean to know a grammatical rule? It is hard for a person to understand a new and unfamiliar grammatical structure and to actively learn it, but it is even more difficult to use it in a natural way when speaking or writing. In this sense, there are two different perspectives on grammar. It can be seen as either a product or a process, i.e. as the form of a grammatical rule or the grammar as a part of language use. These different approaches are often not related, i.e. the teaching of a grammatical rule does not necessarily lead to skills in the use of it in self-produced texts. This explains why teachers are surprised and disappointed when their teaching does not lead to the result that they have expected in the student’s free writing. In Språkdidaktik Ulrika Tornberg refers to knowledge about a grammatical rule as declarative knowledge, and procedural knowledge as the ability to use these rules in communication (2001:62).

(8)

When teaching grammar, you have to take into consideration the developmental stage that the students are at. Pienemann claims that it does not matter how many times a teacher explains a rule; if the students are not ready to learn it, the teaching is of no use (as referred to in

Tornberg, 2001: 64). Students must be mature enough, have understood a structure and have a need for it to be able to use it automatically (ibid.:106). Since there are many aspects of the teaching of a grammatical rule such as subject-verb agreement, I am interested in how ninth graders experience their learning of this rule, what learning strategies they prefer and if they can apply a certain rule in their writing. I am also interested in finding out if they have the ability to reflect on their own written production and how they think when doing so.

1.1 Problem

According to Behaviourism, language development is about the “formation of habits” (Lightbown & Spada, 2002: 34). Behaviourists therefore assumed that the language habits established in the first language would transfer to or interfere with any second or third language. In this line of thinking – referred to as the Contrastive Analysis Hypothesis – similarities between two languages would facilitate the learning of a new language while differences would make learning harder. Chomsky and other scholars challenged this notion and pointed out that many errors produced by learners could not be attributed to the L1. As other researchers followed suit, it was shown that language learning is developmental by nature and that there are many features that are the same in the acquisition of an L1 and the learning of an L2. So called error analysis was a consequence of this more nuanced way of looking at language development: errors can be explained in different ways, some errors are developmental while as some may be linked to the L1. Pronunciation and syntax (word order) are obvious examples of the latter. Consequently, transfer is still a possible explanation why some errors occur, and according to Ulrika Tornberg, positive transfer can facilitate the learning of a grammatical rule in the target language if the students recognize it from their L1 (2001: 65 and 61). Similarly, research differences between the L1 and the target language may in some cases lead to negative transfer or interference. These insights resulted in the first of my research questions:

(9)

- To what extent do some Swedish students think it is difficult to learn a grammatical pattern such as third person s?

This question needs an explanation. There is naturally a major difference between learning and understanding a grammatical structure explicitly in the classroom and using it in

spontaneous speech or writing. Embedded in the question is therefore a curiosity in finding if the students I set out to investigate had the same experience. I wondered if they consciously make use of the grammar they learn explicitly in the classroom. These thoughts resulted in the second research question:

- To what extent do the students think about grammatical rules when using the language in open-ended forms of communication such as in writing?

Embedded in this question is whether they reflect on rules that they have studied and practiced in language exercises when they e.g. write a paper. Can they at all connect their language use to the rules that they have studied explicitly? If so, they would stand a good chance of being able to correct their errors. This thought resulted in my third research question:

- To what extent do the students go back in their written texts to try to correct them?

Another interesting aspect concerns how a grammatical rule is taught in the classroom, and how the students experience the teaching. It is interesting to find out if the students are aware of the existence of different learning styles and to find out if they have any preferences. These thoughts resulted in my final research question:

- How do the students prefer to study grammar, deductively or inductively?

As I discussed in the introduction, learning subject-verb agreement is also a question of the individual students’ language proficiency. I was therefore also interested in finding if or how the ninth graders that I investigated were able to use third person s correctly. As Tornberg has explained (2001:105), grammatical rules are learned in predictable sequences, independent of the learner’s age.

(10)

1.2 Aim

I aim to find out if the ninth graders experience that the lack of interference from Swedish in the case of subject-verb agreement is something that has a negative effect on their learning of this grammatical rule. I am interested in how ninth graders experience declarative and

procedural knowledge in the classroom and outside school and in finding out if they can use their explicit knowledge spontaneously, with the only focus on the subject-verb agreement, in their written production. I also aim to find out if the students have the ability to analyse and correct their answers in their own texts and if they do this spontaneously, if they have their own learning strategies and if they prefer the inductive or the deductive method. The purpose of this investigation is thus to give me as a future teacher a deeper understanding of the students in their learning situation which also can help me to develop strategies on how to approach grammar in my teaching.

(11)

2 Background

It is important to understand the difference between declarative and procedural knowledge. Grammar that is taught explicitly and that leads to declarative knowledge, does not

necessarily lead to students’ proficiency in their spontaneous production, but it can be very helpful on the way of acquiring procedural knowledge.

2.1 Declarative and procedural knowledge

Declarative knowledge is the conscious knowledge about the language, about its formal structure, that teachers mostly teach the students explicitly. Procedural knowledge is unconscious and intuitive knowledge of how to use the language, which students learn implicitly. In short, you can say that declarative knowledge is the learners’ knowledge about the language and the procedural knowledge describes what to do with the language (Tornberg, 2001:62-63). Larsen-Freeman refers to declarative knowledge and procedural knowledge as formal grammars and functional grammars (Carter & Nunan, 2005:34), the latter concepts clarifying the meaning of the two former. Declarative knowledge is also closely related to learning and procedural knowledge to acquisition of a language, which will be presented further in chapter 2.4.

Ellis explains that declarative knowledge involves ‘knowing that’. It consists of such

information as the definitions of words, facts, rules and memory for images and sequences of events. Procedural knowledge is ‘knowing how’. It is represented in memory in terms of ‘production systems” (1990:177). Some researchers “distinguish the mental operations that a person is able to perform effortlessly and automatically”, here describing procedural

knowledge, “from those that can only be performed with considerable difficulty and relatively slowly” (Ellis, 1990:176), describing those students whose formal knowledge still is not automatized. Anderson describes three stages in the learning process:

(12)

 The cognitive stage, where the learner acquires declarative knowledge

 The associative stage, where the knowledge is proceduralized by the detection and correction of errors.

