This paper was first presented at the Society for Caribbean Linguistics Conference in Georgetown, Guyana in 1995. It was presented again later that year at the Mid-America Linguistics Conference at the University of Kansas, and published in Proceedings of the 1995 Mid-America Linguistics Conference. This paper explains the early stages of the process by which a writing system is being standardized for Belize Kriol.

Orthography Development for Belize Creole

Ken Decker
Summer Institute of Linguistics

1.0 Introduction to Belize

The young nation of Belize is located at the intersection of Central America with the Caribbean. It is the only country in Central America with English as its official language, but there is a growing Spanish-speaking population. Belize relates politically and socially with both the nations of Central America and the Caribbean region. It has a 1991 estimated population of 194,300 people, of which approximately 30 percent, identify themselves as Creoles (Barry 1992:67, 165). The English-lexicon Creole of Belize is the most widely used second language of the country.

2.0 Attitudes of Creoles towards their Creole speech

In most cases around the Caribbean, historically, Creole languages have not been accepted as legitimate languages. They have been called "bad," "broken," or "bastard" forms of more dominant languages. There has been considerable research into language attitudes in Creole communities. (for example see: Ferguson 1959, Stewart 1962, and Devonish 1986.) There has been reluctance to consider development of Creole speech varieties because of their perceived limitations and the possible hindrance it may produce to the people's development in the more prestigious languages. However, as LePage (1980:341-2) pointed out, Creole in one social situation may be stigmatized while in another situation it is cultivated for identification.

In recent years the observation has been made that attitudes are changing to be more positive towards the recognition of the Creole vernaculars. Major language development is progressing in numerous Caribbean nations where Creole languages are spoken, such as in Netherlands Antilles, St. Lucia, Haiti, Sierra Leone, Australia, and Belize.

2.1 Reports of negative attitudes towards Kriol in Belize

Negative attitudes toward the use and development of Belize Kriol (BK) have been summarized by Cooper (n.d.) As shown in the following list of objections :

1. Creole is stigmatized; with the development of orthographies and their use, this attitude has begun to shift.

2. Creole has no standard phonology, morphology, or syntax; there are too many varieties.

3. Creole has no standardized orthography.

4. There is no body of Creole literature to draw upon for literacy.

5. Creole literacy would cut off its users from the rest of the world.

Lopez (1991:15-6), in his analysis of the merits of making Belize Kriol the national language in Belize, refutes all of these arguments against the recognition of BK as a language, except the fact that there is no standardized orthography. He points out that the Ministry of Education did discuss the possibility of the development of an orthography in the 1980's, but that the idea was discarded due to the expectation that the financial cost of the development of teaching materials and textbooks would be prohibitive.

2.2 Reports of positive attitudes toward Kriol in Belize

There is much evidence of positive attitudes towards BK use and development in Belize. Most people will acknowledge that BK is used in many social situations everyday. Cooper (n.d.:9) concludes, from a 1986 study, that the usual language of office communication between Belizeans is in Creole, not English. He states that BK is often spoken in school classrooms, and he reported that some of his non-Kriol, English-speaking informants felt alienated from Creole friends when they use English.

There are television and radio programs in which deep mesolectal, if not basilectal, BK is spoken and promoted. Kriol phrases are used in newspaper articles, promotional posters, on t-shirts, and billboards. Musicians are marketing tapes with Kriol songs. Plays have been written and performed in Kriol. There have been numerous poems and stories published in BK. Several of the most noteworthy are: a BK poem titled "Tode and Billy" which was published as early as 1935 (Elliot); a book about BK, including a small glossary, was published by George McKesey in 1974; and a book of BK proverbs was first published by Dr. Colville Young in 1980.

Over the course of recent years, there have been debates in journals, newspaper editorial columns, and other public forums concerning the development and use of Kriol, and whether BK is a 'real' language or a dialect. While there are those who wish Kriol would go away, there are others who feel quite passionate about the validity of the use and development of BK.

