Alexandra Rejstrand
alexer@stp.ling.uu.se
Abstract
The goal of this Master’s Work was to design and implement a CALL program that teaches
the English phonetic alphabet. Before designing the program a survey of similar, already
existing software was conducted. The findings are a few programs for English-speaking
children with the purpose of teaching them how to read English. The program created is not
intended solely for children but for anyone in need of knowing the phonetic alphabet. Hence,
the resulting program lets the user practise the English phonetic alphabet with gradually more
difficult exercises. Since English exists in a variety of forms/dialects, British and American
pronunciation have both been chosen to be included in the program even though the focus is
on the sounds and words that Swedish learners of English generally have problems
pronouncing.
Introduction
This document is my Master’s Thesis in Computational Linguistics, with the specialisation
Computer Assisted Language Learning (CALL), within the Language Engineering
programme at the Department of Linguistics at Uppsala University. The Master’s work
consists of two parts; one practical and one theoretical, where this thesis constitutes the latter
part. The aim of the practical work was to design and implement a phonetics program, now
named Sephonics, with the IDE (Integrated Development Environment) Delphi and has been
carried out in co-operation with Arro & Wartoft AB, located in Uppsala. Also included in the
work was an investigation as to whether, and if so, in which format sound could be included
in the program. This essay is thus the theoretical part that documents the mentioned phonetics
program.
By way of introduction the essay starts off with some background information in chapter 2.
The first section deals with pedagogy and introduces different procedures that can be used
when learning language and traditional methods of teaching the phonetic alphabet and
pronunciation of English. The second section deals with user interfaces and thus describes
different aspects to consider when designing a user interface. The third section introduces the
notion of CALL and its advantages. The fourth section deals with problems of English and
pronunciation in teaching.
In chapter 3, the actual program is described, first with some general information and an
overview and then a detailed description of each exercise.
Chapter 4 describes how the program has been evaluated and discusses the results. A
discussion concerning the choices made when designing the program is also included.
Chapter 5 summarises the essay and mentions possible extensions that could be made to the
program in the future.
2 Background
2.1 Pedagogy
The most common pedagogical methods in language learning are the grammar-translation
method, the audio-lingual method and the natural approach. In short, the grammar-translation
method focuses on the form of the language and does not really care about the message that is
being transmitted through language. Also this method is not concerned about the spoken
language but only the written form. The audio-lingual method introduces phonetic writing at
an early stage and is mostly focused on the casual, every day, spoken language. This method
also cares most about the structure of the language and ignores the content. The natural
approach claims that there is a natural order in which linguistic features are learned and that
understanding precedes production. The affective filter, mentioned by Krashen, is said to
influence the learning in either a positive or a negative way depending on whether the
learner’s feelings towards the learning situation is good (= the filter is inactive and learning is
easy) or bad (= the filter is active and learning is difficult if not impossible).
There is also a more recent method, called the communicative method, that focuses on the
message being communicated instead of what form the message has been given.
2
From this information it seems obvious that Sephonics would be best suited in the audiolingual
method due to its early (in the perspective of the learning process) interest in phonetic
writing. In this method, Sephonics would be a great help for the student in mastering the
phonetic alphabet.
One might say that Sephonics uses a grammar-translation view applied to spoken language
instead of the written language. Since the form of the phonetic output of the learner is
fundamental to whether she will be understood or not, the program focuses on form rather
than content. In fact, Sephonics presupposes that the learner already knows the content of the
used words and does not refer to it at all (except for example sentences in some exercises).
The most effective method to learn a new language is probably to mix some features from all
approaches and get varied teaching with the student in focus. Reid introduced the notion of
learner preferences (Åsa af Geijerstam, p.c.), meaning that different learners prefer to learn in
different ways; a statement that becomes evident when reading Stewick (1989) where seven
(successful) language learners give their views on how they learn languages. The initial
assumption of the book was that since they all are very successful when learning languages
they would go about it in almost the same way. This is not the case, though; they all use
different methods in order to learn. The only common denominator is that they are always
aware of the language they are about to learn and are not afraid of being corrected. According
to Reid, learners can be oriented towards visual, audible, kinaesthetic or tactile ways of
learning. Visual learning means seeing words and language structures on paper while audible
learning means hearing them. Kinaesthetic learning involves touching, in the case of
computers clicking with the mouse and using the keyboard. Tactile learning means that one
wants to make experiments and see for oneself what is happening when some parameters
change.
