[Spelling Progress Bulletin, Fall 1979, pp11-16]
[Note: this page is set to display International Phonetic Alphabet characters. If your browser over-rides this setting, by default or by choice, the IPA characters may not show.]
Katherine Betts: see Bulletins.

Language, Orthography, and the Schwa,

by Katherine P. Betts, Ph.D.*

*Winter Haven, Fl, U.S.A.
Presented at Nene College, Northampton, England, July 1979, International Conference of the Simplified Spelling Society.

This treasure, this symbolic storehouse of all man's recorded knowledge, this uniquely human behavior we call language does not readily yield insights and solutions to its mysteries. To those few whose artistry flows from their pens, we assign immortality. To those of us captivated by the study and analysis of language structure, we offer the inevitable debates arising from different disciplines and from various vantage points.

In the annals of language description and analysis, linguistics is a relatively recent discipline, as is its off-spring - psycholinguistics. Hence, lack of concensus and pluralism are appropriate and predictable. Contrastively, as we know, the battle for changing the vagaries and complexities of English spellings (orthography) has raged for centuries.

Despite their efforts, orthographers and alphabeteers have not changed, to any great degree, the way we spell the English language. (We have, in a limited way, used other alphabets, e.g., i.t.a., for beginning reading instruction.) But their greatest contribution, at this juncture, is the elevated status of orthographic study (e.g., included in college textbooks on reading instruction). That day is here, long overdue ultimate dream of orthographic practitioners has been cohesive, as illustrated by Dewey (1971, p. 6) a few years before his death:
A wholly simple phonemic spelling of English would have only one grapheme corresponding to each phoneme, and only one phoneme corresponding to each grapheme. Our currently accepted T.O. (traditional orthography) is deficient in that it has only 26 letters, 3 of which (c, q, x) are for all practical purposes duplicates to represent about 41 sounds - probably the optimum number for a phonemic notation for general use. Largely in consequence of this deficiency, it is also defective in having a multiplicity of spellings for the sounds and a multiplicity of pronunciations for the spellings. Several symbols for one sound are a major obstacle to writing (more particularly spelling); several sounds for one symbol are a major obstacle to reading. The impact of this confusion is the most obstructive single factor in elementary education - in effect, a roadblock to reading, which is not only itself the most important subject of elementary education but also the medium thru which much of the rest of elementary education is carried on.
But the problems of fruition have, for one, been captured by Wijk (in Haas, 1969, p. 58):
The problem of devising a suitable new system of orthography for English may perhaps at first seem to be a comparatively easy one; but anyone who endeavours to penetrate more deeply into the question will soon find that it is fraught with formidable difficulties. The mere fact that the numerous attempts which have been made to solve it, both by eminent individual scholars and by societies specially founded for the purpose, have all failed to produce an acceptable solution, is in itself a sufficient indication of the intricate nature of the problem.
Is the goal of one-to-one phoneme-grapheme correspondence for English orthography an impossible dream? Or has it been achieved? These and other questions will be explored in this discussion of the schwa /ə/ and its implications.

Admittedly, the schwa /ə/ phoneme is a miniscule element in the tapestry of the English language. Yet it serves as a classic example of problems plaguing any re-definition of English orthography and the teaching of reading/ writing skills.

The sound of a in about, or the schwa (a German modification of the Hebrew word sheva, originally meaning a diacritical marking of a vowel), entered the English language more than nine centuries ago, as explained by Scragg (1974, pp. 11-12):
As a whole, Old English spelling as developed in the West Saxon tradition was much newer a one-to-one relationship with sounds than its Modern English descendant ... The widespread use of a single stable spelling system for an extended period meant that the accuracy of phonemic representation was increasingly disturbed in the eleventh century, and spellings which had a one-to-one relationship with sounds gradually lost it as the phonemic pattern altered ... Vowels in unstressed syllables gradually fell together in /ə/ [schwa], so ... for example, the symbols a, e, o all represent the same unstressed vowel; eleventh century scribes frequently confused these graphemes (and also u) in inflectional endings and affixes.
Some pertinent points regarding stress and its effects on sounds have been summarized by Classen (1919, p.209):
It is found that all language sounds in stressed syllables have not the same [historical] development as sounds in unstressed syllables. This is only another way of saying that all sounds are combinatory, since their development is bound up with stress. But setting aside this point for the moment, it is clear that a sound which is stressed will have more resistance to changes of a certain kind than a sound which is not stressed, for it will not be slurred over in pronunciation, it will be more clearly pronounced and any divergence from the normal will be more noticeable than if it were unstressed. On the other hand, an unstressed sound tends to be slurred and shortened, especially in those languages in which the stress tends to fall on the syllable which really conveys the meaning, for in such languages unstressed syllables are less important from the point of view of significance. Hence there is usually in language less variety of sound and quantity in unstressed syllables than in stressed ones.

