This paper aims to examine children’s writings in Spanish to determine which sound units children represent in writing (to determine if there are syllabic writings), and to analize which letters children use to represent sound units. We also wanted to examine whether consonant letter names influence children’s spellings.

Two experiments were conducted. The main objective in experiment 1 was to examine the sound units children use to guide their spellings without having to center their attention on the quality of the letters themselves. Experiment 2 attempted to examine segmentation units as well, and to examine which letters children use in their spellings. Both experiments were concerned with the possible influence of letter names upon spellings. In both, children had to write 25 common nouns. In experiment 1, children wrote freely. In experiment 2, they were asked to use non-Latin mobile letters. Participants for both experiments were five and six year-old Mexican kindergartners. 

Results show that there were a considerable number of syllabic writings, and more than half the children produced syllabic spellings for more than 60% of words with two or more syllables. Monosyllabic words were often represented with two or three letters. Letter names seem to influence spellings in a marginal way. That is to say, children seem to produce more correct initial letters when they match the initial syllables. However, some syllabic spellings did not use any pertinent letters at all, which suggests children are not representing letter names, but the syllable itself.

During the last two decades, the predominant paradigm for children’s writing development in the Romance languages has been the work of Ferreiro and Teberosky (1982), at least in Latin-American countries. This theory states that children develop through a series of stages. At the pre-syllabic stage, children write strings of letters which are “read”, or interpreted, as a whole word or phrase. Children develop some criteria that will last through their overall development: the notion that words have to be represented by more than one letter; that contiguous representations of the same letter should not occur within a word, and that different words should have different representations. At this stage, children do not match each letter to a sound unit.

In the next stage, children start making systematic correspondences between letters and syllables. That is, children represent each syllable with one letter. At first, any letter represents any syllable. Later on, children start using pertinent letters. More frequently, syllables are represented first by the pertinent vowel letter. Thus, tetra-syllabic words are written with four letters, penta-syllabic words with five, and so forth. The exception is mono-syllabic words. Because children have the notion that words should have more than one letter, children often write short words with two or three letters, and adjust their readings accordingly. For instance, a CVC word may be represented with two letters, and children will often finger-read verbalizing two segments: one CV and another VC, thus converting the monosyllabic word into a bi-syllabic word that shares the same nucleus in both syllables.

During the third stage (syllabic-alphabetic) children represent some syllables with one letter, as in the previous stage, while others (usually CV syllables) are represented with more than one letter. Finally, in the alphabetic stage, children represent CV and most CVC syllables in a complete way, even if orthography is not conventional.

In recent papers, the existence of the syllabic writings, as described previously, has been questioned by several authors. For example, Pollo, Kessler and Treiman (2005) asked Brazilian and American children to spell bi-syllabic words, and found that only 10% of the spellings produced by Brazilian children had two letters. That is, they found no strong evidence for syllabic representations. Also, the authors find that Brazilian children’s spellings present more vowel letters than American children’s. The authors’ claim is that children are influenced by the letter names in their spellings, especially the vowels, which in Portuguese are highly transparent.

This paper aims to examine children’s writings in Spanish in order to examine which sound units children represent in writing (and thus to determine if there are syllabic writings), and to analize which letters children use to represent such sound units. Related to this, we wanted to examine whether consonant letter names influence children’s spellings.

Two experiments were conducted. The main objective in experiment 1 was to examine the sound units children use to guide their spellings without having to center their attention on the quality of the letters themselves. Experiment 2 attempted to examine segmentation units as well, and to examine which letters children use in their spellings. Both experiments were concerned with the possible influence of letter names upon spellings.

Participants for both experiments were five and six year-old Mexican kindergartners. Children belonged to working class families and attended public schools where they received no direct teaching of letter-sound correspondences or phonological awareness training.

In experiment 1, participants were 58 children. Children were asked to write “the best way you can” 25 common nouns (7 tetra-syllabic, 6 tri-syllabic, 6 bi-syllabic and 6 mono-syllabic). Some of the words started with a syllable that matched a letter name, while the others started with the same consonant sound, but the initial syllable did not match the letter name. During the interview, a visible page showed all the upper-case letters of the alphabet. Letters were not in the alphabetical order. If a child asked for a letter, she was encouraged to look at the paper and “choose the one you think best for writing X”. After writing all the words, the child was asked to finger-read each word. At this time, the child could make any corrections she might consider appropriate. No feedback was given by the experimenter at any time.

In experiment 2, children were asked to write the same words than in experiment 2, using non-Latin letters. Children chose the non-Latin, paper letters from a bowl, arranged the letters in order to represent the writing, and finally the experimenter would paste the letters in a piece of paper, respecting the child’s spelling. After writing all the words, the child would finger-print all the words. At this moment, he could add or delete letters. No feedback was given to the child.

Results show that there is a considerable number of syllabic writings, and more than half the children produced syllabic spellings for more than 60% of words with two or more syllables. Monosyllabic words were often represented with two or three letters. Letter names seem to influence spellings in a marginal way. That is to say, children seem to produce more correct initial letters when they match the initial syllables, and they use slightly fewer letters (after editing) when written nouns start with le letter name. Some syllabic spellings did not use any pertinent letters at all, which suggests children are not representing letter names, but the syllable itself.

Educational research and practice concerning beginning literacy has relied strongly upon Ferreiro and Teberosky’s model, in which syllabic writings are an important focus of attention. This study suggests that the syllable is indeed the guiding sound unit in pre-literate Spanish writers.