What to think of when learning a new language
These learner preferences are to be considered when learning a new language, together with
the triangle of learning. The triangle of learning was made by Liberg (p.c.) and poses three
questions, namely: “What is to be learnt?” “Who is going to learn it?” and “How will it be
learnt?”. The answer to each question affects the answer to the other two. How something is
to be taught highly depends on who is going to learn it but not vice versa. Wenden (in
Tornberg (1997:72)) has replaced “Who” with “Why is this to be learnt?” with the
motivation that “who is going to learn” is included in the consideration of how the material is
to be learnt. To make language learning more effective each learner should ask herself these
questions before starting.
• Why do I want to study this language?
• How will I use the language?/What are my needs? (Ordering a café latte in an
Italian coffee shop or closing a business deal with an Italian associate?)
• Do I have to produce (i.e. write or speak) the language myself or is it “enough” to
understand what others have produced (i.e. read or listen)?
• Will I mostly be using the spoken or written language? Or both?
• How much time do I have to learn the language?
• Which learning style do I prefer? (The visual, audible, kinaesthetic or tactile
approach?)
• How do I feel about learning this language? (Motivation, self-esteem, interest in the
language and the people that speaks it, relation to the teacher and other students.)
3
• Is the language I am about to learn similar or not to the language(s) that I already
know? (Learning a language that differs in grammatical structure often makes the
student inclined to apply the already known structures to the new language. This
problem can be reduced with help of the so called contrastive analysis of the two
languages)
The answers are needed in order to optimise the teaching for each individual student. Then, of
course, it is not always practically possible for the instructor to fulfil all the demands of each
student but that is another (largely financial) matter.
2.1.1 Learning procedure
Learning to associate a sound with a sign is mostly accomplished by repetition: the computer
is an excellent aid in learning in that it enables the student to repeat a task as many times as
desired without being even the slightest bit annoyed. The opportunity to repeat is very
important in all training (Higgins & Johns, 1984:37.).
In addition to repetition, Strevens (1965:58) claims that: “Pronunciation teaching uses above
all mimicry and exhortation as its chief techniques. The direct and deliberate imitation of
sounds, either in isolation or in connected utterances that are very short – this is what, for the
most part, comprises this form of teaching.” Mimicry is very much supported by Sephonics
since sound is included in each exercise. Strevens (1965:111) continues:
We can divide the teaching of any pronunciation feature or contrast into three distinct
stages. In the first place we must make sure that the student can recognise the difference
between two sounds or patterns that are not contrasted in his own language. The second
stage is to let the student make the sound himself in isolation. […] The means to achieve
this are firstly imitation of a model, secondly reference to an allophone of a sound in the
native language, or another known language, and thirdly, in the last resort, physical
directions. The third stage is the production of sound in context, as part of a longer
utterance. Success in stage two will not automatically lead to success in stage three.
The first stage Strevens (1965) talks about is very much practised in the program’s exercises
one and two, as I will discuss further in the separate sections about each exercise. In the
second stage the user is left on her own (or, preferably, in the hands of a teacher) due to the
fact that the program cannot assess the users pronunciation in any way. Of course the user
should read the signs out loud when working with the program but the program has no way of
knowing if the user that is able to connect a sound with a specific sign is also able to correctly
pronounce that same sound. The third stage is practised as the user has access to example
words and sentences where the sounds are included to provide context.
The program is intended to be used at home, alone so that the user can listen to a sound as
many times as desired and even say it out loud without having to worry about others hearing
her. Knowing that others can hear you when saying things that you are not sure of how to
pronounce can be very stressful to some people. Quite obviously it would be very noisy if
several users, using this program, sat next to each other and wanted to use the listening
functions. Noise is obviously not a good ingredient in any learning environment. Another
positive effect of using computer programs in the learning process is that the learner can
advance in her own pace without having to adjust to others. This also reduces the stress since
the user is allowed to make mistakes/move forward without having to feel embarrassment in
front of the other members of the class.
4
This is in no way intended as a claim that the program necessarily must be run by a single
user at a time. The learning process might very well be helped by the learners working
together with the program, discussing which answer might be the correct one and helping
each other to get the pronunciation right. The pedagogical gains and losses of co-operation is
an other matter and a discussion of it can be found, for example, in Tornberg (1997:67).