Definition.

The schwa /ə/, e.g., the sound of u in but, one of nine simple (or "short") vowels in the English language, is articulated in the mid-central, lax position in American English dialects, somewhat farther back in British English dialects. It is an unrounded vowel in the respect that the lips do not enter into its articulation. Phonetically, the schwa /ə/ (represented by o in abbot) is differentiated from its stressed allophone /ʌ/ (represented by the o in mother); phonemically, these variant sounds of schwa appear to be allophones of a single phoneme, a family of sounds in complementary distribution. Phonetically, the schwa-plus-r (e.g., er in mother) is considered one phoneme /ɚ/; phonemically (particularly in dictionary respellings), it has been interpreted as two phonemes /ər/. The schwa-plus-r, beyond the scope of this discussion, has been delineated in a companion paper by Emmett Albert Betts. "Implications of Spellings: "Graphic R".

In G & C Merriam's Webster's New Elementary Dictionary (1970, p. 26a), the phonemic definition of schwa is evident:
The sound represented by the symbol ə (called schwa) is one of the most common in the English language. When stressed this sound is spelled with the letter u in cut ... with oo as in blood, and with o as in son, done, and color. When unstressed this sound may be spelled with any of the vowel letters as in about /ə-baut/, silent /'sī-lənt/ maritime /'mer-ə-tīm/, collect /kə-lekt/, suppose /sə-'pōz/, and cylindrical /sə-'lin-dri-kəl/.

Phonemic Basis.

General Comments


The delineation of phonemes (a linguistic abstraction) and the differentiation between phonetics and phonemics (branches of phonology) appear crucial to this discussion. Hence, Gleason's (1961) interpretation:
The phoneme is the minimum feature of the expression system of a spoken language by which one thing that may be said is distinguished from any other thing that may be said. (p. 16)

A phoneme is a class of sounds which: (1) are phonetically similar and (2) show certain characteristic patterns of distribution in the language or dialect under consideration.

The simplest of the patterns of distribution is free variation. The human vocal apparatus operates with an incredibly high degree of precision, but it is still far from exact. If the word key [two phonemes is pronounced, even by a single speaker, a hundred or so times and all the measurable features of each /k/ are measured, it will be found that no two are exactly alike. They will, however, cluster about certain average characteristics ... Any two sounds (e.g., stressed and unstressed schwa) which are always in free variation cannot be two phonemes but only two points within the range that constitutes one phoneme. (pp. 261-267)
Linguistic pluralism regarding the classification of English phonemes was pinpointed by Wise in 1957; his comments, still valid today:
The specific phonemes of the English language have never been completely agreed upon; on the contrary, those who have thought most deeply and effectively on designating them still change their minds occasionally as to what the phonemes are and what they include ... two [definitions] have proved more useful than any others, viz., [Daniel] Jones' statement that a phoneme is a family of sounds, and Bloomfield's that a phoneme is a minimum unit of distinctive sound features. (pp. 74-75)
Pike's (1947, p. 57) metaphor clarifies the differing linguistic objectives of phonetics and of phonemics; as he points out:
Phonetics gathers the raw material. Phonemics cooks it. Practical phonetics provides a technique for describing sounds in terms of movements of the vocal apparatus, and for writing them in terms of articulatory formulas., i.e., as letters of a phonetic alphabet. Practical phonemics provides a technique for processing the rough phonetic data in order to discover the pertinent units and to symbolize them in an alphabet easy for the native to read. The purpose of practical phonemics, therefore, is to reduce a language to writing.

Phonemic Status of the Schwa.