 The autonomous stage, where the learner’s errors disappear and performance is more or less unconscious.

(1990:177).

A consequence of this model is that explicit teaching leads to procedural knowledge. Ellis adds further explanation of these stages: “it is possible for a learner to pay attention without doing so consciously” and controlled processing can “take place with or without any

conscious awareness” (1990:177). As a contrast, Tornberg presents Krashen’s Acquisition-Learning Hypothesis, which claims that learning explicitly about the structure of a language can never lead to procedural knowledge. This she later criticises by saying that this hypothesis has been shown to be untenable and that a deeper understanding of the language structure can result in acquisition, though with a certain delayed effect. She claims that these thoughts have resulted in the teaching of today, which should achieve a balance between function and form and between explicit and implicit teaching (2001:104-105).

2.2 Methods for learning

Students’ success in a school environment mainly characterized by implicit learning is “often measured in terms of their ability to ‘get things done’ in the second language, rather on their accuracy in using certain grammatical features” (Lightbown, Spada, 2002:92). The Grammar- Translation Method is the oldest pedagogical method, used since the nineteenth century, and it is the method that has the most obvious focus on grammar as a product. The focus is on analysing grammatical structures of the target language and the teaching is mostly held in the students’ first language except from isolated phrases that are translated. There is no focus on the use of the language and the teaching is mostly about the language (Tornberg, 2001:103). Tornberg explains that it is important when teaching to bear in mind the distinction between grammar as a process and grammar as a product, since it helps us to isolate the grammar as form, but also to integrate it in the use of the target language (2001:102).

(13)

It is preferable to use a more communicative approach in order to facilitate the achieving of procedural knowledge for learners of a second language. “Communicative, content-based and task-based instructional environments also involve learners whose goal is learning the

language itself, but the style of instruction places the emphasis on interaction, conversation, and language use” (Lightbown, Spada, 2002:92). Communicative and task-based methods are often used in today’s classroom, but the difficulty is to find topics that engage all students and raise their motivation to the point where they learn implicitly. One of these methods, the Direct Method is focused on the use and the process of the target language. The teachers never teach the students grammar deductively and the students are advised to detect the structures of the language on their own.

Pienemann claims that looking upon the students’ production of spontaneous writing shows the students’ interlanguage, which represents their learning development of the target language. This can be the teacher’s guide to understand where in the language development the students are and when they are ready to learn a rule. This is what Pienemann calls the teachability hypothesis, which also shows that there is no use of teaching the rule if the students are not ready (Tornberg, 2001:61 and 64).

Rutherford asks “if language knowledge develops primarily in terms of accumulated structural entities, then what kind of learner production would we expect to see along the way?” (1999:5). Can teachers expect that the student will add this rule “to his ‘repertoire’ of already existing structures” (5) and that he or she will master those structures, one after another, as he learns them?

2.3 Learning, acquisition and the Input Hypothesis

According to Krashen there is a difference between learning a language and acquiring a language, learning being the conscious learning about the language. Explicit learning which leads to formal knowledge is either introduced by the deductive or inductive method and by having errors corrected. Krashen describes acquisition as the subconscious way of developing ability in the language (Alatis,1981:2) and he is convinced that learning cannot lead to

(14)

input is enough to acquire a language structure (Alatis, 1981:3). Robinson, on the other hand, claims that learning without awareness is not possible and that implicit unanalysed knowledge is eventually transformed into analysed knowledge (11).

Krashen discusses the Input Hypothesis as important aspects of acquiring a language (Alatis,1981:3-4). The traditional way to teach a grammatical structure is to first focus on form, and then later to be instructed to try to use it, but the Input Hypothesis advises us to expose the student to “structures a bit beyond his or her current level of competence”

(1983:4). The input should be comprehensible, interesting, not grammatically sequenced and in a sufficient quantity and the focus should be on meaning. Krashen claims that the input must be designed to enable the students to understand the message to be able to acquire the grammatical structures and when these structures reoccur, the students will get a review provided in a natural way. This indicates that constant exposure to a variety of structures is important to satisfy the need of all students and that “it requires more than a few exercises and more than a brief paragraph of input for the acquirer to fully acquire a new structure”

(1980:5). On the other hand, Lightbown and Spada “do not find support for the argument that if second language learners are simply exposed to comprehensible input, language acquisition will not take care of itself” (2002:150).

2.4 The Monitor Hypothesis

The Monitor Hypothesis shows that the fluency performance a learner has in a second

language is due to the acquisition and not to the learning of the language (Alatis, 1983:2-3). In this hypothesis Krashen states that “the acquired system acts to initiate the speaker’s

utterances and is responsible for fluency and intuitive judgements about correctness” (Lightbown and Spada, 2002:38). This indicates that it is procedural, not declarative knowledge, that can result in increased accuracy in students’ spontaneous production.

Tornberg says that we use our declarative knowledge as a key to correct our production, while the procedural knowledge represents the already automatized and subconscious knowledge and she recommends to have the focus in teaching on interplay between the development of the two kinds of knowledge (2001:62, 64). Krashen adds that learners have some use of the

(15)

traditional way of learning grammar, which helps them to produce a more formal accuracy in the already written text, using the Monitor or editor.

2.5 The Natural Order Hypothesis and the Affective filter

Krashen claims that you learn grammatical rules in a certain predictable order, which he calls the Natural Order Hypothesis, and where the third person singular marker –s is among the last grammatical rules you learn. The Natural Order Hypothesis is predictable in a person’s first language, but shows similarities in the learning of second language, and is therefore a hypothesis to trust and to have in mind when teaching English to Swedish learners (Alatis, 1983:2). Lightbown and Spada claim that studies have shown that the rules that are easiest to learn explicitly are not necessarily the first rules to acquire; “the rule for adding an –s to third person singular verbs in the present tense is easy to state, but even some advanced second language speakers fail to apply it in rapid conversation” (2002:39).

As the predictability of the natural order hypothesis is similar to first and second language learners, also many developmental errors are similar. However, some errors, e.g. “the morpheme accuracy/acquisition orders which emerge from second language studies differ somewhat from those established by first language studies” (Singleton, 1989:124).