2.1.2 Traditional methods of teaching the phonetic alphabet
There is, to my knowledge, only one method that can be described as traditional when
teaching the phonetic alphabet. This is the method of letting the class listen to tapes with
recorded sounds and possibly repeat the sounds while looking at a chart of the transcriptions.
This method is not very practical since many students have to share one tape which is only
available while in class. It is not an example of individualised teaching. Learning to match a
sound to a sign is mostly done by repetition which this method does not encourage since the
students cannot practise at any time. The situation with only one tape for many students is
probably most often the case in school. In other language courses (where the students pay to
learn a language) a tape of one’s own is often included in the price (even video tapes might be
included to teach gestures etc.).
This method can be, and often is, combined with giving the students a list of the sounds to be
learnt and attached to them words in the students’ mother tongue where the specific sound is
included. This way it will most certainly be easier for the students to learn to transcribe the
sounds they already know how to pronounce and it gives them the opportunity to practise at
any chosen time. However, it is rare, if it exists at all, that two languages use the exact same
set of sounds and thus the same problem remains; to learn the new sounds the students are left
with the tape in class.
As for less traditional methods there are some applications on the internet (see for example:
Henry Rogers’s and Michael Stairs’s program “Phthong” (http://www.chass.utoronto.ca:8080/
~rogers/phthong.html)) but they are designed for linguistics students and thus do not have the
same purpose as Sephonics. The downside of this particular program, “Phthong”, is that it
does not provide any sound files for the user to listen to. There are also lots of sites with
general phonetic information about how and where the sounds are produced in the oral or
nasal cavity together with exercises on articulatory phonetics. See for example Kevin
Russell’s site (http://www.umanitoba.ca/linguistics/russell/138/sec1/ipa.htm), The virtual
CALL library site by Matthew Platts (http://www.sussex.ac.uk/langc/CALL.html) and
Hartmut Traunmuller’s guide to phonetics (http://www.ling.uu.se/staff/hartmut/tur.htm).
The above mentioned sites are all free to use but of course there are also commercial
applications on the internet. The following two sites both sell phonetics programs for Englishspeaking
children, seemingly with the purpose of teaching reading skills: http://www.
hookedonphonics.com and http://www.phonicsworld.com. A program for Swedish that reads
the words out loud for the user has been made by Allemansdata AB. This program is intended
to teach children up to 9 years old how to read and spell in Swedish.
2.2 User interfaces
An interface is the part of the program that displays information to the user. It is also the part
that should give the user a possibility to interact with the program. A simple interface is that
of MS-DOS which consists of a command line where the user is supposed to type commands
that the program executes. This sort of interface demands that the user knows which
commands are available. Another type of interfaces are the graphical user interfaces, GUI. A
5
GUI presents the user with a more visual and more easily understood interface where all
commands are shown by means of buttons and other devices.
There are (at least) two ways of looking at an interface; a visual and a functional approach.
The visual approach concentrates on how the screen is composed, how the used colours
interact and where controls and functions are placed. Simply put, it emphasises the visual
appeal of the layout to the user. The functional approach, on the other hand, concentrates on
how the user can interact with the program and how the program helps the user in that
interaction.
Of course the two approaches mostly overlap. One could say that the functional view is the
theory behind the visual view, which would then be the practical part. An example: the place
of the buttons placing the functions wanted by the user affects the way the user can interact
with the program. If the buttons are placed in a way so the user does not see them, that will
have a negative effect on the user’s ability to interact with the program and perform the
wanted task.
Designing an interface in a user-friendly fashion means making it as easy as possible for the
user to do whatever she might want to do. Nygren (p. 13. my translation) puts it this way:
“The goal when designing the layout is readability, clarity and consistency.“ Heckel (1984)
goes as far as talking of designing friendly software as an entirely new craftsmanship. There
are lots of things to take into consideration when designing an interface and most of all it
comes down to who is going to use it and for what purpose. As Chapelle (1997) puts it:
CALL developers need to consider how software can provide learners with opportunities
believed to facilitate Second Language Acquisition. In other words, it is useful to view
multimedia design from the perspective of the input it can provide to learners, the output
it allows them to produce, the interactions they are able to engage in, and the L2 tasks it
supports. Because CALL software can actually play a role in input and interaction, it is
useful to consider it as a participant in L2 tasks.