That the phonemic status of the schwa /ə/ has been diversely interpreted in extant dictionaries and by eminent scholars cannot be denied. The schwa is the most frequent vowel sound in English discourse (i.e., speech); its phonemic status, clouded by several factors: (1) the use of two symbols (/ə/ for a in sofa, /ʌ/ for u in cut) in phonetic alphabets, (2) a plethora of symbols in dictionary respellings (e.g., Emmett Betts, 1973, p. 13, identified 12 dictionary pronunciation symbols used between 1944 and 1953 for the schwa sound), (3) diverse interpretations in orthographic studies (e.g., Venezky, 1970, versus Dewey, 1971), (4) the ambiguous relationship of the schwa and its stressed allophone to the phonetic schwa-plus-r /ɚ/ (e.g., broth(er)) and its stressed allophone /ɝ/ (e.g., b(ir)d), and (5) shifts of stress in speech utterances (e.g., ham and eggs versus ham 'n eggs).

If one accepts the premise that the schwa and its stressed allophone are separate phonemes, then logically one also accepts the notion of separate phonemes for: the unstressed (e.g., moth(er)) and stressed (e.g., b(ir)d) allophones of /ər/, 8 other simple vowels, as well as 27 additional vowel nuclei (not all of which appear in any one dialect) articulated as off-glides with one of three semi-vowels (h, w, y), plus two diphthongs (as in (ou)t, b(oi)l), and the on-glide /yü/ (as in c(u)te). All of these, beyond the scope of this discussion (see Gleason, 1961, or Trager Smith, 1957); all of these, only the vowels!

A few years ago, Emmett Betts (1973, p. 13), discussing the schwa as part of a comprehensive article on the phonemic basis of word perception, made this observation:
"It will be noted that the schwa /ə/ is used phonemically in both the 1956 and 1970 editions of Webster's New Elementary Dictionary. This phonemic approach simplifies the use of pronunciation symbols and, therefore, makes the pronunciations more accessible to both child and adult."
Further on, he provided a list of scholars who address the schwa phonemically and a list of scholars who have a phonetic orientation. An independent compilation made for this discussion is a bit longer and includes all the names on his list, with a change for Trager and Smith, who are now on the phonemic side of the fence. Also, Robert Hall, depending upon his objective, appears on both lists.

Thus the schwa /ə/ for designating both stressed (e.g., m(u)d) and unstressed (e.g., (a)bove) allophones is used by
John B. Carroll (Language and Thought), 1964,
W. Nelson Francis (English Language, 1965),
H. A. Gleason, Jr. (An Introduction to Descriptive Linguistics, 1961),
Robert A Hall, Jr. (Sound and Spelling in English, 1961),
Archibald A. Hill (Introduction to Linguistic Structures, 1958),
Chas. F. Hockett (A Course in Modern Linguistics, 1958),
Herbert Landar (Language and Culture, 1966),
Donald L. Lloyd and Harry Warfel (American English in its Cultural setting, 1963),
Kenneth L. Pike (Phonemics, 1947),
Clifford R. Prator, Jr. (Manual of American English Pronunciation, 1957),
Paul Roberts (Patterns of English, 1956),
Peter H. Salus (Linguistics, 1969),
Norman C. Stageberg (An Introductory English Grammar, 1965),
Richard L. Venezky (The Structure of English Orthography, 1970),
Henry R. Warfel (Language - A Science of Behavior, 1962).

Authors espousing the phonetic interpretation of the schwa /ə/ and, therefore, classifying the schwa and its stressed variant /ʌ/ as separate phonemes include:
Arthur J. Bronstein (The Pronunciation of American English, 1960),
Jon Eisonson and Paul H. Boase (Basic Speech, 1956),
Louis H. Gray (Foundations of Language, 1939), Robert A. Hall, Jr. (Introductory Linguistics, 1964),
Claude E. Kantner and Robert West (Phonetics, 1941),
John S. Kenyon (American Pronunciation, 1950),
Ralph R. Leutnegger (The Sounds of American English, 1963),
Albert H. Marckwardt (Introduction to the English Language, 1942),
Dorothy Mulgrave (Speech, 1954),
Thomas Pyles (The Origins and Development of the English Language, 1964),
Charles K. Thomas (Phonetics of American English, 1958),
Axel Wijk (Rules for the Pronunciation of the English Language, 1966).

