By Ýpek Serengil

 

Second Language Learning: Performance

Analysis and Discourse Analysis

 

For quite a long time, errors of L2 learners were deeply analyzed, though the learning process was ignored. In the last twenty years the focus shifted to the process of learning. A mistake can be corrected, but it will not be erased from the mind unless the source of it is studied. Performance analysis allows us to see the development of a learner, the changes in her/his way of using the language gives us cues about L2 learning. This work aims at presenting the syntactic, morphological, lexical and phonological studies of L2 development.

 

METHODOLOGICAL ASPECTS

One of the greatest problems in the description of actual language development is because of the fact that linguistics has not developed the descriptive devices, which could represent the dynamics of this process. There are not many distinguished studies in this field. The major studies in this field belong to Corder(1971), Nemser(1971) and Selinker(1972).

The common point of these three are: the L2 learner constructs an internal grammar on the basis of the L2 input he receives; a grammar which in subsequent stages or “varieties” keeps being reconstructed and will approximate a certain target variety of native speakers of that language more and more, although it will rarely be identical to it. However, most of the L2 studies do not supply us with concrete hints to describe how these varieties change. Heidelberger Forschungsprojekt is an attempt to describe this development over time. This project supports the idea that varieties do not only have a diachronic, but also a synchronic dimension. The quality of L2 use is not only determined by time, but also by circumstance. Thee are no single style speakers. Every speaker shifts linguistic variables, if the following factors change:

* social status of addresser- addressee

* language medium (spoken-written)

* topic of discourse

* linguistic task

* physical surroundings

* amount of attention paid to speech

Another point to keep in mind is that language change does not manifest itself in the form of development, leveling or regression may also show up. Selinker names these two negative changes “fossilization”. The possible causes of these are:

-         low motivation for learning L2

-         age

-         limited range of L2 input

In addition to the mentioned methodological problems, the differences between the researches are based on the choice of period of time, informants, and data collection.

 

PERIOD OF TIME

 

There are two research types related to the time period: longitudinal and cross-sectional. The properties of longitudinal research design (the language behavior of a group of informants is registered for a certain period at specific intervals- 4 weeks, 6 months) are listed below:

  1. It gives a picture of language development over time
  2. It allows statements on the order of acquisition
  3. It is time consuming, and also not very economical
  4. It is difficult to apply to large groups

The properties of cross-sectional research design (one single sample of the language behavior of a group of informants is taken) are listed below:

  1. Time factor is eliminated
  2. It allows statements on the order of accuracy
  3. Most researchers prefer it as it saves time and money

To avoid the disadvantages of both types a combination of these might be used, which is called quasi-longitudinal design (cross-sectional samples of language behavior of different groups of language learners-at earlier or later stages- are collected simultaneously and compared with each other).

 

INFORMANTS

 

Statements about language development can be based on observed behavior of different groups of one, some or many informants. Some well-known studies of L1 acquisition are based on single or multiple case studies. Case studies have many disadvantages when compared with large groups.

DATA COLLECTION PROCEDURES

 

Statements about language development may be based on data, which may have been gathered in very diverse ways. There are many methods of data collection, ranging from observation of natural behavior to the most artificial experimental tasks.

 

SYNTACTIC DEVELOPMENT

 

In this section two projects concerned with syntactic development will be discussed: HDP by W. Klein and a project by Cazden.

HDP: a large group of adult Italian and Spaniard L2 German learners’ (2*24) language learning behaviors is analyzed in an L2 environment; data collection is based on naturalistic techniques; and a cross-sectional design is used.

Cazden: six speakers of Spanish (2 adults, 2 adolescents and 2 children) L2 English learners behaviors in L2 Environment is analyzed; data collection is based on naturalistic techniques; and a longitudinal research design is used (duration of 10 months).

 

HDP

 

The HDP started in 1974 in Germany, Heidelberg area. The informants were adult immigrant workers from Italy and Spain. Data collection was both based on participant observation and non-standardized interviews. Participant observation was systematically done in factories, pubs and an aliens registration office. The corpus was restricted to 100 sentences per informant. Klein and Ditmar first of all introduce the concept of a variety grammar as a tool for the description of second language learning. The proposed variety grammar is probabilistic and context free.