Jacob Nielsen has written many publications about user interface design and human computer
interaction and is considered somewhat of a guru within these fields. In one article he very
clearly states ten heuristics for usability in interfaces. The article summarises the 30 elements
that Heckel has outlined in his book “The elements of friendly interface design” and the same
heuristics appeared in Dix (1998:414) with slightly different terms as a help to evaluate
software. Nygren and Preece (1994) also agree. Below is a list of the heuristics outlined by
Nielsen with my own comments added.
Visibility of system status – It is important for the user to know what the program is doing and
especially to know that the program reacts to the user’s actions. This is something that Nygren
also states as very important. It is essential that the program reacts to the user’s actions
immediately so that the user knows that the program has received the command.
Match between system and the real world – the program should be as integrated with the real
world as possible, meaning that the user should not have to learn new concepts in order to
interact with it but be able to use her existing knowledge/language.
User control and freedom – if the user has control she feels secure and a secure person is a
better learner. (See section 2.3.2, Advantages of CALL, for reference.) The user should be
given the power to correct mistakes, i.e. the possibility to undo and redo.
6
Consistency and standards – one sign, one meaning. Do not change the meaning of buttons
with the same name or the user will most likely be confused and make unnecessary mistakes.
Not to mention the frustration and confusion such mistakes cause.
Error prevention – every error prevented is a victory. If errors are prevented you do not have
to think about what to tell the user to make her correct it. Error messages are difficult to
formulate in such a way that the user understands what is expected in order to get on with the
program.
Recognition rather than recall – the user should not have to recall information as of how to
use the program. It is better to have that information visible to the user, or at least easy
accessible.
Flexibility and efficiency of use – there should be short-cuts for the most frequent user actions
for the more experienced users as well as declarative ways of doing things for the less
experienced users.
Aesthetic and minimalist design – no unnecessary information should be visible to the user
since it competes with the important information that you want the user to observe. Lots of
frills distract the user’s attention from the important information.
Help users recognise, diagnose and recover from errors – giving the user instructional and
understandable error messages prevents her from ever making the same mistake again. And
error prevention is good.
Help and documentation – might be needed even though it is best if the program can be used
without it. If needed, the help should be as short as possible while still instructive.
As far as possible these heuristics have been incorporated in the program to make it as user
friendly as possible, see section 3.3, Detailed description, for details about how the heuristics
have been applied.. Thus, a good interface is not noticed, it is just there and lets the user
interact with the program without intervention.
2.3 Computer assisted language learning (CALL)
A discipline that combines the two above mentioned areas (pedagogy and user interfaces) as
well as other areas (artificial intelligence, computational and applied linguistics, psychology,
human-computer interaction) is CALL, which is an abbreviation for Computer Assisted
Language Learning. CALL stems from the notion of CAL which means computer assisted
learning in general.
CALL systems can be divided in two types, frame based and intelligent ones, the latter
referred to as ICALL and discussed later on. Frame based systems can be linear, branching or
generative. Sephonics is a branching system in that it not only accepts an answer but lets the
user try again. In all exercises more or less sophisticated help is available to the user e.g. in
the second exercise where the right answer blinks if the user has not found it within 4 clicks.
The program is also generative since it presents new questions to the user each time an
exercise is run.
7
ICALL systems were, according to Bowerman, “born out of the shortcomings of frame-based
CALL systems” and include the following modules.
The expertise module, which contains the domain knowledge that prevents the system from
merely matching answers against those predicted by the programmer and enables it to analyse
the users input.
The tutoring module, which contains the systems’ “repertoire of teaching operations”, making
it possible for it to gear the helping facilities in the right direction for each student’s specific
problems, i.e. give context sensitive help.
The student module, which contains the model of each student’s progress. The model is
needed in order for the system to know which of its teaching operations to choose for each
student.
The user interface module, that enables the user to comfortably interact with the program.
2.3.1 Different types of CALL
Computers can be used to assist language learning in many different ways. Similarly, CALL
programs can be viewed from different perspectives. Programs can be either tools or tutors,
the computer can take part in the interaction or leave it to humans to interact with each other
via the computer, so called CMC, computer mediated communication.
Programs designed as general working tools might well be used to assist language learning.