Singleton states that in first language learning, the most efficient learners are older children (1989:78), but concerning second language acquisition, research shows no evidence of this. However, she claims that the younger the learner is when beginning to learn a second language, the better, but it is difficult to compare different school settings “since second language exposure-time involved in studies focusing on formal learning situations never approaches that involved in long-term naturalistic settings” (1989:137). She adds that there is no research that contradicts the hypothesis that those who begin to learn a second language early in life “generally achieve higher levels of proficiency than those who begin later in life” (1989:137).

Rutherford explains that the differences of language structure between L1 and L2 in second language learning make the student “constantly confront the need to express familiar or

(16)

routine thoughts by means of an initially alien code and, in turn, to interpret and understand that same code used to express the (unpredictable) thoughts of the person with whom he is trying to communicate” (1999:7). Tornberg also claims, that the learning of an unfamiliar grammatical structure makes the students struggle hard: “Dessa elever måste således kämpa sig till en insikt om en språklig kategori i engelskan som känns främmande och onaturlig i deras eget språk” (2001:65). Selinker claims that interference, e.g. the phenomenon where the learner transfers the structures of his first language to his second language, has a more

positive than negative effect on the learner than other researchers have believed. The learner is more likely to transfer familiar structures than unfamiliar ones, which facilitates the learning (Tornberg, 2001:61).

Avoidance, or what Lightbown and Spada call simplification, is a developmental error that shows that the student is on her way to understand the structure of the language (2002:75). They also claim that if the learners do not recognize the rule they “may learn a second language rule but restrict its application” (85) or prefer not to use it (86). Motivation and success are important issues in the learning of a language. The students with a higher motivation are often more successful and success in itself generates motivation (Lightbown and Spada, 2002:52). According to Krashen, the Affective Filter, anger, anxiousness,

boredom, can block the students’ motivation and thereby their acquisition. When the filter is up, the students’ motivation is low and when it is down the students are highly motivated 2002:39-40).

2.6 The role of procedural and declarative knowledge: a summary

As mentioned before, Krashen’s way of juxtaposing acquisition and learning has been severely criticised. Findings from immersion projects in Canada have shown that a focus on function only will not lead to any full control of the grammatical system (see Lightbown & Spada, 2002). Krashen’s hypothesis about acquisition and learning was supported by many scientists in the eighties, and even the Swedish syllabi instructed teachers to not focus on grammar in their teaching, since it, according to Krashen, had no effect on the student’s language development. Teachers were instead advised to have a focus on communication in their teaching. Over time, the acquisition and learning method, which cannot be tested

(17)

through science has been criticised (Tornberg 2005:63,105). Ellis explains that when it comes to older children and adults, and above all learners on a higher level, an explicit knowledge about grammatical structures is necessary in order to gain implicit knowledge (Tornberg, 2005:31). This means that the student’s knowledge about the grammatical structures (declarative knowledge) of a language leads to the ability to use these structures in production, procedural knowledge.

(18)

3 Method

I conducted my investigation by first giving the students in two classes of ninth graders a writing assignment, which instructed them to write at least half a page describing the day of a mental patient at a mental hospital. It was important that they wrote spontaneously and in the present tense, since my aim was to investigate only the present tense in student’s writing. To ensure this, I gave them specific instructions of acting as a person who studies and takes notes of what the patient is doing at the moment, and not what he did before. I told them that my investigation was about how they use the English they have learned, but I did not tell them that the writing assignment had anything specifically to do with verbs and subject-verb agreement. Then they would have had too place much focus on grammar, which I wanted to avoid. The purpose was to get a hold of their spontaneous production, which challenged me to create a setting for their story that was interesting enough to make them think of what they were writing and not on how they were writing it. I exhorted them to write down as many interesting things about this inventive patient as they could think of in order to take their minds of the form of the language. I also told them that this was not a test and that they should use the English they had learned and try not to ask questions or use the dictionary during the writing process.

After the students had handed in their texts I read through some of them and specifically looked for texts that were interesting for my interviews. I wanted texts which had enough verbs written in the present tense, which had verbs that agreed with a variety of subjects, and texts that had both correct and incorrect forms of the verb. I asked four students, two from each class, two boys and two girls if they wanted to take part in an interview. I handed out a note for them to bring home in order to get their parents’ consent. After about a week I conducted the interviews in a small classroom with only one window facing an atrium where no other students were allowed. I started by asking if the student had another native language than English, which is relevant for this investigation. This was also the main reason why I chose this particular school, where most students have Swedish as their native language. I discussed about eight of the verbs in the student’s text and exhorted the student to give me his or her explanations of the choice of verb form. I hoped they would provide me with enough

(19)

information on how they think when writing spontaneously. The next step was to ask them to correct their choice of verb form by asking them if they wanted to change anything. In the last part of the interview I and the interview subject had a discussion on how to learn a

grammatical rule such as the subject-verb agreement.

This method of investigation, first to make the students write a spontaneous text and then advise them to reflect on what they have written has been inspired by the method called “introspection” (Malmberg and others, 2002:9), or “the think aloud method” (2002:9), described in the project STRIMS’ investigation I huvudet på en elev. The way this method was mostly used in this investigation was by monitoring students when they faced

grammatical problems in sentences. They were instructed to think aloud when trying to figure out the right grammatical form and when trying to find out what was wrong and what was right in the sentences. The purpose of using this method was to get access to the student’s thought processes in their writing production (2002:9). The introspection method was tried, but without success, already in the nineteenth century with the German psychologist Wilhelm Wundt as the most well known advocate. Nowadays introspection is used as a qualitative model of research (2002:9). The method of retrospection, also used in the STRIMS’

investigation, is described as a means of making the students go back to and reflect on their own earlier written production, something that takes a very central part of my investigation.

(20)

4 Results

The difficulty in presenting the result of the written texts was the great variation in the amount of examples of verbs in the present time agreeing with the third person singular. One pupil had no examples and three pupils had one to three, with one or two errors, while the rest of the students had five to 32 examples. In all the texts I found four cases of avoidance and one case of overuse, a student who attached all verbs with the marker –s, regardless of their agreeing subjects. I could also recognise in some student texts the marker –s on the verb agreeing with first person singular, and also those agreeing with the third person plural, “nurses”. The use of the progressive –ing form was also a problem, since most students used it without the preceding “is” or “are” and it seemed that many students use it to simplify their construction of the verb phrase and to avoid the deciding of which verb form to use for each subject.

