Second Language Research Methods. Herbert W. Seliger

are intended to supplement the content of the chapters in which they occur, and to help the reader to think and experience rather than simply read and go on to the next chapter.

      1 What is research?

      Race made a slight grimace. ‘I’m used to that. It often seems to me that’s all detective work is, wiping out your false starts and beginning again.’ ‘Yes, it is very true that. And it is just what some people will not do. They conceive a certain theory, and everything has to fit into that theory. If one little fact will not fit it, they throw it aside. But it is always the facts that will not fit in that are significant …’

(Agatha Christie: Death on the Nile 1938)

      Introduction

      One might say that knowing how to look at research problems and how to carry out research is related to the state of mind of the researcher. The state of mind of the researcher reflects, to some extent, the world in which he or she lives. What researchers believe, what they accept as forms of knowledge, is often a reflection of their social and cultural context. While we would like to think that research is objective and always seeks to show the ‘truth’ or the ‘facts’ about some phenomenon which is being investigated, we are not always conscious of the influences around us.

      We are living in what we think is a period of scientific objectivity. However, Aristotle, Ptolemy, Galileo, Copernicus, and Einstein all felt this way about their own times and accepted as scientific facts things which later generations would disprove or criticize.

      This chapter will be concerned with establishing some ways of thinking about research in general and about second language research in particular, by discussing the following questions:

      – How is research similar to and different from naturally occurring cognitive processes involved in learning?

      – What are the differences between conclusions reached on the basis of ‘common sense’ and intuition on the one hand, and scientific procedures on the other?

      – And, finally, when we have reached our conclusions, how do we know if they are reliable? That is, how do we know when we really know something?

      Research as a natural process

Research and reality

      Many of us have images of the researcher dressed in a white smock, spending hours in a laboratory or facing a blackboard filled with arcane mathematical notation, suddenly crying out with great emotion, ‘Eureka!’ and making a discovery that will forever make his or her name remembered down through the generations. What is closer to the truth is that, as part of natural human activity, we all carry out activities that have the basic characteristics and elements of research.

      Not all research is carried out in laboratories or involves complicated statistics. In fact, ‘research’ is so common that it surrounds us on a daily basis. Every day, we carry out activities that have the same basic components as academic research. We carry out these activities because we seek answers to questions about phenomena that occur in our daily lives.

      Let us imagine the following scene: An infant is in a playpen, fenced in from the surrounding world, not yet in control of the language that will enable him to communicate with those around him. The child’s parents are sitting in the same room and the child sees an object on the other side of the room. The child wants the object but cannot get out of the playpen. Nor is he able to say, ‘Mom, please hand me the ball.’ The child views the object and attempts to voice a sound he has come to associate with it. ‘Baa,’ he calls. No response. ‘Baa,’ he calls again. This time one of the parents notices the child. ‘What is he saying?’ asks the parent. The mother suggests to the father that the child is hungry. ‘No, I don’t think so,’ answers the father, ‘he has just eaten.’ ‘Maybe he needs to be changed,’ offers the mother. ‘Didn’t you just change him after he ate?’ asks the father.

      After many guesses (or hypotheses) about what the child is trying to express and the rejection of these possibilities for various reasons, the parents guess that the child wants something. The child’s word ‘baa’ resembles something in the room, a brightly colored ball. The mother develops a new hypothesis. Perhaps the child wants the ball. She crosses the room, picks up the ball, holds it out to the child and says with rising intonation, ‘Ball?’ The child indicates that this is what he has been trying to say by responding with a smile and a squeal of excitement.

The basic components of research in the real world

      We note that this everyday experience contains the following components:

      1 There is some form of behavior that is not clearly understood: the child vocalized.

      2 The behavior is observed and found to be special for some reason. Questions are raised regarding the reasons for the behavior: why did the child, who had previously been quiet, begin to vocalize?

      3 Several possible explanations for the unusual behavior are sought and related in some way to previous knowledge or past events: the baby’s unique behavior is related to experiences that he had just had, such as eating or being changed.

      4 One of the possible explanations is considered to be the one that most probably explains the behavior. This becomes the parents’ ‘hypothesis’ about the behavior, based on both observed facts and previous knowledge. The other hypotheses are rejected.

      5 To test this hypothesis or answer the research question, more data need to be collected about the baby’s behavior in other instances of vocalization. In this case, the parents’ experiment seems to indicate that the child was indeed vocalizing the word ball.

      Another way to be certain that the baby was saying the word for ‘ball’ would be to wait for a second occurrence of the sound ‘baa’ and pick up another object in the room to check the baby’s reaction. Several repetitions of this procedure would lead to the confirmation or rejection of the hypothesis.

      Notice also that the parents’ hypothesis was formed on the basis of an observed language phenomenon that occurred naturally. Another way of testing the hypothesis would be for the parents to try to elicit a vocalization from the child. They might decide to see whether the baby responds differently when shown different objects and whether he responds consistently with the sound ‘baa’ when shown the ball. For example, if the parent held up the ball and the child responded with some other word, the parents would have to revise their thinking about the baby’s language development. They might conclude that:

      1 The hypothesis about the relationship between the vocalization ‘baa’ and the ball is premature. That is, the parents were imputing to the child linguistic abilities that did not really exist.

      2 The child is at a stage in his language development in which he sometimes uses the appropriate word for the object and at other times says whatever comes to mind. In other words, it may be that the child is experimenting with his language ability.

      The above demonstrates that research questions can be investigated from many different perspectives by using different procedures such as observation, questioning, experimentation, and elicitation. Once hypotheses have been formed, tested, and confirmed or rejected, it may be necessary to repeat the experiment or reconfirm the conclusions by researching the question using different means. As we shall see in Chapter 2, each of these different perspectives affects the way the research is carried out.

      Research might also be conducted internally in what Einstein referred to as ‘mind experiments’. In this case, questions are raised and answers found by using reason and logic. A second language learner may listen to others using the language he or she is trying to acquire and think about the way native speakers use the language. It has long been accepted that the process underlying the acquisition of a second language grammar is hypothesis testing. The following is an actual example of an internal experiment carried out by an eight-year-old second language learner and later reported to her parents.

      The child was in the process of acquiring Hebrew as a second language. One day, she was playing with her Hebrew-speaking friends as they were climbing up a rocky hillside. The second language