One of these authors, Charles K. Thomas (1958, p. 58) pinpoints some problems with perceiving the schwa in unstressed syllables:

Many people remain completely unaware of the existence of /ə/, partly because of the variations in spelling conceal it, partly it is often interchangeable with unstressed /i/, and partly because of the natural tendency when an unstressed syllable is examined, to add stress to it, thereby changing its quality. The vowel /ə/ is, however, one of the most frequent in English, and its use is essential to good English pronunciation. Proper balance between emphatic and unemphatic material is as important in speech as is balance between singer and accompanist, or between foreground and background in a painting.

Thomas (1958, pp. 10-11) also concurs with Emmett Betts in reference to dictionary respellings of the schwa:
... With rare exceptions we are visually minded; we rely more on our eyes than on our ears. We feel uncertain about an unfamiliar word till we can visualize its spelling, however odd that spelling may be. Our dictionary makers encourage this visual tendency, else we should not have so many symbols in all but the most recent dictionaries for the unstressed vowel [schwa] common to (a)count, sod(a), sil(e)nt, Apr(i)l, c(o)nnect, and circ(u)s.
Arthur Bronstein (1960, pp. 180-181), another of these phoneticians, has addressed the complex phonemic status of the schwa, noting the absence of "a satisfactory and generally adopted conclusion" on either side:
The /ə/ vowel is the lax, central vowel that can occur in any position of a word ... It is probably best described as a sound made with the articulators in neutral position, with neither spread nor rounded lips, and with the tongue neither forward nor back ... It is variously called the schwa sound, the indeterminate, weak, obscure, or unstressed vowel ... the schwa sound may be spelled with any vowel [letter] ... The variations of the sound are dependent an the phonetic surroundings of the vowel. It is not an unstressed variety of other vowels, for any stressed vowel may also have an unstressed form ...

/ə/ is the vowel commonly found in the monosyllabic definite and indefinite articles, prepositions, conjunctions, pronouns, and helping verbs as well as many other words not so easily classified: a, an, the, but, or, for, from, of, her, them, shall, was, can, as are normally spoken with /ə/, unless stressed. Many other words possess this intermediate vowel, that cannot be assigned to any other phonemic entity ...

Because of the extensive unstressing of syllables in our language, /ə/ is our most commonly used vowel ...

It is probably not necessary to mention that many linguists do not agree with the conclusion that recognizes /ə/ as a separate phoneme ... The widely followed Trager-Smith system ... describes the unstressed and stressed vowels of above and under as /ə-bəv/ and /ənd-ər/, and there is strong phonemic justification for this on the basis of complementary distribution (i.e., /ə/ and /ʌ/ do not signal differences in meaning) ...
Some linguists, then, prefer using /ə/ as a separate phoneme in American English, recognizing /ə/ and /ʌ/ as belonging to /ə/.

Syllabic l, m, n.

In relation to syllabic l m, n, - as in coup(le), har(um), poll(en) - the equivocal status of the schwa /ə/ has yet to be resolved. Evidence of ambiguity in this area abounds, particularly obvious in dictionary respellings of word forms with one of these syllabic phonemes. Differences are found, not only across dictionaries but also in different editions of the same dictionary. A consistent rationale for the examples below has not been discovered:

Words G. & C. Merriam
Webster's
New Elem. Dict.
G. & C. Merriam
Webster's
Secondary Dict.
Random House
Collegiate
Dictionary
Thorndike-Barnhart
World Book Dict.
(2 vol.)
 1975197019591969 1979

people/'pē-pal/(same)/'pēp-l/ /pē'-pəl//pē'-pəl/
bottle/'bät-l/(same)(same) /bot'-əl//bot'-əl/
shuttle/'shət-l/(same)(same) /shut'-əl//shut'-əl/
hovel/'həv-əl/(same)/'həv-l/ /huv'-əl//huv'-əl/
counsel/'kaüns-əl/(same) /'kaün(t)s-l//koun'-səl//koun'-səl/
column/'käl-əm/(same)(same) /kol'-əm//kol'-əm/
bottom/'bät-əm/(same) /'bat-m//bot'-əm//bot'-əm/
custom/'kəst-əm//'kəs-təm/ /'kəst-m//kus'-təm//kus'-təm/
token/'to-kən/(same) /tok-n//to'-kən//to'-kən/
carton/'kärt-n/(same) (same)/kar'-tən//kar'-tən/
pigeon/'pij-ən/(same) /'pij-n//pij,-ən//pij'-ən/

One may note in these examples that syllabic l, m, n, have been respelled (often for the same word) in three ways (e.g., /-əl/, /-l/, /-ə/(The term same in the above chart refers to agreement with the respelling in the first column.) Incidentally, the syllabication of these respellings also varies (e.g., people, custom, carton).