4.1 Students’ difficulties in learning a new grammatical pattern

To what extent do some Swedish students think it is difficult to learn a grammatical pattern such as third person s? The ninth graders that I interviewed seemed to be focused on and have the most experience of explicit learning. They all had similar opinions on how a person learns grammar and how one learns how to use it. All four students conveyed a clear message: that you learn grammar by being taught the rules deductively and by doing exercises where you practice the rules and by repetition. They also had a very similar way of personally referring to the use of grammatical structures in their own self-produced texts, listening inwards and intuitively knowing what sounded right and wrong. None of them seem to think it’s difficult to learn an unfamiliar grammatical pattern, you just need to study it and use it.

The student in interview 1 (see appendices) thinks that to know a grammatical rule is to know how to use it in various situations (appendices). He lets us know that it is very easy for him to learn grammar and that he just reads through the explanation of the rule and then he knows it. This is how he claims that he learns a grammatical rule, but on the other hand he admits that

(21)

he normally does not study the rules, which does not stop him from being confident in his ability to see what is wrong and right in his own production; “Det kommer liksom i huvudet … jag pluggar inte in regler och så för at t… jag känner liksom vad som är rätt, liksom … man ser när det är fel vad man skriver och sånt” (47). He thinks he can master grammar well this way.

The student in interview 2 is also very confident in himself when using grammar even if he doesn’t know the rules, and he thinks that he can hear very well when his English doesn’t sound correct. He seems to think his experience in using English outside the classroom is relevant for his learning. In his experience looking at films and listening to music can facilitate the learning of a foreign language (51).

The girl in interview 3 thinks it is most important to know how to use the grammatical rule, especially in writing where the demand for a more formal language mostly is required (54). However, she has the opinion that procedural knowledge regarding grammatical rules isn’t very important in oral communication if only one can make oneself understood. She mostly talks about the explicit method as a mean to learn grammar and she agrees with the boy in interview 1 that the easiest way to learn a grammatical rule is to learn it, practice it and repeat it, all in the traditional way. The difficulty with learning rules is that you can mix them up and get confused, which is also why she finds it hard to learn them.

The girl in interview 4 says that she is insecure of the grammatical rules, and she claims that the reason is that she probably missed out on most of the grammar teaching in eighth grade since she was in Scotland (56-57). As the other interview subjects she relies a lot on how it sounds when making her choice of verb form, without referring to a specific rule. She is also the only one of the four students who avoids completely the use of this rule; none of the verbs in her text has the morpheme –s. To her, making errors and have them corrected is an

important aspect of learning, and writing short texts to practice rules that she experiences as a useful method when learning grammar (58). The importance of writing short texts instead of practicing grammar in cloze sentences is that the texts provide with a more meaningful content. As the girl in interview 3, she also thinks the difficulty with learning grammatical rules is that you often mix them up (57). Her opinion differs from the other girl concerning the effect of grammar on meaning; she believes that if you know how to apply grammatical

(22)

4.2 Students’ reflections on grammar in spontaneous writing

To what extent do the students think about grammatical rules when using the language in open-ended forms of communication such as writing?

The Krashen Monitor Hypothesis indicates that acquisition, not learning, is what can result in accuracy of students’ spontaneous production, such as the written texts of the students that I have investigated, and it also shows us that the students can correct their own language using the monitor as a mean to create a fluency in their writing, without referring to their declarative knowledge. This agrees with the result of the interviews that I have had, which revealed that none of the students thought of any grammatical rules while writing. I asked them for the reasons to their choice of verb form and we talked about each word twice, but none of the students referred to a grammatical rule at any time. Instead they explained that they listened inwards to their inner voice in order to get the verb forms correct. The student in interview 3 showed the best example of this method, mumbling to herself frequently during the interview, trying to hear what sounded right.

The boy in interview 1 begins by explaining that he does not know the reason to why he chose the verb make in the sentence “he makes wierd apelike sounds” (45), other than that it sounds more correct, but then when I ask him if it would sound just as correct without the -s, he doesn’t know. He does not know either if it sounds better with “try” instead of “tries” in the sentence “the nurses tries” (45), and in “nurses seem” (45), he does not know why he chose the verb without the –s. In the sentence “he pretends he is an airplane” he thinks that he chose “is” because it is only one airplane and even if the sentence is correct, he does not find a connection between the verb and the correct third person subject. The second time I ask him about the sentences, he seems more certain that it should be “try” instead of “tries” and he still thinks that “the nurses seem” sounds better than “the nurses seems” (46). I ask him how he thinks when writing a text, i.e. an essay, if he thinks of grammatical rules when writing. He answers that sometimes he might think of rules, but mostly he just writes down what feels right. He is aware of his own capacity to get the verb forms correct, “alltså det är ju bara att skriva på och ibland så kan jag ju tänka efter, vilket blir bra” (47).

(23)

The boy in interview 2 corrects one of his errors “he throw” into “ he throws”, giving the reason “för annars blir det fel, låter fel… he throws something at me” (48), here referring to what sounds correct, repeating the phrase to himself. He understands that if there are two nurses he has to use “are” and not “is”, but at the same time he is talking about prepositions, which has nothing to do with this. He thinks it sounds correct with “Peter’s eyes becomes dark” (438) the first time I ask him about the phrase, but the second time he thinks it sounds better with “Peter’s eyes become dark” (49), only because it does not sound correct with the additional -s.

The girl in interview 3 mumbles frequently during the interview when I ask her to explain the choices of verb forms and if she wants to change them. She immediately corrects herself in the sentence “he start”, which she changes into “he starts” and in the same sentence she also corrects “run” into “runs” so that she get the phrase “he starts and runs” (51). She has made the same error as the boy in interview 2, “he got eyes that reminds me” (52) and instinctively she corrects “reminds” into “remind”, but when I remind her that I haven’t marked the words because they are wrong, she gets uncertain again. Later during the interview when we return to the word she is certain that it should be without the –s, with the explanation that it has to do with “me”, but then again she feels uncertain and changes it back to “reminds” (53). After I have informed her of the third person –s rule and she has applied the rule to some of the verbs in her text, she suddenly says “men jag är inte så säker på det längre. Jag tycker det låter bättre med start” (53) going back once again to the “start” she has discussed earlier. It is uncertain here if she means she no longer believes in the existence of the third person –s rule or if she only refers to the verb alone.