These programs include word processing programs, e-mail programs and corpus programs
and they help the user by letting her get in touch with the target language and people who
speak it as their mother tongue. The internet can be a big resource in getting access to texts in
the target language even though it was not made to aid language learning.
CALL programs specially designed to aid teaching, on the other hand, still do mostly the
same thing as text books but in a more interactive way. There are text based programs for
doing drills and cloze exercises as well as graphically advanced programs that take the user
on a trip to a country where the target language is spoken naturally and the user has to carry
out a mission where she needs knowledge about that language.
Simple drill programs are, however not to be discarded due to lack of graphics or layout. They
can be just as effective in learning language as a more advanced program. First of all
everyone does not have the equipment (or finances) to use advanced techniques so the less
advanced programs might get used more widely. Second, learners might feel that they want a
program that teaches them a specific feature and not a ‘simple game’, i.e. they are not
motivated to use the game version even if it teaches them the same thing as another program.
Third, it is often the simple, straightforward programs that are fun even after having used
them for a long while, for example solitaire and other programs like it that appeal to the user’s
stubbornness and willingness to try “just one more time” to get it right.
2.3.2 Advantages of CALL
CALL tackles questions of how computers can best be used to help people learn language i.e.
how computers can make it easier, more effective and more fun to learn a language. The
CALL software’s task of facilitating language learning is mainly due to the facts that learning
a language takes a lot of time and for many language students learning is not fun, sitting in a
8
classroom with at least 25 other students trying to get the teacher’s attention. Computer
programs, then, offer an individualised way of learning since the level of difficulty can be
adjusted according to the specific level the particular student is at right now. Advanced
computers also offer the ability to include moving pictures and sound in the programs, which
ought to make the learning better since sound is a vital part of a language. This means that
computer programs can include most of the different learner styles discussed in section 2.1,
Pedagogy a feature which makes the programs not just more varied but also better adjusted to
different learners.
Even though CALL is no longer a new discipline, computers have not been used to teach
language to a great extent for a very long period of time and it is still uncertain whether they
are really useful. Higgins and Johns (1984:86) discuss this question and conclude the
following:
We do not yet know whether the computer is a suitable supplement or aid to the vast bulk
of the work done in classrooms or by learners on their own. We do not know whether all
students should be urged to spend time on CALL activities, or whether it should remain a
voluntary activity. We do not know whether CALL has equal value for students at all
levels, or whether it favours one particular level.
This statement was made in 1984 when most CALL programs consisted of small, text-based,
drill exercises. However, it is still true. What do we really know about the effect/effectiveness
of CALL? It ought to be wonderful for reasons of variation and motivation but is it? Does it
really contribute to an increase in learning and if so to what extent? The different results of
studies of the effect of CALL on language learning are brought up by Gustavsson (1999:11),
where she claims that some studies find that CALL has a positive effect, some that it makes
no difference at all to the learning and some studies show that CALL even has a negative
effect on learning. Thus, the question of the effectiveness of CALL can probably not be
answered until CALL programs have found their natural place in the process of learning a
new language and an objective technique for studying learning is developed. However, a
complicating factor here is the commonly noted fact that the whole learning situation changes,
so that it may well be impossible to isolate the contribution made by CALL (Pedersen 1998).
The most commonly mentioned advantages of using CALL in teaching are listed by
Gustavsson (1999).
Variation – by simply being something out of the ordinary in regular teaching.
Motivation – computers are still exciting and connected with fun for most students.
Individualisation – by offering different levels of difficulty for the students to choose from
and also letting them work in their own pace. Features like these suit fast learners, who might
want to proceed faster, as well as slower ones, who are given the possibility to take the
needed extra time.
Lessening the demands – students do not feel as much pressure from the computer as they do
from the teacher and thus can relax more and concentrate on the learning. Secure and
comfortable students learn more than insecure ones.
Student control – the students are given the opportunity to control their own learning.
9
Lessening the burden for the teacher – by letting some students work with the computer the
teacher gets more time to help the rest of the class.
Multimedia –several senses work together, which increases learning.
Interactivity – by making the student interact with the program she becomes more active than
she would be by just listening to a lecture.