2.3 Research regarding Belize Kriol

There has been some linguistic analysis of BK. Most notably Young (1973), Escure (1978, 1981, 1991), and Hellinger (1973). In the area of sociolinguistics, the work of LePage et al (1974) is quite well known. Other noteworthy sociolinguistic research not cited elsewhere in this report are Escure (1982) and Kenan et al (1977). There is much research that is yet to be done.

Marlis Hellinger wrote several articles concerning the literary future of BK. In her article, "The future of Belizean Creole" (1974:14), she discusses a number of factors favoring the development of Kriol:

1. Creole is as strong as ever in all...functions of the community.

2. Vital cultural activities...are still carried on in Creole.

3. A number of Belizean writers have started to use Creole as a literary language....

4. First steps have been taken in the educational field to give Belizean Creole its proper place in the school curriculum.

5. Belizean Creole has gained scientific recognition by local and foreign linguists.

Twenty-five years later these factors are still true. These local developments, as well as the development of other Caribbean Creoles, creates an environment that is ripe for development.

3.0 Various methods proposed for writing Creoles

Literary development in Creole languages has been breaking out all over the Caribbean. Significant development has occurred in Haitian and St. Lucian French-based Creoles, Papiamentu of the Netherlands Antilles, and the Islander Creole English of San Andres and Providence islands. More tentative beginnings have been made in Grenada, Guyana, Trinidad and Tobago, and Jamaican Patwa. The Creole community of Trinidad and Tobago has surely been given an encouragement through the excellent work of Lise Winer (1990).

3.1 The Phonemic Model

In Winer's 1990 proposal of orthographic standardization for Trinidad and Tobago Creole English (TC), she presents three models. First is what she calls the Phonemic Model. This model is similar to systems proposed by Devonish (1986) and Cassidy (1978) for other Caribbean Creole English varieties.

The Phonemic Model adapts a basic linguistic phonetic system for the Creole language, thus creating a "one symbol to one sound" system. This type of system is felt to enhance initial literacy, and appears quite different from the orthography of the lexical source language. The Phonemic Model has the disadvantage that it must be standardized for one specific dialectal pronunciation and may lose historical and morphological relationships. Possibly the most important point cited by Winer is that a Phonemic Model would have low social acceptability. Berry (1970) claims that "an alphabet is successful in so far and only in so far as it is scientifically and socially acceptable." Hellinger (1974:26) proposes that social unacceptability must be the reason why Cassidy's orthography for Jamaican Creole has not been accepted. She relates that, "in the introduction to Sibley's delightful book [Quashie's reflections in Jamaican Creole] McLaughlin explains that Cassidy's phonemic system, which 'may subsequently become standard orthography for the Creole' (Sibley 1968:xii), has been altered in the direction of the English model 'for the sake of intelligibility.'"

3.2 Difficulty with phonemic alphabets

Linguists and literacy specialists of the Summer Institute of Linguistics (SIL), in their work with developing languages around the world, have identified some problems with phonemic alphabets. Phonemic alphabets are generally praised for their ease of learning by the beginning reader. However, Grimes and Gordon (cited in Dawson, 1989:1) point out that the best alphabet for a fluent reader is different from the needs of the beginning reader. Phonemic alphabets have been shown to be easy to learn for the new reader, who reads by decoding sound by sound. The fluent reader reads by identifying words and phrases as whole chunks of information. Therefore, the fluent reader needs words to maintain recognized shapes regardless of the changes in pronunciation due to morphophonemic variations of a word.

3.3 The Historical-Etymological Model

Winer next discusses the Historical-Etymological Model. In this model, words are spelled as they have been spelled in the historical form in the language from which the words have come. A phonemic representation would be used for: new words, words with no historical precedence, and words with unknown etymologies. While being highly acceptable and accessible to those favoring the lexical source language, it preserves all the orthographic inconsistencies of the historic word forms and adds new forms in the phonemic representations. It also maintains the appearance that the Creole is subordinate to the lexical source language.