At one point she shows that she is focused on meaning and not form. To the question how she thought when she wrote “one of the doctors that works” she answers “dom arbetar, dom liksom arbetar, works” (52). She says she doesn’t think about grammatical rules when she is writing, they mostly make her confused and she mixes them up. She writes what she feels is the best to write, what sounds natural and what sounds correct. Sometimes she goes back to check on her writing, “ibland när jag är osäker går jag tillbaka och tittar på vad det skulle vara. Men det är ingen regel som jag brukar tänka på. Jag blandar ihop allihopa, så att…” (53). When correcting her text during the interview, at first she seems to have a clear idea of

(24)

what is wrong and right, starting with correcting the sentence “he start and run” into “he starts and runs”, but later on she shows an increasing insecurity in her instincts.

The girl in interview 4 also struggles with the verbs “start” and “run”. She wants to change “he start “ into “he started” and to the question how she was thinking when she wrote “I run” and she, as the other girl, refers to meaning and not form, “jag tänkte väl att jag sprang” (55). Repeatedly she wishes to change the verb form into the past tense, even with a –t as the inflectional morpheme (56) and even if we mostly talk about verbs in present tense. She also wants to change the verb itself, “go” into “walk” (56) even if I try to explain when the –s should be there and not. This girl is very insecure about grammar and spelling and she also does not trust herself when listening to what sounds right and wrong (57).

4.3 Student’s inclination to analyse grammar in self-written texts

To what extent do the students go back in their written texts to try to correct them?

The student in interview 2 only follows his instincts when writing a text, and sometimes he goes back to check on his writing, then only listening to what sounds right and wrong.

During the interview the girl in interview 3 mumbles a lot to herself to listen inwards to what sounds right and wrong concerning choices of grammatical form. This is also her explanation of how she works with her language during the writing process. On the question if she has the ability to mostly get it right when listening to her inner voice, she explains “inte alltid. Jag är inte säker men jag kan göra mig förstådd. Man förstår vad jag menar men det är inte alltid rätt verbform” (54). She understands that listening inwards to the inner voice to get the

grammatical forms correct is easiest for a person who listens a lot to English (54), which shows that she has some insights of how you achieve procedural knowledge. She does not have as much experience of English outside the classroom as the boy in interview 2, but she seems rather confident in her abilities of communicating in English, telling us that she has travelled a lot and taught Ethiopians how to speak English. On the question what it means to know a grammatical rule, she answers, “det hjälper ju en om man kan det naturligt men man kan ju ändå göra sig förstådd… när man pratar med engelsmän och sånt, de förstår ju ändå

(25)

vad man menar. Även om man kanske inte så, böjer dem exakt rätt. Men om man ska skriva och så, så är det ju bra om man skriver rätt” (54). She repeatedly says she does not think about grammatical rules when writing and she, as well as the other students, does not remember the third person –s rule when I try to remind her of it.

The girl in interview 4 says she usually goes on writing and she tries not to get stuck when she feels insecure and sometimes she goes back later to correct her language (57). The most important issue to her when correcting her own text is that there should be a connection in meaning and structure between the items and she makes her choices based on how it sounds.

4.4 Preference and experience of learning methods among students

How do the students prefer to study grammar, deductively or inductively?

The student in interview 1 (see appendices) has the insight that learners need to practice the rule that they have learned explicitly, by using the rule in different ways; “prova sig fram, arbeta, försöka med, skriva och sånt… använda regeln” (47). He does not remember having tried the more implicit way of learning a grammatical rule, i.e. using a text as a starting point to identify the rule on his own by trying to discern its pattern, and he seems to be satisfied with the explicit way of learning even if he does not remember the rules (47). He does not seem to understand that the English he hears, writes or reads perhaps has an effect on his learning of grammar. He only refers to the practicing grammar learned explicitly.

The student in interview 2 thinks that it would be too hard to try to deduce from a text a grammatical pattern, and he thinks that the teacher should explain the rule and hand out exercises with cloze sentences (appendices p.50). Here he informs us of a good method to learn a grammatical rule: “… alltså stenciler med meningar, för då hade man lärt sig sätta in den i meningar, de olika verben … ja, då ska du ju kunna det. Sen hade man kunnat få regeln också att kunna sätta in i pärm, eller så, engelskpärmen. Då hade man nog kommit ihåg den” (51). This student seems to be aware of the distinction between learning a rule and knowing how to use it, that you can learn it explicitly and by repeating it eventually learn how to use it, but on the other hand he explains that he has learned how to use grammatical rules by

(26)

watching English programs on TV (51). He seems to think his experience in using English outside the classroom is relevant for his learning; “alltså i dagens samhälle så är det ju mycket engelska och det är bara att titta mycket på filmer… så lär man sig” (50). He does not know the rules, but he claims that he can hear very well when his English does not sound right and he is very confident in himself.

The girl in interview 3 prefers the explicit method as a mean to learn grammar and that the easiest way to learn a grammatical rule is to learn it, practice it and repeat it. On the question what she thinks about learning a rule applying the inductive method, she imagines it would be useful in the sense that students have to think more for themselves, but she does not remember having experienced this method herself.

The student in interview 4 is the only one among the interview subjects who can remember having an experience of the learning of a rule inductively, an experience from her four months spent in Scotland. In groups the students picked out the verbs in past tense and formulated the rule, which seemed to be a method frequently used in their learning of grammar.

Unfortunately she was not able to join in the exercise because it was her first week in the school and she had a difficult time. She thought it a difficult exercise and did not think it was useful at the time, but during the interview this is her suggestion for a good exercise (58).

4.5 Results from the interviews of the third person –s rule

Examples of third person singular -s and the errors

0 5 10 15 20 25 30 35 1 3 5 7 9 11 13 15 17 19 21 23 25 27

Examples of third person singular -s Errors

(27)

Krashen states that the third person –s is an easy rule to learn (as referred to by Robinson, 1996:38), and one of the students I have interviewed seems to agree with him; “…engelskan är ändå rätt lätt för det är ju bara ett –s på slutet…” (boy in interview 2, p.50). None of the students I interviewed recognized or could explain the third person –s rule even if I tried to refresh their memories, but the two boys had a very good idea of when to use it and they had confidence in their ability. They all had minor problems applying the –s to the verb agreeing with third person plural, but none of them applied the –s to any other subjects.