Morphophonemic Alternations.

The phonemic status of the schwa /ə/ can also be observed in morphophonemic alternations (e.g., phonemic variations of morphemes). Of the several options in this complex category, the following examples illustrate the effects of shifts in syllable stress in pairs of identical word forms. Merely by shifting primary stress from the first to the second syllable in each pair of words, one may observe a shift in form class, in meaning, and in pronunciation (i.e., the formerly stressed vowel shifts to schwa /ə/). (The respelling symbols are from the 1975 edition, G. & C. Merriam's phonemically based Webster's New Elementary Dictionary.):

contract
convert
convict
content
entrance
moderate
annex
/'kän-trakt/
/'kän-vərt/
/'kän-vikt/
/'kän-tent/
/en-'trans/
/'mad-ə-,rāt/
/'an-eks/
(noun) vs.
(noun)
(noun)
(noun)
(verb)
(verb)
(noun)
contract
convert
convict
content
entrance
moderate
annex

/kən-'trakt/
/kən-'vərt/
/kən-'vikt/
/kən-'tent/
/'en-trəns/
/'mad'ə'rət/
/ə-'neks/
(verb)
(verb)
(verb)
(adjective, verb, noun)
(noun)
(adjec.)
(verb)

In each of the above, (and other) examples, the schwa /ə/ phoneme contrasts with each of the stressed vowel phonemes to signal differences in meaning. However, no example could be found in which the schwa /ə/ contrasted with its stressed allophone /ʌ/ to signal different meanings, an important criterion in determining separate phonemes.

Graphemic Basis.

General Comments.


As we know, alphabetic symbols - or graphemes - are the other side of the phonic coin. The degree to which graphemes represent phonemes has been assigned various terms: relationships, correspondence, or fit. As orthographic scholars and practitioners have so often pointed out: phoneme-grapheme relationships in the English language are notoriously complex, often irregular, sometimes inscrutable. However, they are not irrational; else none of us could have learned to read and write English.

In addition to the phonemic reference of graphemes, Gleason (1961, pp. 409-411), for example, illustrates their morphemic reference:
A writing system consists of a set of graphemes plus certain characteristic features of their use. Each grapheme may have one or more allographs ... the relationship of graphemes to allographs is similar to that between phonemes and allophones...

The most familiar type of grapheme is that with a phonemic reference ... The reference of a grapheme may be single-valued or multi-valued. These complexities are merely instances of the intricate fit which exists between the English writing system and English phonology...

A second type of grapheme has morphemic reference. This is the case with English &...

Another somewhat different instance of an English grapheme with morphemic reference is English 'in boys' [' refers to possession; morphemic -s refers to plural] ... Boys, boy's and boys' are phonemically identical, but are morphemically distinct.
Other queries regarding graphemic reference will be raised in relation to the schwa; for example:

1. How do relational versus marking functions of graphemes (see Venezky, 1970) enter into decisions regarding specific phoneme-grapheme relationships?

2. How does the "silent-letter" debate affect similar decisions?

3. How should graphemes within word forms be segmented to provide generalizable data regarding phoneme-grapheme relationships?