A method similar to the Historical-Etymological Model was proposed for BK by Richard Hadel (1974). I have tested texts written like this in BK with Belizean Creoles, and as Hadel predicts, people can read it quite easily, but the informants nearly all said, "but it's not Kriol!" Having some appearance of difference from the lexical source language is important.

3.4 The Modified English Model

The third model described by Winer is called the Modified English Model. This model is described as retaining the spelling for words shared by both the lexical source language and the Creole, only salient features would be changed. For example, English 'through', following BK pronunciation would be spelled 'chrough'. In this example, only the first phoneme in BK pronunciation is different from English pronunciation. Words that have an established spelling would retain that spelling, and other words would receive a phonemic spelling. This model has the advantage of being more accessible to those already literate in English, but, as with the Etymological-Historical Model, it maintains all the inconsistencies of English spelling as well as adding new variations found in the phonemic spellings.

Winer reports that this model is closest to what has been used in TC literary works. This is true for Belize also. However, writers have usually been unsystematic in their spellings. In BK texts, I have often found the same word spelled different ways in the same text, or even the same sentence. This will tend to discourage the reader and make the language look like an inferior system compared to the lexical source language.

4.0 Principles for orthography design in English-based Creoles

Tom Crowell, SIL's former International Literacy Coordinator, suggests that in most situations "alphabet symbols are chosen for thirty percent linguistic reasons and seventy percent because of non-lingistic motivations." (cited in Henne, 1991:12) therefore, anyone who would endeavor to design an orthography should follow principles that have already been discovered. William Smalley (cited in Henne, 1991:12) identified five factors that are important in guiding orthography design. They are:

1. Maximum motivation for the learner and acceptance by his society and controlling groups such as the government.

2. Maximum representation of speech.

3. Maximum ease of learning.

4. Maximum transfer.

5. Maximum ease of reproduction.

Winer (1990:252-3) lists seven principles to be considered in the standardization of an orthography for TC. These principles are pertinent to the Belizean situation also.

1. Practicality. The English (Roman) alphabet should be used without new characters, without diacritics or accent marks.

2. Consistency. In a phonemic system, each letter or letter combination signals only one distinctive sound.

3. Pronunciation-based spellings. Spellings based on pronunciation rather than on presumed etymologies...should be preferred.

4. Historical precedent. Where well-established spellings are familiar and accepted, they will be maintained....

5. Pedagogical support. Given the opportunity and requirements for writing in tc in schools, spelling should support literacy in tc as a first language, and also in standard English as a second....

6. Readability. Readability should be maximized, primarily for tc speakers, and secondarily for English speakers...

7. Linguistic independence. TC (and BK) should be perceived as a legitimate language, different from, as well as similar to, English.

There is quite a bit of overlap in Smalley and Winer's lists. "Compromise" could easily be added as another principle required in the design of an orthography. In a Creole continuum, you cannot have complete linguistic independence from the lexical source language and conformity with it.

Smalley's "maximum transfer" and Winer's "pedagogical support" both relate to the aspect of conformity to the national language. It has been found that people in minority language communities often, eventually, want the writing system for their language to conform to perceived norms of the national language. Dawson (1989:9) warns alphabet makers: "do not underestimate the desire of a community to conform to the national language. There are cases where an orthography has been used acceptably for twenty years, but when readers become more and more acquainted with their national language through the school system those features of their own orthography which did not conform to the national language are rejected."

Although there are many different attitudes held by different people to the type of orthography they need, and many sociolinguistic factors to be considered in the design of an orthography, the orthography that conforms to the national language will probably receive the greatest acceptance (Henne, 1991). Gralow (1981:10) points out that people in minority language communities will eventually need to transfer their reading skills to the national language. If the writing systems for the two languages are similar, this can only facilitate learning to read the second language. In a Creole continuum situation, I believe that spelling conformity should be given greater weighting, due to the lexical similarity between the basilect and acrolectal forms.