The boy in interview 1 had 22 examples of third person singular –s and all of them were correct even if the verb was not close in the sentence to its agreeing verb; “Then he makes wierd ape-like sounds and begins to swing in the lamp” (text 23, appendices, p. 59). In this sentence you could easily understand that “makes” is easy to get right, standing close to “he”, but “begins” must be harder to get right, standing on its own in the sentence further away from its subject. Even if he had all the inflectional morphemes correct to the third person, he had overused them and put them agreeing to the third person plural; “the nurses decides to” and “the nurses tries” (text 23, appendices, p. 59). The other boy also had an example of the marker –s attached to the verb agreeing to third person plural. He got “his look turns” correct, but “Peter’s eyes becomes” was incorrect, and he did not have an explanation for it (interview 2, p. 48). He had 18 examples of the third person –s, but had only one third person singular –s error in the first sentence; “he throw” (text 24).

None of the girls seemed to have a feeling for when it was appropriate to use the rule and they were both very insecure of when to add the third person –s to the verb. The girl in interview 3 made the same error as the boy in interview 2 and had as he put a third person singular –s agreeing with “eyes” (text 25). She had only 7 examples of verbs agreeing with third person singular, but 4 errors, two of them directly after the subject and two further away in the sentences. She had managed a more difficult sentence; “one of the doctors that works whith the patient comes back”, but did not manage the easier; “he start screaming and run” (text 24). After I had reminded her of how the rule works, she seemed to understand that you only put an –s on the verb agreeing to the third person singular, and immediately after that she wanted to take away the –s from “reminds” in “he got eyes that reminds me”(53). When she

explained the reason for this, she thought that the verb should agree with the subject standing closer to it; first person singular, instead of the third person plural. The verb form was correct,

(28)

The girl in interview 4 also had 7 examples of verbs agreeing with third person singular, but she also had a complete exclusion of the inflective morpheme –s, which made her a case of avoidance. After I had reminded her of the rule and asked her to explain it she was able to do that, and she understood that her error “it do not” should be “it does not”, but the adding of “not” complicated things when she thought that “it does” is correct, but not “it doesn’t”. During the interview she pointed to “do not” and explained that she first had written “doesn’t”.

4.6 Results of written texts, third person –s

The results of the written texts showed great variation in the amount of examples of verbs in present time agreeing with the third person singular. Some students had a difficulty sticking to the present tense and had a majority of the verbs in past tense. Some had written almost two pages handwritten text and some only a little more than half a page. The five pupils who had no examples or up to three examples of verbs in present tense agreeing with third person singular had one or two errors, while the rest of the students had five to 32 examples. Of the three students who had five examples of those verbs, two were cases of complete avoidance, as the girl in interview 4 (text 26). All of them, including text 26 with its seven verbs in present time and its seven errors, consisted of just a little more than half a page of text. The fourth and last case of avoidance was a text with eight verbs in present tense and eight errors, all verbs agreeing to the third person singular standing next to the subject.

Most of the students had five to 14 examples of verbs in present tense agreeing with third person singular and all had one to ten errors, except four students who had no errors. One student, who had no errors on the verbs agreeing with third person singular, had a sentence in which he suddenly had the marker –s on a verb agreeing with first person singular. This type of error is seen in some of the other texts as well. Another student had put the third person singular –s on a verb agreeing with third person plural, as in some other texts, presented below.

(29)

The students who had the most verbs agreeing with third person singular, 18-32, had few errors in comparison to the larger amount of text that they presented. The two students who both had 18 examples of those verbs also had only one error each, but one of the students had overused the marker –s and put it on all the verbs; “the nurses tries… the nurses takes… and gets” (text 17). The student text that had 22 correct examples of the verbs agreeing with third person singular also had two examples of verbs agreeing with a plural subject.

4.7 Results of written texts concerning other examples of subject-verb

agreement

The progressive, the –ing form was, along with the verbs in past tense, a reason for the lack of verbs in present tense and it also simplified the students’ construction of the verb phrase and helped them avoid the deciding of which verb form to use for each subject. About all of the ten students who had examples of are/is and have/has in their texts had used them all correctly. The only student who had two examples of incorrect use of “are” had agreed those to third person singular; “what the hell are he doing?” and “everybody are quiet” (text 14, appendices, p. 40), and the only incorrect use of “have” also agreed to the third person singular; “he begins to eait, like he newer have seen foud befour” (text 12). All the rest of the 11 students who had used have or has, had used them correctly. The students who had little or none examples of verbs in present tense, had either mixed the present and past tense in every other sentence or begun writing in present tense, later to change into the past. The use of “was” and “were” were more or less frequently used, but they were all used correctly.

(30)

5 Discussion and Analysis

Many researchers have discussed declarative knowledge and procedural knowledge and most of them seem to agree on the definition of them, originally identified by Anderson, declarative knowledge referring to knowledge that a learner has about the target language and procedural knowledge being the knowledge a learner has of how to produce language effortlessly. The relationship between declarative knowledge, explicit learning and deductive teaching is easy to make sense of: the teacher explains the grammatical rule, the students practice it in short sentences, which eventually may lead to knowledge about the language structure. This is the method used in the Grammar-Translation Method, which Larsen-Freeman refers to as Formal Grammars (Carter, Nunan, 2005:34) and which is no longer modern, but still often used by teachers. In the 1970s, researchers observed that this pedagogical method did not help learners to communicative contexts outside the classroom, which led to an interest in communicative approaches (37).

One of the main purposes of this investigation was to find out if the ninth graders that I have investigated have reached a level in their interlanguage where they are susceptible to learn the third person –s rule. The Pienemann Teachability Hypothesis helps us find out if the students we teach are ready to learn a certain grammatical rule or not by analysing students’ texts to establish in which developmental stage they are. I think it is difficult to find that out from just looking at one student text, as is the case in this investigation, and I think you need to look at the students’ production repeatedly in both writing and speaking. This has to take place during a longer period of time to get a clear idea of their interlanguage. What we can deduct from studying the student texts from this investigation is that they all are in different stages in their development of the third person –s. This means that I, as a teacher, have to pay attention to my students’ developmental stages and to when it is suitable to discuss with them a certain grammatical rule.