Within the HDP the focus is on the development of specific syntactic categories and the development of word order characteristics. For the syntactic description of learner and native varieties, a context free grammar was used. The proposed grammar contains 101 context free rules grouped into 15 rule blocks. All applications of all 101 rules were counted per speaker, and on this basis the probabilistic weights of all rules within each rule block were calculated: these give a “syntactic profile”. A special cumulative syntactic index was formed in order to standardize the profile per informant. The combination of the following eight criteria determines the syntactic index per informant:

1. - /- Presence of subject

2. - /- Presence of verb

3. Complexity of VG with/out Aux/Mad

4. Complexity of verb complements

5. Degree of pronominalization

6. Complexity of NP

7. Type of determiner

8. Type of adverbial complex

Finally the influence of various personal and environmental factors determining the success of L2 learning in the domain of syntax were studied. Of the factors analyzed, the following six proved to be the most important ones:

    1. contacts with Germans in leisure time
    2. contacts with Germans at work place
    3. formal professional qualification
    4. age, at time of immigration
    5. attendance at school in the past
    6. duration of stay in Germany

No significant relationship was found between the level of syntactic performance and sex or mother tongue. The most important factors are the two contact variables. Duration of stay seems to be important only for the first two years. The learning process then slows down, or even stops, and may start moving again when other environmental factors change.

A number of concluding remarks on the HDP should be made:

a)      Implicational analysis offers a better method of dividing the informants.

b)      As cross-sectional method is used we cannot talk about a healthy conclusion on acquisition order.

c)      The HDP has a strongly descriptive orientation. Little attention is paid to explaining the orders found.

d)      The language base is normalized in HDP. The spoken characteristics of German are ignored.

e)      The limited range of utterances is not suitable to the detailed techniques used.

 

CAZDEN PROJECT

 

This is an L2 English Learning analysis; it is far too limited when compared with the HDP. It, however, has a better presentation of actual L2 data. It is largely based on naturalistic data collection techniques. The data were collected through spontaneous conversation, elicited conversation, pre-planned sociolinguistic interactions.

Each data collection session involved three people: the subject, the experimenter, and a bilingual transcriber. This is a longitudinal (10 months) project, which is limited to a multiple case study of six Spanish speakers- learning English.

According to the findings of this project, negatives are acquired in four stages. The development of WH-questions seems to take place in the following stages of the learning period. 

 

MORPHOLOGICAL DEVELOPMENT

 

R. Brown’s longitudinal multiple case study of L1 development (14 grammatical morphemes were used) with three children led to the following conclusions:

Many L2 order studies aimed at the development of grammatical morphemes; such as articles, prepositions, and pronouns. These have minor roles, but they are most frequently used in language.

 

LEXICAL DEVELOPMENT

 

Although the L2 speaker’s lexicon determines to a large extent his possibilities of communication in the second language, research into lexical development is quite limited. This is probably due to the reality that interesting problems of lexical development do not emerge until relatively late in L2 development. Levenston presents a few interesting questions to be discussed:

Levenston has presents a few formulas related to learning L2 lexicon:

  1. Avoid two forms what appear to be identical: (the- L1 Eng) and (le/la- L2 French)
  2. Try words which may be generalized to a large number of contexts
  3. Avoid words, which present semantic difficulty.

A longitudinal study in L2 lexical development is Yoshida. This project is carried on 7 months with 3-year old Japanese L2-english learners. In addition to observation, comprehension and production tests were administered in the 5th, 6th, and 7th months. As a result; the subject’s comprehension proved to be far superior to his/her production. The average vocabulary use of a child was 260 words. However, main verbs were so frequently used, auxiliary am-is-   

 

Realized                                                    Intended

Miki is long fish                                       Miki caught a long fish

You are chocolate?                                   Do you have chocolate?

 

PHONOLOGICAL DEVELOPMENT

 

There are interesting findings in phonology in an attempt to explain the slips of the tongue. For instance when an L2 learner is under pressure to communicate, s/he seems to attend to utterances selectively, considering only certain phonetic features, and ignoring minor variations in the pronunciation of words, etc. The results prove that L1’s phonological systems interfere with L2’s system.

Wode distinguishes two basic phonological processes in the learning of oral productive L2 skills:

-         If there is enough equivalence between L1 and L2 elements, it will be sufficient for L2 learners to substitute the familiar L1 elements in L2 speech;

-         If there is none equivalence between L1 and L2 elements, the developmental process may mirror gradual L1 learning patterns to a considerable extent.

In both cases the L2 learner uses his mastery of L1 phonology as a basis, L2 phonological elements seem to be scanned by the learner for equivalencies and non-equivalencies: L2 elements that are within the as yet unknown and crucial range of equivalence are matched and substituted and by their respective L1 equivalent, whereas other L2 elements undergo developmental changes which are often parallel to changes in L1 learning. As a result, the phonological errors in L2 learning differ systematically as a function of the phonological system of L1. 

