Research method of
onomatopoeia and of phonetic symbolism
Junka AKIYAMA
１． The introduction
In language, the relation
between the sign and the signification is arbitrary. But, some words have the
phenomenon the phonetic symbolism what have some relation between the sign
and the signification. It is in the onomatopoeia that this phenomenon appears
notably. The onomatopoeia is the language what not only has the ability to
express the sense and the action in the dimension of movement or action
, and but also can express the movement of feeling. The onomatopoeia is the
word expressed by the direct imitation of sound, or the word that express the
sensuous impression with the sound of the word, we can see the relation non
arbitrary between phone and signification. In this report, Ill introduce the
research method of onomatopoeia and of phonetic symbolism.
２． The research method
In the past research or phonetic symbolism
and of onomatopoeia, the researchers used very variable research method. And
Ill show some researches method with the example.
２－１．SD Method・Factor Analyze
We often use this method in
the research of phonetic symbolism and of onomatopoeia. The SD Method (Semantic
Differential method) is the method defined by the psychologist, OSGOOD. This
method estimate a personal impression for a matter with the pair of opposite
adjective, each pair having a scale, and we try to distinguish the semantic
structure of the object by the extent of each scale. The SD Data is structured
in three factors, Evaluation , Potency and Activity . Then, The
Factor Analyze is one of the statistical methods, to describe the structural
character with the least common variable in the correlation between some scale
value or the value what quantifies some semantic characteristic. And we develop
its randomness in the plus and minus direction on the dimension of each factor
or objective factor about the structured group of scale, and we use this method
for estimate meaning.
２－２．The research method of
phonetic symbolism
First, Ill show a research, the correspondence figure and
sound . Many researchers have used this method, and it makes a series in the
domain of research of phonetic symbolism. The object of this method is to analyze
the phenomenon of phonetic symbolism under this procedure; the researchers
present one or some word non-signification for a figure, and they make the
subjects estimate them or choose a best fitting word. The first research of
correspondence figure and sound, its the research by KOHLER(1929), and the
following researches often refer this experience. He presented a curved figure
and a straight figure (figure 1) to the subjects. Also he presented two word, maluma, and takete, and he made the subjects choose which words suit for these
figures. As the result of this experience, 97% of these subjects chose maluma for a curved figure and 94% chose
takete for a straight figure.
Many researchers
have used this experience, but TAYLOR&TAYLOR（1965）
criticized this method ; he argued that we cant analyze which sound has
which effect with this method. For example, when we analyze phonetic sound in
the change of vowel, we cant have a significant result, sinon we examine all
posible combination.
Secondly, Ill
present researches of objective phonetic symbolism . This word is used by
TAYLOR&TAYLOR（1965）, and the
purpose is to find objective traits of phonetic symbolism in certains natural
language. As the basic procedure is the comparison of words ; the
researchers present to their subjects their mother tongue and a strange
language for them who correspond their mother tongue, and they make their
subjects choose a word suitable for mother tangue in the groupe of strainge
language. But it is true that we have some doubts about the result, becaouse
this experience have some problems in the condition of subjects and of
experience, the choise of strainge language or words, and in the translation.
To avoid the effect by mother tongue, some researches compaired between strange
languages, but it is rare and it often had negative results.
Last, I refer the researches with SD Method,
the experience by Oyama＆Haga(1963). They used SD
Method with correspondence of figure and sound . And they presented to their
39 subjects five words non- signification in the form of CVCVCV (for example,
ramara rimiri ), and made them estimate with same SD scale. They used a,
e, i, o, u as vowels and r, m, k, t as consonant. They got three factors:
Stability , Clearness and Potency . And they concluded that for
example u and o have the symbolism deeper, further, more full, softer,
hotter, wetter and smoother than i and e . The following tables are the
result of this experience.
In the analyze of phonetic symbolism, the researchers
have done various experience and some experiences got the positif result for
universal phonetic symbolism, but there are also some problems that the result
depends on the condition of subject or exoerience. So it is difficult to
analyze phonetic symbolism objectively.
２－３．The research method of
onomatopoeia
In the research method of onomatopoeia, there is some
sorts, for example, to analyze the phonetic symbolism in onomatopoeia, or to
analyze the function of onomatopoeia. First, Ill present the research of
Murakami(1980), he analyzed phonetic symbolism with Multi-quantity analysis. He
chose 65 onomatopoeia covering broadly the sound in japanese and did 5
analysis. The first analysis is to abstract the factor of sound, and from the
second analysis to the fourth analysis are to abstract the factor of
signification. The subjects are same in all analysis. The fifth analysis is to
examine the relation of the factor in the preceding analysis. As the result,
there was some relations between the phonetic factor and the factor of
signification, and it showed the phonetic symbolism. The following table is the
detailed result.
Second, Ill present the research of OSAKA, he examined
the function of onomatopoeia. This is an experience with SD Method and
Factory Analysis . He did the chief ingredient analysis with collecting
words respondent to onomatopoeia by subject, to analyze the deep structure of
496 japanease onomatopoeia. As the result, eight ingredients was abstructed :
action, laugh, liquid, hearing, mental condition etc. And also, he abstructed
two-syllabled repeated onomatopoeia from japanease onomatopoeia, and he
classfied them by 5 sense and under-classfied them again. In this phase, he put
onomatopoeia by its resemblance in the three-dimensional space. This
three-dimensional space was drawn in the method of the percepective
representation, if the space position is closer and closer, they are resemble
more and more, if it is further, they are resemble less and less. Figure 2 and
3 are an example of the resulut of this analysis.
There is also the researches with Magnitude estimation
method in the research of OSAKA. This method is a sturucture of sense scale
developped by Steavens. In the case of sense scale, we present to subjects some
sense stimulus by intensity and different quolity, and we put these subjective
intensity on the semantic scale in the form of number. First, he chose the 9
group of words what compose a cluster, and he presented to 19 subjects these
words as stimulus and made them estimate the impression of thse words. He asked
also them to estimate only a dimention of potency as a rule. The result of
analysis of estimate value, he conclueded that there is some significative effect
to image statistically. He has done another researches in the view of
pshycology, for example, the effect for memory, childrens acquirement of onomatopoeia or how to make
onomatopoeia by children.
The above is the research of onomatopoeia, and I think
that they have done very valious researches about onomatopoeia because there is
different view between the analysis of phonetic symbolism in onomatopoeia and
the analysis of the functions of onomatopoeia. In the first researches, most of
them use nearly same procedure and we can find the universal phonetic
symbolism, but in the second researches, the researchers analyze them more
individually ; about own language.