[Journal of the Simplified Spelling Society, 1993/2 pp8,9 later designated J15]
[See Journal and Newsletter articles by Patrick Groff.]

Some Empirical Data on the mat-mate System.

Patrick Groff.

Patrick Groff is Professor of Education Emeritus of San Diego State University, and will be reporting regularly in the JSSS on spelling matters in America.

In New Spelling 90 (Fennelly, 1991) considerable comment is given to what is described there as "the mat-mate problem" (p7). "English has evolved a complicated - and an inefficient way of dealing with it." Fennelly, et al. (1991, p6) observe. The mat-mate spelling pattern "in fact is at the very heart of the spelling problems of the English language," they conclude (p6).

The crux of this problem, New Spelling 90 advances, is the single vowel letter - consonant(s) - final <e> spelling of words (eg mate). The final e in such words signals that the speech sound to be given the first vowel letter is its "long" vowel sound. The final <e> is not sounded.

New Spelling 90 recommends that spellings such as mate be reformed so that the final <e> here is placed alongside the first vowel letter in these words. Thus, mate would be spelled maet. This decision is based on "a statistical analysis of current spelling," New Spelling 90 (p1) explains. In coming to this determination about the spelling of words that contain "long" vowel sounds, there seemingly was no consultation made, however, of the empirical evidence about learners' abilities to spell and read the words in question. Such information is in short supply. There apparently are only two readily available sources of these data.

The first is the New Iowa Spelling Scale (Greene, 1955). The NISS reports the percent of students at grades two through eight who could spell correctly 5507 commonly used words. The second source is Children's Knowledge of Words (Dale & Eichholz, 1960). It presents the percent of students at grades four, six, eight, ten, and twelve who could read and were familiar with 12302, 6130, 4460, 3431, and 2027 commonly used words, at these grade levels respectively. The NISS reports that 52 percent of fourth-grade students could spell correctly monosyllabic words that have the final e spelling pattern described above (eg, mate, theme, hike, nose, cute). The NISS also reports that 50 percent of fourth-graders could spell correctly monosyllabic words with a cluster of two contiguous vowel letters (rather than two separated vowel letters as is the case in the final <e> spelling pattern) (eg nail, seal, seed, tries, soap, pour, thrown). The second vowel letter in both these categories of words signals that the first vowel letter should be given its "long" speech sound. Certain unclassifiable words were omitted from both these calculations, and all others that were made (eg weave, seize, noise, choose).

When the fourth-grade students in the NISS spelled words in these two categories (eg, mate vs. nail) onto which the suffix ing had been added, they were more successful with words of the latter category (eg nail) than with the former (eg, mate). On the average, 52% of these students could spell correctly words such as raining. On the average, only 39 percent could spell words such as smiling. It is also pertinent to note that on the average 13% of the students who could spell words such as rain could not spell their inflected (ing) forms. This inability increased to 22% for words such as smile-smiling.

There was no significant difference between fourth-grade children's ability to read words in the two categories discussed here (eg, mate vs. nail). The CKW study reported that 77% of these students could read monosyllabic words in the final <e> category (eg, mate). Of these students, 76% could read words in the vowel letter cluster category (eg nail). The data from both NISS and the CKW studies appear to support the New Spelling 90 advocacy of the reform of spelling represented by the change of the spelling of mate to maet. In this respect, it is obvious that spelling final e words (eg mate) requires the learning of only a single rule regarding the initial vowel letter - final <e> connection. On the other hand, with vowel letter cluster words (eg, nail) the learner must remember that various letters are used in the secondary position of the clusters to signal that the initial letter should be given its "long" vowel sound. These secondary letters in the cluster can be <a> (as in seal and soap), <e> (as in seed, lies, and hoe), <i> (as in paid), <o> (as in door), <u> (as in four and soul), <w> (as in bowl), and <y> (as in days).

Despite suffering the disadvantages of being less predictably spelled, the vowel letter cluster words (eg nail) were spelled (by NISS students) and read (by CKW students) almost as well as were final <e> words (eg, mate). When spelling monosyllabic words inflected with ing, students had more success with vowel letter cluster words than ones with final e endings. These accomplishments become all the more remarkable when one discovers that in the NISS the ratio of monosyllabic predictably spelled words with final <e> endings to unpredictably spelled ones in this category was 3.2 to 1. For vowel letter cluster words (eg, rain vs. bear) this ratio was 1 to 1.13.

The explanation for the impressive ability of students to spell and read vowel letter cluster words as well or better than they can words that end in final <e>, despite the apparent unpredictable spellings of so many of those in the former category, is not altogether clear. We know at present more about the developmental stages of children's spelling ability than we do about why these stages evolve in the order that they are observed to follow. We know, for example, that children ordinarily learn to spell words such as coat and wheel before they do words such as home and knife (Barnes, 1992). Moreover, in attempting to spell home, for instance, children typically will write hoem before home (Barnes, 1992).

It is significant, at this point, to note that children's inventions of plausible spellings of words reflect the kind of thinking on this matter employed by the learned authors of New Spelling 90. This notably similar cogitation by both parties is to discover a tenable system for spelling words that is based on the knowledge about phoneme-grapheme correspondences that respectively is available to them. The common goal for the young and the mature here is to bring reason to bear on a practical problem. We must infer that young as well as older humans inherently are rational beings. This deduction explains to some extent why young children are eager to spell words linearly by segmenting them into a sequence of speech sounds, and by regularly symbolizing each sound with a letter or digraph that has a reasonable connection to it.

The agreements between New Spelling 90 and young children as to the plausible way to spell words has some immediately practical implications for the Simplified Spelling Society. It is becoming evident (Ehri, 1992) that it is wise to teach students to spell phonetically before they are expected to spell in the conventional ways. The best expert opinion available to schools as to what phonetic spellings should be taught comes from the SSS.

By acting as a source to schools' spelling programs in this manner, the SSS can enhance its prestige, help improve children's chances of mastering traditional orthography, and gain some favourable publicity. The movement toward the reform of spelling doubtless can be intensified more by the Society's active engagement in educational affairs than from an attitude by it of splendid isolation. When the time comes that society elects to accept spelling reform, the SSS wants to be on the school scene, ready and able to assist educators to implement this transformation.


Barnes W G W (1992) ''The developmental acquisition of silent letters in orthographic images' in S. Templeton & D. R. Bear (Eds.) Development of orthographic knowledge and the foundations of literacy Hillsdale NJ: Erlbaum.

Dale E, Eichholz G. (1960) Children's Knowledge of words Columbus OH: Bureau of Educational Research and Service, Ohio State University.

Ehri L C (1992) 'Review and commentary: Stages of spelling development' in S Templeton & D R Bear (Eds) Development of orthographic knowledge and the foundations of literacy, Hillsdale NJ: Erlbaum.

Fennelly L R (Ed.) (1991) New Spelling 90. Croydon, England: Simplified Spelling Society.

Greene H A. (1955) New Iowa spelling scale Iowa City IA: Bureau of Educational Research and Service, University of Iowa.

Back to the top.