4. How important is frequency-of-occurence in reporting the spellings of sounds and the sounds of spellings?


Graphemic Status of the Schwa


In stressed syllables, the schwa /ə/ is spelled by:
1. u as in fun, rug, run, luck (the most frequent spelling)
2. au as in rough, country, trouble (highly frequent spelling)
3. a as in ton, other, mother (frequent spelling)
4. o-e as in come, love, done (infrequent spelling)
5. oo as in flood, blood (rare spelling)
6. oe as in does (rare spelling)
7. a as in tam-tam (rare spelling)

In unstressed syllables, spellings of the schwa /ə/ include:
1. a as in away, about, cereal (the most frequent spelling)
2. e as in wanted, pavement, taken (highly frequent spelling)
3. a as in pilot, carbon, atom (frequent spelling)
4. i as in habit, civil, devil (infrequent spelling)
5. u as in upon, column, hocus-pocus (infrequent spelling)
6. ai as in captain, fountain, mountain (infrequent spelling)
7. au as in glorious, famous (infrequent spelling)
8. y as in analysis, paralysis (rare spelling)
9. eo as in luncheon (rare spelling)
10. ai as in captain, fountain, mountain (infrequent spelling)
11. au as in restaurant (rare spelling)
12. io as in fashion, legion, region (infrequent spelling)
13. ei as in forfeit, counterfeit (rare spelling)
14. u-e as in capsule (rare spelling)

Therefore, it seems that three spellings (u, ou, o) of the schwa predominate in stressed syllables; all other spellings of this sound are infrequent or rare. Interestingly enough, three spellings (a, e, a) most frequently represent the schwa in unstressed syllables; all other spellings, again, are infrequent or rare. Although frequency of occurence is not unique to this discussion, it remains an important criterion in evaluating the spellings of sounds and the sounds of spellings.

Next, conclusions regarding the number and variety graphemes representing the schwa /ə/ vary, being highly dependent upon the approach taken. If, for example, the schwa and its stressed allophone are classified as separate phonemes in reporting spellings of this sound and if (as has often been done) the schwa-plus-r (excluded from this study) is separated and reported as spellings of schwa, then understandably the data regarding the spelling of this sound will be considerably different from study to study.

There are, also, other sources of variety in reporting the spellings of the schwa /ə/. For instance, one can be quite tidy, accounting for ambiguous vowel graphemes in word forms, and assign split digraphs (e.g., o-e in come) and split-digraph combinations (e.g., u-ue in brusque) to spellings of schwa /ə/. Or one can espouse the "silent-letter" syndrome and thus eliminate the e in come and the ue in brusque from any further consideration. (Actually, all letters are silent; their function, complexly symbolic!) Still another way to deal with these troublesome orthographic features is to describe the e in come and the ue in brusque as serving neither a marking function (i.e., signaling the sound of a previous vowel in the word) nor a relational function (ie., representing a phoneme) and to classify these graphemes in terms of graphotactics (i.e., serving only a spelling convention).

Other Orthographies.

Many have addressed their efforts to orthographic change. Some have published their rationale and proposals for change; others have not. Fortunately, a compilation of fifty-plus orthographic proposals (Emmett A. Betts, Editor, Orthographies; 1974) was completed during two years of Phonemic Spelling Council activities. This compilation has provided a rich resource for an analysis of the schwa in stressed and unstressed syllables: consistency in the treatment of the schwa and recommendations for spelling this phoneme. The purpose of this analysis is not to endorse any one proposal over another, but rather to draw some conclusions regarding the interpretation of the schwa.

Corpus.


All of the authors submitted their rationale, their proposed alphabet, and an approved transliteration of Lincoln's Gettysburg Address (177 words, including the title). Altogether, there were originally 54 orthographic samples in the 1974 edition (the first one, T.O., or traditional orthography). Of the 54, 5 were deleted: two at the request of the authors; three, shorthand systems requiring special training to read.

Procedure.


l. The T.O. sample was examined to determine frequency of occurence of the schwa /ə/ in stressed and unstressed syllables, including syllabic l, m, and n.

2. The proposed spellings for the schwa were examined in the rationale and in the sample.

3. The results were tabulated, classified, and converted to percentages without reference to specific authors.


Results.


1. In the T.O. sample, 62 (or 35%) of 177 words in the Gettysburg Address include the schwa /ə/ in their pronunciation.- 9 in stressed syllables and 53 in unstressed syllables.

a. In T.O. stressed syllables, the schwa /ə was spelled most frequently by u (n=5), less frequently by o-e (n=2), o (n=l), and a (n=1).

b. In T.O. unstressed syllables, the schwa was spelled most frequently by e (n=12) with 3 of the remaining e spellings also shifting to unstressed /i/ (e.g, r(e)maining, d(e)votion, r(e)solve. Less frequent spellings were o (n=9), a (n=7), and i (n=2).