The question Rutherford asks “if language knowledge develops primarily in terms of

accumulated structural entities, then what kind of learner production would we expect to see along the way?” (1999:5) is a question that we as teachers should ask ourselves once in a

(31)

while. It is important to know why we chose certain methods, what we want to achieve with them and what we expect from the students. So if you teach the third person –s in the traditional way, then how and when do you want the students to produce the rule? Do you expect that the students will add this rule to their “repertoire’ of already existing structures” (1999:5) and that they will master those structures, one after another, as they learn them? The result of this investigation shows us that a teaching only on form is not enough for learners to learn the structures of a language; therefore, the teaching also needs to include exercises in conversation with focus on meaning.

None of the students contributing to this investigation could spontaneously refer to a grammatical rule while reflecting on their choice of verb form in their texts, and they constantly referred to if it sounded correctly or incorrectly. All of them had a difficulty to connect their written text to the knowledge of a grammatical structure. None of the four students I have interviewed remembered the third person –s rule even if I reminded them of it, which indicates that deductive teaching of grammatical rules is not sufficient for learners to achieve procedural knowledge and that other learning methods, such as a more

communicative approach are essential. Many teachers present to their students a text that has a certain grammatical structure which he or she wishes to teach them by making them aware of it. However, this is not the right method according to Krashen, who claims that focus should be on meaning, not on form. Communicative and task-based methods are often used in today’s classroom, but the difficulty is to find topics that engage all students and raise their motivation to the point where they learn both vocabulary and grammatical structures implicitly. Tornberg recommends, however, a teaching with a focus on interplay between form and function (2001:62, 64), which I think is preferable. As a teacher you should bear in mind the importance of communication, which of course is the goal of language teaching.

It is sometimes difficult, but imperative, to motivate students into learning the language we are teaching and to find ways of lowering the Affective Filter, which can stand in the way of the students’ learning of the language. A good way is to get to know your students and to find out which their interests are. We also have to detect the students anxiety or boredom and help those students in order to raise their motivation and thereby their success. As we understood from the girl in interview 4, she told of anxiety when being a student in Scotland. She said

(32)

To make our students successful learners we must raise the motivation and deal with the issues such as anxiety. Many students have an aversion to speak in a whole class environment and they should be given the opportunity to speak in small groups.

Lightbown and Spada claim that the rules that are easiest to learn explicitly are not

necessarily the first rules to acquire and as an example they mention the third person –s rule. This is yet another exhortation to us as teachers that teaching according to a communicative approach with the focus on meaning and not on form is imperative.

Rutherford explains that the student who has to learn a unfamiliar grammatical pattern has to “constantly confront the need to express familiar or routine thoughts by means of an initially alien code” (1999:7), which means that it is harder to understand a familiar pattern and thus to learn it. Tornberg seems to agree with him, that the learning of an unfamiliar grammatical structure makes the students struggle hard; “Dessa elever måste således kämpa sig till en insikt om en språklig kategori i engelskan som känns främmande och onaturlig i deras eget språk” (2001:65). She also informs us that transfer has a positive effect on students in the sense of the transferring of familiar grammatical patterns that facilitates learning. In Swedish we do not have the structure of agreeing the verb form to the subject, and therefore it is harder to acquire this pattern than i.e. the use of determiners.

The student in interview 1 is very confident in his use of grammar, but this confidence is not due to the learning of grammatical rules explicitly, since he admits that he does not study grammatical rules actively in the way that he actually thinks you learn grammar. He thinks it is useful to practice grammatical rules by using them. He relies on his instinct of what sounds right and wrong, which seems to help him make the right choices even if he does not know how to explain it. However, he does not agree the verb with the correct subject in the sentence “he pretends he is an airplane”. Here he thinks that the verb “is” agrees with “airplane” and not with “he”, but he still gets the sentence right. He sometimes reflects on his writing during the process and, as he expresses it himself; the result is good (42).

The boy in interview 2 is very interesting because he has a different opinion on how you learn grammar, one that to some extent confirm those of Krashen. He seems to believe in the Input Method, telling us that there is a lot of English that you can learn from in the Swedish society today and his advice is to watch films with English speech, with or with subtitles in English.

(33)

He thinks that the English he has used in communication outside the classroom has helped him in understanding the structure of the language. But still he talks about learning grammar explicitly, but more in the less abstract way of learning it rather than using it in

communication. He states, as the student in interview 1, that he relies on his procedural knowledge to get the grammar right in production.

Singleton states that in first language learning, the most efficient learners are the older

children (1989:78), but concerning second language acquisition, research have no evidence of if the most efficient learners are the younger or the older children. She adds that some

evidence shows that the younger the learner is when beginning to learn a second language, the better, but it is difficult to compare the different school settings “since second language exposure-time involved in studies focusing on formal learning situations never approaches that involved in long-term naturalistic settings” (1989:137). She also states that there is no research that contradicts the hypothesis that those who begin to learn a second language early in life “generally achieve higher levels of proficiency than those who begin later in life” (1989:137). The boy in interview 2 shows that Singleton has a point; he has been in contact with English in a naturalistic setting since he was four and he can master grammar far better than a lot of the other students of my investigation. He is one of the students who has 18 examples of verbs agreeing with third person singular, but only one error, which makes him one of the five students who have the best result.

The difficulty for second language learners is that they mostly do not have the access to a naturalistic setting for their learning. Lightbown and Spada emphasize the importance of explicit teaching in this classroom environment when the teaching approach is on

communication with a focus on meaning. When the teacher “attempts to simulate ‘natural’ communication in conversational interaction, the students focus naturally on what they say, not how to say it. This can result in a situation where learners provide each other with input which is often incorrect and incomplete” (2002:150).

Therefore, if we do not have the access to a naturalistic setting for second language learning, we need to keep a certain focus on form in our teaching. “There is increasing evidence that learners continue to have difficulty with basic structures of the language in programs which offers no form-focused instruction ”(Lightbown and Spada, 2002:150).