One difference of note between the two lists is that Smalley includes "acceptability" as a factor, whereas Winer discusses acceptability as an end to the principles. Either way, the acceptability of an orthography to its users is of utmost importance.

4.1 Application of principles to the Belizean context

In Belize, among those active in promoting Kriol, there is a great desire that written BK appear different from English to establish that it is a different language. However, this desire for the appearance of the linguistic independence of Kriol may eventually conflict with the acceptability of the orthography to the majority of Creoles. When the linguistic independence becomes established in the minds of Belizeans, would an orthography that is radically different from English continue to be the best system for the other needs of the language? In fact, it is possible that even at this time, an orthography having maximum appearance of being different from English, may not be acceptable to most Belizeans.

An orthography which diverges from the 'one symbol - one sound' principle may prove to be the best system. In Winer's extended description of the consistency principle, she notes exceptions to the principle, for example: proper nouns, to avoid cumbersome spellings of common words, to distinguish homophones, and for unassimilated loanwords.

In Belize, there is a high degree of awareness of English. Many English word forms, or spelling norms, are already considered as acceptable forms even for those words in Kriol. One feature of BK is the phonological similarity of many words. For example, BK /fe:t/ is said for both 'faith' and 'fate'. Although these words are pronounced the same it would be quite important to have different ways to spell them to reduce ambiguity. There would also be conflict with consistency when considering words which have spellings established by historical precedence, ie. the small, yellow fruit called /krabu:/ has an established spelling of 'oo' for the /u:/ sound. The consistency offered by a phonemic system may also be in conflict with the readability of a text for the experienced reader. As discussed in section 3.2 above, pronunciation-based spellings are helpful for the beginning reader, but may not help the fluent reader.

5.0 The Rule-based Phonemic Model

The first Belize Creole orthography workshop was held June 16 and 17, 1994 in Belize City. Twenty-five of the most active and influential people involved in BK development were gathered to begin the process of standardizing a writing system. The participants were first given a short introduction to phonetics, so that we could talk in more linguistic terms. Next a list of principles, similar to Winer's list of principles, were discussed to help focus the attention of the participants on the overall guidelines necessary for this standardization process. Then four models were presented. The first three models were similar to the three models described by Winer. The fourth model is a compromise between the Modified and Phonemic Models.

I call the fourth model a Rule-based Phonemic Model. The key idea of the Rule-based Phonemic Model is to maintain the more common spelling conventions of English. This is done by choosing the most representative ways that sounds are symbolized in English. Rather than having a 'one symbol - one sound' system, we accept that there will have to be more than one way to represent some sounds, specifically vowel sounds for reasons to be outlined below. Consonants maintain the 'one symbol - one sound' correspondence. In principle, the two or three different ways of spelling each vowel are chosen from the way those sounds are written most commonly in English.

The major argument against leaving the 'one sound - one symbol' dictum, is that the system will be more difficult to learn. However, I believe that a phonemic system that associates unfamiliar symbols with the sounds of a language may prove difficult to learn also. Wiesemann (1989:19) points out that most any orthographic system, even ones that are not very systematic, can be mastered, given enough time. "however, the easier the writing system is, that is, the closer it is to the sound perception of the native speaker, the more quickly it can be learned and by more people." (italics mine). It is the sound perception of the speakers with which we are trying to coordinate. Which symbols do the people associate with which sounds? In a phonemic spelling system, the sound of /i/ would probably be represented by the letter 'i'. This would be fine in a country with Spanish as the national language because the /i/ sound in Spanish is written with an 'i'. In English, the /i/ sound can be written at least eleven ways. For example: baby (y), he (e), sea (ea), see (ee), machine (i), field (ie), key (ey), either (ei), people (eo), amoeba (oe), and in Belize 'caye' (aye). The letter 'i' used for the /i/ sound is not a very common spelling. In English, the more commonly recognized symbols for the /i/ sound are 'ee' and 'ea'. So, in BK we choose to write the long /i:/ sound with 'ee' or 'ea'. We reduce the number of ways to spell the /i:/ sound from eleven to two.