2. The various recommended spellings of the schwa /ə/ in 48 orthographic proposals were tabulated and classified in the table which follows; the results reveal some interesting trends, as well as variability within individual proposals and across proposed systems of spelling:

a. An overwhelming majority (83%) of the proposed orthographies opted for the u spelling of schwa in stressed syllables. (Two of these added diacritical marks to u. One also used the same symbol u to represent /ü/ as in n(ew).) Note:- new is also pronounced /nyü/.

b. The remaining 17% recommended 6 different symbols for stressed schwa; two of these used the ə symbol for their spelling of the sound.

c. Almost half (or 40%) made no provision for the schwa in unstressed syllables, instead using their stressed representation of the vowel grapheme (as in contin(e)nt). However, about a third of this number did provide for syllabic l (as in peop(le)).

d. A few proposals (13%) used m for syllabic /m/, and 15% used n for syllabic /n/ (as in nati(o)n)).

e. Interestingly, 60% proposed symbols for unstressed schwa; of this number, two-thirds recommended u for spelling schwa in unstressed syllables.


The Schwa /ə/ in 48 Proposed Orthographies for English.

Stressed Syllables Unstressed Syllables
GraphemeNo. Spelling of "but"Grapheme No.Syllabic l,m,n
u38but *No provision
No provision
12N= 25
%= 52
  6
13
  7
15
u  1but except for /l/  7  
 N= 19(or 40%)
u  1butu 18(or 38%)
N=40(or 83%)
ə  2bətə   4
 (or 4%)
v  1bvta   2
e  1bet (apostrophe) '   1
.  1b.ti   3
  (alternate with ə)  
ʌ  1bʌte   1
 29 (or 60%)**
q  1bqt
o  1bot
   8(or 17%)

*No provision indicates that the schwa /ə/ in unstressed syllables was ignored, i.e., transliterated as the stressed representation of the vowel grapheme (e.g., the e in judgem(e)nt).

**12 of 29 (or 41%) of orthographic proposals having a symbol for unstressed schwa /ə/ were not consistent in the use of this spelling.

Implications for Reading and Spelling.

The title originally opted for this paper, "Implications of Spellings for the Schwa," was revised after a preliminary review of orthographic studies and relevant linguistic publications; the approach, deemed inadequate. As a result of the broader base which was developed and the diversity of positions therein, one can draw a variety of implications for reading and writing.

Of course, the one which has captured the motivations of almost all orthographic practitioners has, after all these years, been aptly summarized by Bloomfield (1933, p. 500):
Although our writing is alphabetic, it contains so many deviations from the alphabetic principle as to present a real problem, whose solution has been indefinitely postponed by our "educationalists' " treatises on methods of teaching children to read ... The primers and first reading books which embody these doctrines, present the graphic forms in a mere hodgepodge, with no rational progression. At one extreme, there is the metaphysical doctrine which sets out to connect the graphic symbols directly with "thoughts" or "ideas" - as though these symbols were correlated with objects and situations and not with speech sounds. At the other extreme, are the so-called "phonic" methods, which confuse learning to read and write with learning to speak, and set out to train the child in the production of sounds - an undertaking complicated by the crassest ignorance of elementary phonetics.
Of many implications for reading and writing, here are some major ones which merit consideration:

1. English orthography is multi-faceted, having phonemic, morphemics and grammatical bases. Therefore, one can simplify one facet (e.g., phoneme-grapheme relationships) and, at the same time, complicate another facet (e.g., morphophonemic change, as in s(ig)n versus s(ig)nal).

2. Frequency of occurence (not only of words but also of Phonemes and graphemes) appears to be a significant variable in analyzing the implications of sounds of spellings and spellings of sounds in orthographic studies and in developing materials for reading/writing instruction.

3. Lack of concensus regarding the phonemes of English and the functions of graphemes (elaborated in Graphemic Status of the Schwa) within word forms causes some difficulty in forming valid generalizations across orthographic studies. Ibis vacillating pluralism underscores a need for greater emphasis on the theoretical underpinnings of orthographical proposals and of reading/writing instructional materials.

