O. INTRODUCTION
Child self-identification is an interesting topic to
educators because it helps them investigate when children happen to identify
themselves as full members of a community. In fact, when a child produces the
first personal pronoun "I", he, at the same time, produces a piece of language.
It is this " I " that makes him, identify as an individual.
Along that trail, many scholars such as Weeks, T.E. (1979),
Chomsky N. (1975), Halliday , M.A.K. (1979), Romaine, S. (1986) and others
have been interested, for a variety of reasons, in how and when children,
throughtout different stages of human development, from birth to adolescence,
do acquire useful language in which they need to communicate, first as
individual and then as members of their communities.
O.I. Problem statement and hypothesis
The interest of this study is justified in the sense that as
linguists have not yet been able to answer satisfactorily the question " When
does self-identification appear in children ? " i.e. when a child, as a human
being, can use the language to identify himself as an individual with the first
personal pronoun " I "; when he says, for example, " I am a boy ". This
research paper will try to provide a tentative answer.
Therefore, the hypothesis of this piece of work is stated in
such words as «All children aged from two years can identify
themselves as unique in a mirror», saying " Mie huyu " (This is
me).
0.2. Method of working
In order to answer the above question, there are two major
sources of data for this study. The first comes from the extensive body of
literature on the aspects of child's physical growth and his language
development, from 0 : 6 to 6 : 00 years of age.
The second source of data is from our investigation. We
obtained first a group of 40 children aged from 0 : 6 to 6 : 00 years. From
this number we, secondly, selected 15 children to whom we presented a mirror
and recent photographs and we paid attention to their different reactions and
answers to the questions we put them.
Furthermore, a questionnaire was conceived for children who
could answer our questions. On the contrary, parents whose children could not
speak helped us get information concening them. The questionnaires are given in
annex to this work.
In all cases, stratified and poised sampling is the method
which helped us find the number of children who were our subjects. This method
is the process by which a certain number of strata is divided so that the
variability of the elements of each stratum is more homogeneous than those
belonging to the different strata. In this case, the elements in the strata of
the sample must correspond to the number of elements of the population.
The formula we used was:
p = n or
N
p = n x 100
N
0.3. Delimitation of the topic
This work concerns children aged from six months to the age
of seventy-two months or six years (6 : 00), because these constitute our focus
as parent and educator. This age range is made up of four categories. The first
comprises three children aged from 0:6 to 1 : 5 years. The second is about five
children aged 1 : 6 to 2 : 00 years. The third one is made up of four children
aged 3 : 00 years. The last category is made up of three children aged 5 : 00
to 6 : 00 years.
The geographical areas concerned in this work are Mbobero and
Bagira since these two places share Kiswahili and Mashi, the two languages that
helped us get our data.
| 