The ability to run the program at any time (if access is given to it outside of class, of course)
is another great advantage of CALL. It makes the students less dependent on the teacher,
although human interaction cannot be replaced by computers. When finding that a particular
structure gives you problems, which is bound to happen sooner or later when learning a new
language, you most likely will want to practise that structure with someone that already has a
command of it, until you get it right yourself. This procedure very soon becomes boring for
the person already in control of the structure. Using a CALL application with the ability to
run the program as many times as wanted, practising the same phenomenon over and over
again, saves a lot of energy for all involved.
These advantages apply to well-designed CALL programs. To know if a CALL program is
well-designed (i.e. does what it is supposed to do in a user-friendly fashion), it will be
valuable to investigate the following topics, most of which are brought up by Chapelle
(1997:22):
• the quality of the interface (see section 2.2, User interfaces, for reference of
how to evaluate the interface);
• the quality of the program’s language processing;
• the learners’ apparent (dis-)satisfaction with the program;
• the learners’ ability to engage in cross-cultural exchanges.
A well-designed CALL program does not abandon the user when she has given an incorrect
answer but helps her cope by hinting what is the wanted input. This type of feedback is called
intrinsic and is mentioned in the TELL consortium’s list of general design principles
(Laurillard 1996) as a feature that should be used if possible. Sometimes it is very hard to
design and implement exercises that allow intrinsic feedback, which is the reason why most
exercises give the user extrinsic feedback, meaning explicit information about what is wrong
and what is expected instead. The pedagogical gain with intrinsic feedback is that it makes
use of the fact that perception is faster than cognition. Hence, the user needs less time to
realise what to do if she can simply look at the screen to see what is wrong instead of having
to read and understand an explanation to know what has happened. Sephonics gives semiintrinsic
feedback in the exercises 3.3.5, Exercise 5 – How Is The Word Spelled? and 3.3.6,
Exercise 6 – How Is The Word Transcribed? (See Laurillard 1996 for an example of an
exercise which offers completely intrinsic feedback.)
2.4 Phonetics
2.4.1 Introduction to IPA
One of the purposes of Sephonics is to teach the English phonetic alphabet but the primary
purpose is to let the user practise it. The English phonetic alphabet is a subset of the
International Phonetic Alphabet (IPA, in short), designed by the International Phonetic
Association, also abbreviated IPA, in order to have a unique way of describing the
10
pronunciation of all languages, including those with no orthography (i.e. written form of
language). With IPA, a specific speech sound is represented by only one sign. This is not the
case in the English orthography as we will soon see. IPA organised the signs in two charts,
one for vowel sounds and one for consonant sounds. The signs are placed on the charts
according to how and where in the speech organs they are produced. Interested persons can
find out more about IPA at their web site (http://www.arts.gla.ac.uk/IPA/ipa.html). The charts
used in Sephonics are somewhat different from the ones made by IPA. Here follows a
description of the modifications made to the charts in order for them to better fit the program.
Firstly, the charts in this program only show the signs relevant for British or American
English depending on the user’s choice of target language whereas the original IPA charts
include signs for all sounds possible for the human speech organs to produce.
Secondly, both the charts have been compressed because of the lack of screen space in the
program. In the consonant chart this is done by omitting some columns in the original IPA
chart and merging some columns that do not use all lines so that the given space is used more
economically. In the vowel chart compression is achieved by using the original chart and
simply putting the signs closer together in a square instead of an uneven quadrilateral.
Thirdly, most of the consonant sounds can be divided into pairs of two sounds that are
produced in the same way, with only one difference: one sound is voiced and the other is
voiceless. Examples of such pairs are the sounds [p] - [b], where the latter is voiced and [t] -
[d], where again the latter sound is the voiced one. There are consequently two wellstructured
ways of displaying these pairs: either one after the other on the same line or one
underneath the other in the same column. This program uses the latter way because it is more
space-saving.
Fourthly, the charts in the program also include diphthongs and affricates (signs composed of
two other signs) which are not included in the original IPA charts. This latter modification is
due to the visibility of the signs on the screen. It would have been too messy with four smaller
charts. This presentation gives the screen a more uniform look. The digraphs’ place in the
charts are based on their first included sound. Thus, the digraph [eI] is placed to the right of
the single letter [e].
11
The IPA charts. [Reproduced by courtesy of the International Phonetic Association, IPA (c/o
Department of Linguistics, University of Victoria, Victoria, British Columbia, Canada), from
the website of the IPA: http://www.arts.gla.ac.uk/IPA/ipa.html.]