 

DISCOURSE ANALYSIS

 

The basic idea of discourse analysis is that the study of language in context will offer a deeper insight into how meaning is attached to utterances than the study of language in isolated sentences.

The context of language may be considered both from a linguistic and from a social perspective:

-         In most situations of language use, utterances will be preceded and followed by other utterances, resulting in a dialogic or monologic text;

-         In all situations of language use, specific social relations between speaker and hearer will guide the structure of these utterances.

Hatch and Long and Brown give a survey of central research questions in discourse analysis and discuss such topics as a description of different speech acts in different communicative settings, and a description of conversation rules. Hatch and Long list several system constraints for communication, such as: back-channel feedback, turnover signals, or norms a la Grice. Although these norms are universal, the way they are employed may vary between specific social, cultural or linguistic groups.

Until now, the attention has been focused on mainly on the structure and function of na6tive language input to non - native L2 learners. Now we will discuss the term “foreigner talk”.

 

FOREIGNER TALK

 

This concept has been developed to describe the adjustments in the speech of a native speaker of a language who is in verbal contact with a non- native speaker or learner of that language. Some aspects of foreign talk are listed below due to the research studies of Henzl, Meisel, Ferguson and DeBose, etc.

Morphosyntactic strategies:

-         avoid subordinate, embedded clauses,

passive constructions, SV inversion

e.g. She was seen at shoplifting by two young officers.

-         omit function words, prep., art., aux.

e.g. look me, listen me

-         produce well formed utterances by avoiding false starts, unfinished sentences, slips of 

the tongue

-         produce short sentences

Lexical strategies:

-         avoid colloquial expressions, regionalisms, slang words, idioms

-         repeat words or phrases

-         reformulate words by using easier synonyms

-         use confirmation checking devices

-         use words which belong to the “native stock” of the L2 learner.

Phonological strategies:

-         speak slowly by: using many and long pauses between words, avoid reduces forms: e.g. He’ll be here at 10:30- use He will be here at 10:30.

-         Speak loudly by increasing the volume for key words

-         Speak with exaggerated pronunciation.

-         Speak with exaggerated intonation.

These strategies are mainly used by teachers in class communication or in a communication between a native and a nonnative. However, the L2 learner may find these strategies insulting. There is also another question to be discussed: Should a learner be exposed to the real language spoken outside, in daily life? Should S/he learn the language in its most understandable ways and decode the real life discourse later? As Foreign Language teacher what sort of an attitude should we take?

 

TURN TAKING

 

This means turn shifting between speakers. A research is done in this field with adult L1 english speakers and L2 english speaking children. The following findings emerged:

  1. Turn allocation devices were present in their speech from the beginning.
  2. The proportion of utterances containing TA devices nearly doubled or even tripled during the period of study.
  3. Two general classes of YA devices were used: questions and attention directors.
  4. the individual variation in the relative frequency of each of these devices over time was considerable.

For the L2 learner, The development of TA devices is of crucial importance, because such devices enable him to elicit information about the new code he is trying to learn. However, prior linguistic knowledge will facilitate the use of TA devices. An analysis of turn taking patterns in the classroom has been carried out by Allwright. He studied turn taking behavior in two groups of initial ESL- learning adults. His observations are based on audiotapes, collected biweekly session, for a period of ten weeks. A remarkable finding is that the Russian learner- Igor’s success in getting turns as frequently as he does depends more on an inability to make himself understood than on any ability to develop a topic. However, it is impossible to generalize this idea since it is not proved in several other situations.

 

CONVERSATIONAL CORRECTION

 

Verbal interaction involves all sorts of attempts at understanding and being understood. One important attempt is correction. There are two types of correction: self correction and other correction, both of these are highly organized:

The following observations about the structure of SCs and Ocs were made:

Self- correction:

-         word replacement: We talked about this school- university

-         rearrangement: Wha-What kind of English do they- what do they say?

Other correction:

-         modified repetition:

A: Where did you buy your clothes?

B: Where did I buyed it?

-         question words querying an entire proposition: hmmm, what?, where?, when?

-         Partial repetitions: About five hours? Too many what?

-         You mean- reformulation: Yesterday? You mean after school?

 

By Ýpek Serengil

 

Home / ELT Materials / Coursebook Reports / Learn Turkish / Comics / ELT Conferences / Private Lessons / Online Translation / Links /