4. Methodology in reading and writing instruction is allied, but does not have a linear correspondence with consistent phoneme-grapheme relationships. Many other factors, beyond the scope of this discussion, must be taken into account in developing, planning, and delivering effective reading and writing instruction.

In Conclusion.

This discussion has focused on a somewhat exhaustive examination of the schwa /ə/: its definition, phonemic basis, occurence in syllabic l, m, and n, morphophonemic alternations, variability in dictionary respellings, graphemic basis, and a mini-study of its interpretation in 48 proposed orthographics for English. However, the undercurrent of this discussion propels the schwa in terms of its broader implications for reading and writing. Thus the schwa has served as a classic example of several controversial facets of the English phonemic-graphemic system, as well as an example of the morphemic basis of English spellings. Also pondered upon - but briefly - have been the effects of syllable and phrase stress on English phoneme-grapheme relationships which shift in discourse (as they should). Furthermore, syllable and phrase stress combine with pitch and juncture to form the melody - or intonation - of language.

Consistency and simplicity of phoneme-grapheme relationships in the English language are viable objectives, worthy of pursuit, particularly for the beginner attempting the acquisition of reading and writing skills. That English spellings are notoriously complex in their representation of speech, is a valid premise. That several languages (e.g., Spanish, Greek) have a more nearly consistent phonemic representation in their writing systems must also be accepted. However, a one-to-one phoneme-grapheme correspondence is a goal in conflict with the melody of the English (or most any other) language and the many dialects it represents.

In any event, may our efforts be guided by empirical evidence in the classroom, by professional objectivity, and by the practical application of a sound theoretical structure. Whatever the state of our knowledge, we have achieved it, via the miracle of language, by standing on the shoulders of giants in their field, as they have stood on the shoulders of those before them*. It has been said that when our information is structured, it becomes knowledge and that, finally, the appropriate application of that knowledge is wisdom. To bring complex, worthwhile goals to fruition requires the collaborative efforts of many and, above all, a genuine concern and love for mankind.

References.

Betts, Emmett A. "Reading: Phonenic Basis of Word Perception," Spelling Progress Bulletin, XIII, no. 4, (Winter 1973), 10-16.

Betts, Emmett A. (Editor and Compiler). Orthographies, 1974. Coral Gables, Fla: Univ. of Maimi (Xeroxed), 1974.

Bloomfield, Leonard. Language. New York: Henry Molt & Co., 1933.

Bronstein, Arthur J. The Pronunciation of American English. An Introduction to Phonetics. New York: Appleton-Century-Crofts, Inc. 1960.

Classon, E. Outlines of the History of the English Language. New York: Greenwood Press Publishers, 1919.

Dewey, Godfrey. English Spelling: Roadblock to Reading. New York: Teachers College Press, 1971.

- Relative Frequency of English Spellings. New York: Teachers College Press, 1970.

Gleason H. A. Jr. An Introduction to Descriptive Linguistics. New York: Holt, Rinehart & Winston, 1961.

Haas, Wm. (Editor). Alphabets for English. Mont Follick Series, Vol. 1. Manchester, Eng.: Manchester University Press, 1966.

Kantner, Claude E., and West, Robert. Phonetics. New York: Harper & Bros., 1960.

Pike, Kenneth L. Phonemics. Ann Arbor: Univ. of Mich. Press, 1947.

Prator, Clifford H. Jr. Manual of American English Pronunciation. New York: Holt, Rinehart & Winston, 1957.

Ripman, Walter, and Archer, William. New Spelling. London: Sir Isaac Pitman & Sons, Pitman House, 1948.

Scragg, D.G. A History of English Spelling. Mont Follick Series, Vol III. Manchester, Eng.: Univ. of Manchester Press, 1974.

Thomas, Charles Kenneth. An Introduction to the Phonetics of American English. New York: The Ronald Press, 1958.

Venezky, Richard L. The Structure of English Orthography. The Hague, Netherlands: Mouton, 1970.

Wijk, Axel. Rules of Pronunciation for the English Language. London: Oxford Univ. Press, 1966.

Wise, Claude Merton. Introduction to Phonetics. Englewood Cliffs, N. J.: Prentice Hall, 1957.


Back to the top.
See Appendix. List of 54 Orthographies: 1974.