12
A picture of the exercise called “The Phonetic Alphabet” displaying the phonetic signs used in British
English.
2.4.2 English pronunciation
When coming across an unfamiliar word in English it is often difficult to know how to
pronounce it since in English a written letter or combination of letters can be pronounced in
many different ways (Wells, 1990).
An example presented by Wells is the English letter ‘s’, which can be pronounced [s], as in
sense [sen’s], [z] as in rises [raIzIz] or [Z] as in pleasure [pleZE]. At the beginning of words ‘s’
can also be pronounced [S] as in sugar [‘SUgE]. The point with this listing is that in English
there is no rule for the pronunciation of a specific letter or combination of letters: each word
must be considered separately, as stated by Wells (1009:631). At the beginning of the book
(p. xi) Wells claims that “English spelling is notorious for its shortcomings. Knowing the
orthography of a word does not enable one to predict its pronunciation with any confidence.”
This situation does not occur in all languages. However, Swedish does have this form of
peculiarity in, for example, the different ways to spell the so called sje- ([��]) and tje- ([S])
sound. The former sound can be spelled in these ways (among others): ‘sju’, ‘sked’, ‘skjorta’,
‘stjärna’, ‘dimension’, ‘nation’, ‘choklad’. The latter sound could be spelled in the following
(among other) ways: ‘tjej’, ‘kjol’, ‘kärna’. The difference between Swedish and English is
that Swedish does have some rules that tell you when these letter combinations should be
pronounced as these two special sounds and when to be pronounced as the string of letters the
combination is made out of.
13
Then, how is one to know how to pronounce an English word? A simple way to find out is to
look the word up in a dictionary (or listen to a native speaker, which could be trickier to find,
though). In a dictionary of English the pronunciation is described by means of the IPA and it
is fairly obvious that if the reader is not familiar with the phonetic alphabet she is not at all
helped by that description.
Because of this inconsistency in the orthography English pronunciation is a difficult thing to
learn and teach and that is why this program focuses on the English phonetic alphabet even
though the IPA covers all sounds possible for the human speech organs to produce.
2.4.3 Phonetics in school
An obvious place to obtain the knowledge of how to use dictionaries is school, but from what
I have experienced the phonetic alphabet is rarely taught in Swedish schools, and if it is
taught it is done very briefly. This opinion is based on my asking people in my nearest
surroundings, friends and family, if they got any training in reading English phonetic symbols
at school. This informal survey revealed that the teacher generally introduced the phonetic
alphabet as another way of writing but there was never any homework or testing of this
knowledge. Later on in school it was taken for granted that everyone knew how to use (read
and write) the phonetic alphabet. Nowadays it is not even brought up.
To get a more empirically based understanding of the education of the phonetic alphabet in
the Swedish school system a teacher, currently working at mid primary school level in a
Swedish school in Uppsala, has been consulted and some textbooks of English used at that
same level have also been examined. The teachers (Margareta Westergren Axelsson and
Göran Rönnerdal) at the Dept. of English at Uppsala university have been asked what
phonetic knowledge their beginner students have when they begin their studies of English at
university level. Margareta told me that, unfortunately there is no testing of the students’
phonetic knowledge in any way before they are given an introductory course in English
phonetics. However the students are tested after the phonetics course since they are required
to have some knowledge of phonetics in their future studies of English, which indicates that
phonetics is important when learning language. In this way the department can be sure that
everyone has sufficient knowledge of phonetics to continue their studies but they do not know
how much the students already knew before they took the course and what they learned from
the course they were given.
To retrieve this latter knowledge, a simple questionnaire was made for the students on the
mentioned introductory phonetics course. Hopefully the questionnaire would give some
knowledge as to how much the students had learned about the phonetic alphabet in their preuniversity
schooling.
118 students answered the questionnaire, the outcome of which was not quite as expected
since many students had misinterpreted the purpose of the questions and answered that they
had been taught the phonetic alphabet at the university even though the question was if they
had been taught about it in their pre-university studies.
59% said that they have been taught the phonetic alphabet, frequently remarking, however,
that the teaching was done very briefly. Many also said that the serious teaching had taken
place only at the university, which was not the question. 54 % said that phonetic writing had
been used in language class. One common remark here was that phonetic writing existed in
the books but was not actually used in class. 26% said that they do not currently use phonetic
14
writing when learning language while the majority said they use it sometimes. 62% said they
believe they would be helped by doing so.