The spelling rules get a bit more complex. BK has a difference between long and short vowels, 'long' and 'short' in terms of the length of time the vowel is held. The participants elected to use the letter 'i' for the short /i/ at the end of some words. This diverges from English which rarely uses the letter 'i' at the end of a word. In the new BK system, the letter 'i' is also used for the / I/ (the high, front, open vowel) sound between consonants, as in big, fish, or ship. It is hoped that these spelling conventions, guided by rules, will be easier to learn than the present complex milieu of English spellings.

Another feature of Kriol not found in English is the heavy nasalization of some vowels. At the workshop it was decided that we would write 'hn' after the nasalized vowel. When we applied this convention to all the words in which it seemed relevant, it appears they are all morpheme final positions, for example: waahn 'want', frahn 'from', and sohnbady 'somebody'.

Certain spelling rules that are predominant in English are also used for Kriol spelling, such as the 'silent e' at the end of words which influences the vowel quality of the preceding vowel, such as: win and wine, bon 'bun' and bone, and fat and fate. Rules like this are then extended to other words that have "weird" English spellings; for example, this 'silent e' rule is applied to 'light' making 'lite', which the Kriol speaker recognizes due to the familiar rule, and feels that it is written as it is pronounced.

Applying these rules to some words can create unusual spellings that are not immediately recognizable. For example, the long 'o' sound and 'oa' spelling of 'boat' and 'oak' applied to 'ocean' creates the spelling 'oashan'. Dropping the 'silent h' in words like 'wheel, wheat, and white' produces 'weel, weat, and wite'. While these words initially look unusual, we hope that the spellings can be accepted as more representative of the BK pronunciation.

The participants at the workshop felt that they would prefer at this time to keep proper nouns spelled as they are in English or Spanish. As suggested by Winer for the other models, well established spellings for some words will be maintained, even if they do not follow the rules.

Applying this system maintains consistency while diverging from the 'one symbol - one sound' standard. It maintains conformity with the national language while creating an appearance that a Kriol text looks different from an English text. Even though there may be two possible ways to spell a certain word, there would be fewer possibilities than in English, and the amount of variation hasn't hindered most English speakers from mastering that system. It is a system which should assist the new learner who does not know English because it follows rules. The spelling rules learned by the new Kriol reader can be transferred to learning English. Testing has already shown that Belizean Creoles, already literate in English, have very little difficulty reading a text in this new orthography the first time they see it.

For a more thorough description of the system, with examples, see the Kriol Spelling Guide web page.

6.0 Plans for future development and testing

The Rule-based Phonemic system gained approval from the participants at the First Belize Creole Orthography Workshop. A committee was formed from the participants to the workshop to apply the agreed-upon rules to long lists of commonly used Kriol words and test their appearance, recognition and ease of readability. This resulted in the publication of 4000 words in the "Bileez Kriol Glassary an Spellin Gide" in 1997. The next step will be to test the ease of teaching the new system. It will be a long process of fine tuning the system but we are confident that we have made a good solid start towards the standardization of an efficient and effective orthography for Belize Kriol. A similar system has also been accepted by the Creole-speakers on the islands of San Andres and Providence north of the Colombian Caribbean coast.

I will also be interested in getting responses to this type of a system from speakers of other English-based Caribbean Creoles. The Rule-based Phonemic system seems to be more adaptable to forming a pan-Caribbean Creole spelling system because it is not as dialect specific as a phonemic system.

References

Barry, Tom. 1992. Inside Belize. Albuquerque, New Mexico: The Inter-hemispheric Education Resource Center.

Berry, Jack. 1970. The making of alphabets. Readings in the sociology of language, ed. By J. Fishman. The Hague: Moutan.

Carrington, Lawrence D. 1976. Determining language education policy in Caribbean sociolinguistc complexes. International Journal of the Sociology of Language 8:27-44.

