P. Coirier et C. Golder, WRITING ARGUMENTATIVE TEXT - A DEVELOPMENTAL-STUDY OF THE ACQUISITIONOF SUPPORTING STRUCTURES, European journal of psychology of education, 8(2), 1993, pp. 169-181
The development of argumentative writing is studied here by examining
the structural organization of argumentative texts It is assumed that
the ability to construct supporting relationships, that is, a conclusi
on statement supported by argument statements, is acquired gradually w
ith age. The following stages of acquisition are postulated: (1) a pre
argumentation stage, where at first no explicit position is stated, an
d then an explicit position is stated but is not supported by an argum
ent; (2) a minimal argumentation stage, where a position is explicitly
taken and supported by one argument; and (3) an elaborated argumentat
ion stage, where at least two unrelated supporting arguments are used,
and then two related arguments are used. Two corpora were gathered un
der similar conditions: collective debate in the classroom, followed b
y individual essays written on the chosen topic. The first corpus was
produced by 147 children aged 7 to 14, and a group of 34 college stude
nts. The second corpus included the protocols of 92 children aged 11 t
o 16 the essays were graded for the presence or absence of each struct
ural level, and then classified at the highest structural level exhibi
ted. The resulting classification largely confirmed the hypothesized o
rder for the stages of argumentative development. The minimal argument
ative structure (standpoint + one supporting argument) was mastered by
nearly 90% of the 7 and 8 year-olds. The most elaborate structural le
vel in our model (two related arguments) was attained later: less than
one out of four 7-8 year olds versus three out of four beyond age 14.
Techniques involving more complex argumentative relations such as ref
utations and counterarguments, or restriction of one argument by anoth
er, are mastered even later and seem to be strongly linked to the natu
re of the issues under debate. Three main conclusions can be drawn fro
m these results: precocious argumentative skills exist in children bef
ore age 11 or 12, argumentative discourse complexity continues to incr
ease up to age 14 and beyond, and the characteristics of the referenti
al domain of argumentation have an impact on this structural elaborati
on process.