However, most of the students have answered correctly to the three transcription questions
(97%, 66% and 83%), which is probably because many of them have studied phonetics earlier
at a university level and also in the present course where they filled in the questionnaire.
Why is phonetics not taught in school?
So why is it that the phonetic alphabet is not taught to any extent in Swedish schools? Perhaps
the reason could be found in the primary school curriculum (Skolverket 1996), where it is
stated that pronunciation is an important part of the linguistic communication and that the
students’ pronunciation is established at an early stage and observed during the complete
course of study. The model for the students’ pronunciation should be correct British or
American English. The goals are set to understanding simple and clear speech and being able
to participate in simple conversations after three years of study. After six years the students
should be able to understand clear British and American speech, actively participate in
conversations about ordinary topics and orally communicate something they heard, read or
experienced. Furthermore the students should be used to consulting a dictionary and grammar
as a help when reading and writing on their own. The grading concerning the student’s ability
to speak and write is determined by how free, varied and confident the student can be when
using the language. (Translation of the demands/goals stated in the syllabus for English.)
Thus, the school directives issued by the government emphasise the importance of a good
pronunciation, but in actual school practice this is often neglected. There could be several
reasons for this but only a few will be mentioned here. It might be so because it is more time
consuming for the teachers to assess spoken language than written. The teachers’ ability to
pronounce English themselves might also influence their willingness to include this factor in
the judging of students’ grades. If the teacher cannot produce native-like spoken English
(which, according to my experience with English teachers, is often the case) it is not fair if the
students should fail English class if they cannot either, since they have not had a satisfactory
model to learn from.
References
Bowerman, Chris. ICALL: An Underview of the State of the Art in Computer-aided language
teaching. Retrieved 1999-11-22 from the Word Wide Web: http://valhalla.sunderland.
ac.uk/cbowww/PAPERS/icai2.txt
Chapelle, Carol. 1997. CALL in the year 2000: still in search of research paradigms?
Language Learning and Technology 1 (1), 19-34. Retrieved 1999-11-22 from the World
Wide Web: http://polyglot.cal.msu.edu/llt/vol1num1/chapelle/default.html
Dillon, George. Phonetic Resources on the net. Retrieved 1999-06-29 from the World Wide
Web: http://faculty.washington.edu/dillon/PhonResources/PhonResources.html
http://www.weber.u.washington.edu/~dillon/PhonResources.html
Dix, Alan. et. al.1998. Human Computer Interaction. Prentice Hall Europe.
Gustavsson, Sara. 1999. Datorstödd språkinlärning i praktiken – utveckling och utvärdering
av ett CALL-program. Department of Linguistics, Uppsala University.
Heckel, Paul. 1984. The Elements of Friendly Software Design. Sybex.
Higgins, John and Tim Johns. 1984. Computers in Language Learning. Hazel, Watson and
Viney Ltd.
Johansson, Stig and Göran Rönnerdal. 1985. English Pronunciation. A Workbook.
Stockholm, Studentlitteratur.
Laurillard, Diana. 1996. The TELL Consortium – Formative evaluation report. University of
Hull. Retrieved 1999-11-09 from the Word Wide Web: http://www.hull.ac.uk/cti/
eval.htm
Malmberg, Bertil. 1967. Uttalsundervisning. Almqvist & Wiksell. Stockholm.
Nielsen, Jacob. http://www.useit.com/papers/heuristic/heuristic_list.html.
Nygren, Else. Grafiska användargränssnitt, några tips. Retrieved 1999-09-06 from the World
Wide Web: http://www.hci.uu.se/papers/60/
Østern, Anna-Lena. 1992, nr 3. Språkglädje och språklig medvetenhet – Träningsprogram för språklig medvetenhet. Rapporter från Pedagogiska fakulteten vid Åbo Akademi,
Institutionen för lärarutbildning.
Pedersen, Jens. 1998. Informationstekniken i skolan. En forskningsöversikt. Rapport 98:343.
Skolverket.
Platts, Matthew. 1999. The virtual CALL library (a list of downloadable programs).
Retrieved 1999-06-29 from the World Wide Web: http://www.sussex.ac.uk/langc/
CALL.html.
Tidak ada komentar:
Posting Komentar