(34)

The student in interview 3 has some experience in using English outside the classroom from the time she lived in Africa and from travelling to other countries and she seems to have a clear idea of that the most important issue when learning a language is that you make yourself understood and that grammar is something less important in communication if it does not help in conveying a clear message. She seems confident in her abilities of communicating in English, but less confident in herself when it comes to grammatical rules. She clearly listens inwards to her inner voice when I ask for her reflections on the grammatical choices she has made in her text. This was the student which most obviously demonstrated the techniques of the Think Aloud Method during the interview. She was mumbling, thinking aloud, even if we could not always hear what she was saying. She argues that it helps if you know how to use a grammatical rule, but it is not necessary if the listener understands what you are saying. This student seems to think that grammatical rules are in the way of the fluency in production, since she mixes them up and they make her confused. As the other students, she works her way through the language structure when writing a text using her instinct, not thinking of the grammatical rules, which indicates that procedural knowledge is not always a result of deductive teaching.

The girl in interview 4 is more insecure in her intuitive abilities in constructing a text with accuracy in grammar. As the other girl she refers spontaneously to meaning and not form and she seems to think she can avoid making choices of verb form by changing the verb itself. When I ask her if she wants to add an inflectional morpheme to the verb, she instinctively wants to add –ed and even –t to the verb. She avoids completely the third person –s in her text, which is seen as a developmental error, and which shows that she is in a developmental stage in her learning of this rule. She shows signs of insecurity, which can explain her avoidance and which also can stand in the way of her communicating in both writing and speaking. She is the interview subject that seems to have the most focus on meaning, which can give her a purpose for communicating, but I am convinced that she could benefit from explicitly being taught language structure to help her gain more confidence. Even though she has naturalistic setting experiences, they are not over a longer period of time and she also talks about anxiety which could have stood in her way of acquisition during her time in Scotland.

All of the interview subjects had minor problems with applying the –s to the verb agreeing with third person plural, but none of them has applied the –s to any other subjects. This could

(35)

indicate that spontaneously learners have a difficulty to discern how many people are

involved in the subject or they see i.e. “nurses” as a collective. Could the error that two of the interviewed students have in common be due to alliteration, only with the repetition of the final consonants instead of the initial; “his eyes becomes” (text 24) and the less frequent repetition “he got eyes that reminds me” (text 25)? Then this could explain why the girl has managed the sentence “one of the doctors that works”, but not “he start screaming and run” (text 25), where the repetition of the consonant is initially and not in the end of the word. We see the pattern again in “the nurses tries” (text 23, appendices, p. 59). The girl in interview 3 becomes confused when I explain the rule to her and she agrees the verb “becomes” to “me” and wants to take away the –s and the other girl also gets confused when she does not see that “doesn’t” is combined by “does not”. It seems that the more I talk to the girls about the rules, the more confused and uncertain they get. The instincts they relied on initially in our

conversation, they become gradually less secure of. This could indicate that the explicit teaching of grammatical rules can make already insecure students loose their confidence in themselves.

The students that I interviewed seem to believe that analysed knowledge, e.g. declarative knowledge, can turn into procedural knowledge, a belief that is not in line with how they correct and reflect on their choice of verb form in their texts, when they listen inwards to hear what sounds right and what does not sound right. Their method of constructing language in spontaneous writing is by using their interlanguage, which also is the goal in the teaching of a second language. One of the interviewed students states that he thinks he has learned grammar by watching English programs and films and he even experiences the third person –s rule as easy to learn. He describes one of those natural settings for learning that we can promote our students to use in daily life and that teachers can introduce in the classroom.

The interview subjects of this investigation, when asking for their reflections on grammatical choices and for how they learned grammar, two of the students gave answers that reflected on a very distinctive way of thinking. The former question revealed the students intuitive,

unconscious, procedural knowledge and it was a lot harder for the students to describe. The latter was easier to describe and revealed their knowledge on different methods on how to learn grammar explicitly, which I believe is built on their experiences and assumptions of my expectations. They were giving me what they believed was the correct answer to that

(36)

makes them make grammatical choices in their spontaneous writing, but their intuition. They show that they can reflect on their texts, both during the writing process and during the interviews a week later, but not doing so referring to grammatical rules. In the case of

correcting their errors; firstly, they did not know if they had the right or the wrong verb form, since they did not know the rule (which confirms the Krashen idea that declarative knowledge can be used by learners to correct a formal text), and secondly, if they thought something was wrong, they corrected the error in the same way as described above; intuitively. From their answers to my questions on their choices of verb forms and from the fact that their questions during their writing process were not on form, but on content, I concluded that they did not have their monitor on in the sense of correcting formal language, but only to get a more fluent language and intuitively a higher level of accuracy.

The assignment I gave the students was designed in a way that would encourage them to write without thinking on form and by that avoid that they had their monitor on. When the first group was writing I discovered that in spite of my attempts, some of them wrote their stories in past tense anyway, so I told them that they had to write as if things were happening right now. I found that it was very difficult for the students to stick to one tense and many of them mixed the tenses. Four students had almost all the verbs in past tense and their texts were very difficult to include in this investigation since its focus was on verbs in present tense. I was afraid that advising them to write in present tense would disturb them in their spontaneity, but the texts and the interviews showed that this was not the case. The fact that many of the students during the writing process forgot which tense they were supposed to write in indicated that they were focused on what they were writing rather than on how they were writing their story. Another thing that strengthened this understanding was that the questions they asked were related to the content and not to the form. This indicates that the students did not think of using their declarative knowledge, but that their aim was to get a fluent language and intuitively a higher level of accuracy.

With his Natural Order Hypothesis, Krashen suggests that you learn grammatical rules in a predictable order and he claims that one of the last rules you learn is the third person –s rule (J. E. Alatis, 1983:2). The students I have investigated seem to be at very different levels in their interlanguage concerning this rule. Some had acquired it very well, i.e. the student who had 13 examples of verb agreeing with third person singular with no errors or the boy in interview 1 who had 22 examples, also he with no errors. We can also detect four students

(37)

who were in the developmental stage of Avoidance, or as Lightbown and Spada calls it, Simplification. What seems significant is the use of or contact with the target language outside the classroom, and also if it is a long-term or a short-term contact. Lightbown and Spada also claim that if the students do not recognize the rule from their own language, they may eventually learn the rule, but “restrict its applications” (85).