A Critique of Inventive Spelling

Dr. Patrick Groff, NRRF Board Member & Senior Advisor

Dr. Patrick Groff, Professor of Education Emeritus San Diego State University, has published over 325 books, monographs, and journal articles and is a nationally known expert in the field of reading instruction.

When students are puzzled, when writing, as to how to spell a word they have several options. For example, they can substitute a word for the one they cannot spell. Two, they can stop and seek help from classmates, the teacher, dictionaries, etc. Three, they can leave a blank space, continue writing, and fill in the word later.Four, they can spell the word phonetically, i.e., can apply their knowledge of speech sound-letter relationships to its spelling. Five, they can “invent” an idiosyncratic spelling for the word.

In schools in which phonics information is taught in a direct and systematic fashion, it is common for teachers to direct children to use the fourth option. This is effective advice since the experimental research indicates that it leads to the greatest progress in students’ spelling skill that is possible. When direct and systematic instruction of phonics information is provided children, they ordinarily also are required to spell a word correctly immediately after using phonics information to recognize (decode) it.

In this manner, the words that children learn to spell are ones they just have learned how to decode. The correct spelling of a word in this way also reinforces the child’s ability to read the word later on.

Since the advent of the “Whole Language” (WL) approach to literacy development, option five above has become the procedure of choice. The guiding principle of WL is that school children best learn to spell in the same informal, individualized, and natural manner they learned to speak at home, as preschoolers. Therefore, there is very little, if any, direct and systematic teaching of spelling in bona fide WL classrooms.

One of the suppositions of WL is that children learn to spell better by inventing eccentric spellings of words, than by being taught how to spell words correctly in a direct and systematic way. It is found, that if urged to do so, beginning spellers ordinarily will create spellings in the progressive stages: (1) letters are used randomly—candy is spelled SCZ; (2) some of the correct consonant letters are produced—bottom is spelled BT; (3) a vowel letter also is produced—bottom is spelled BODM; and (4) the total word is spelled phonetically—dress is spelled DRES.

It is assumed in WL that if children are allowed to invent the spellings of words they will learn to spell them correctly faster than otherwise is possible. Advocates of inventive spelling thus argue that if children’s “natural” progress through the above “developmental” stages of learning to spelling is interrupted by direct and systematic teaching of correct spellings, children’s abilities to learn to spell correctly will be handicapped.

It also is charged that direct and systematic spelling instruction will interfere negatively with children’s volume of writing, and their abilities to compose creatively. Invented spelling is touted as a far more humane and pressure-free way to develop children’s correct spelling, and to stimulate their written compositions.

However, the novel supposition made by WL about the supposed necessity of replacing direct and systematic spelling instruction with invented spelling has not been experimentally verified. The great preponderance of empirical evidence on children’s spelling development indicates that children learn to spell correctly faster if taught to do so in a direct and systematic way.

The experimental evidence also reports a positive relationship between the quality of children’s written compositions and their ability to spell words correctly. The creative written normally is more in control of the mechanics of writing, it is found, than is the unimaginative one.

It is thus significant that the California 1996-1997 state budget, page 528, Chapter 162, Section 24.03 stipulates that no state funds can be expended for methodology that “systematically uses or encourages inventive spelling techniques in the teaching of writing.” California parents who discover their children are taught to use invented spelling, instead of being taught to spell in a direct and systematic way, thus immediately should bring the above state budge item to the attention of their respective school boards.

As well, invented spelling also turns out to be an unwarranted demand on busy teachers’ time. In this regard, a number of books and articles recently argue that to be effective a spelling teacher also must infer correctly what children are thinking when they invent spellings. The authentic deductions that supposedly can be made in this regard are voluminous and highly complex. Nonetheless, to provide proper instruction to children teachers must master how to make these esoteric inferences, it is held. It typically is said that invented spelling is an irreplaceable diagnostic tool that provides a window on children’s developing knowledge of orthography.

In her book, Beginning to Spell, (1993) Rebecca Treiman warns in this respect that teachers who do not infer correctly what is “the logic behind children’s [invented spelling] errors” will instruct spelling in a way that “may be worse than no instruction at all.” Deriving what children putatively are thinking about when inventing spelling will “determine what help the child needs in order to spell better,” Treiman insists.

However, after teachers devote great time and effort to this end, Treiman offers nothing unique nor practical that they should do to “help” a child spell. For example, typical invented spellings are her as hr, and neck as nak. For the former, Treiman recommends that children “must memorize the E of her.” Nothing new here. Then, her advice that for the misspelling of neck “the child might be assured that /e/ does sound similar to /a/, but that /e/ usually is spelled with E,” is far too abstruse instruction for first-grade children to profit from. There appears to be a desperate but unsuccessful search for utility of invented spelling data.