Cassidy, F. G.. 1978. A revised phonemic orthography for Anglophone Caribbean Creoles. Paper presented at the conference of the Society for Caribbean Linguistics, Cave Hill, Barbados.

Cooper, Vincent O. n.d. Sociolinguistic change and political reality in Belize. Unpublished ms.

Dawson, Jean. 1989. Orthography decisions. Notes on Literacy 57.1-13. Dallas: Summer Institute of Linguistics.

Devonish, Hubert. 1986. Language and liberation: Creole language politics in the Caribbean. London: Karia Press.

Elliot, Rev. James A. C. 1935. "I am with you always:" and other poems - religious, romantic, and humerous. London: Arthur H. Stockwell, Limited.

Escure, Genevieve. 1978. Vocalic change in the Belizean English-Creole continuum and markedness theory. Berkeley Linguistic Society iv: 283-292. Berkeley: University of California.

____. 1981. Decreolization in a Creole continuum: Belize. Historicity and variation in Creole studies, ed. by A. Highfield, and A. Valdman, 27-39. Ann Arbor: Karoma Publishers, Inc.

____. 1982. Contrastive patterns of intragroup and intergroup interaction in the Creole continuum of Belize. Language in Society 11:239-264.

____. 1991. Serialization in Creole oral discourse. Development and structures of Creole languages, ed. by F. Byrne, and T. Huebner, 179-192. Amsterdam: John Benjamins Publishing Company.

Ferguson, C. 1959. Diglosia. Word 15:2.325-340.

Gralow, Francis. 1981. Some sociolinguistic considerations in orthography design. Notes on Literacy 33:8-14. Dallas: Summer Institute of Linguistics.

Hadel, Richard. 1974. Comments (on M. Hellinger, How to write Belizean Creole). National Studies 2:4.30-32.

Hellinger, Marlis. 1973. Aspects of Belizean Creole. Folia Linguistica, vol. 6:1/2.118-135. The Hague: Mouton.

____. 1974. The future of Belizean Creole. National Studies 2:3.11-15.

____. 1976. Creole as a literary language. Belizean Studies 4:6.19-31.

Henne, Marilyn. 1991. Orthographies, language planning and politics: reflections of an SIL literacy muse. Notes on Literacy 65:1-18. Dallas: Summer Institute of Linguistics.

Kenan, Keith; John Sodergnew; and Robert French. 1977. Speech and social prestige in the Belizean speech community. Sociocultural dimensions of language use, ed. by Ben G. Blount and Mary Sanches. 35-50. NY: Academic Press.

Le Page, R. A.. 1980. Theoretical aspects of sociolinguistic studies in Pidgin and Creole languages. Theoretical Orientations in Creole Studies, ed. by A.Valdman and A. Highfield, 331-367. NY: Academic Press.

____. Pauline Christie; Baudouin Jurdant; A. J. Weekes; and Andree Tabouret-Keller. 1974. Further report on the sociolinguistic survey of multilingual communities: survey of Cayo District, British Honduras. Language and Society 3:1-32.

Lopez, Andrew. 1991. Belize Creole as an official national language: an analysis. SPEAR Reports 7.6-24.

McKesey, George. 1974. The Belizean lingo. Belize: National Printers Limited.

Stewart, W. A.. 1962. Creole languages in the Caribbean. Study of the role of second languages in Asia, Africa and Latin America, ed. by W. A. Rice, 34-53. Washington, D.C.: Center for Applied Linguistics.

Wiesemann, Ursula. 1989. Orthography matters. Notes on Literacy 57:14-21. Dallas: Summer Institute of Linguistics.

Winer, Lise. 1990. Orthographic standardization for Trinidad and Tobago: linguistic and sociopolitical considerations in an English Creole community. Language Problems and Language Planning 14:3.237-268.

Young, Colville N. 1973. Belize Creole: a study of the creolized English spoken in the city of Belize, in its cultural and social setting. York, England: University of York dissertation.

_____. 1980. Creole proverbs of Belize. Belize City: Colville Norbert Young.

How Do We Write Kriol?