LIN 4740 Linguistic Theory 1998

Assignment No 1 History and Philosophy of Linguistics

Teacher: Dr. Keith Allan
Student: Stephen Morey 11772026

Write 2,500 words on one of the following:
Discuss how the development of a writing system is evidence of linguistic analysis.

Due Date: 1/4/1998
 
 

1. Introduction

The development of writing systems appears to show linguistic analysis in several ways:

a) word classes

b) phonetic processes - consonants, vowels and tones, 

c) the relative importance of marking each of a) and b).

To illustrate this, section 2 will examine the marking of word classes, of consonant phonemes and of tones as shown in two writing systems - the Ancient Egyptian and the Thai, developed in the 3rd Millennium BC and in the 13th century AD respectively. 

I have also examined the writing systems devised by Shong Lue Yang and Matthew McDaniel, the process of whose work it has been possible to examine in more detail.

2.

2.1 Semantic Classes

Egyptian writing developed over a period of around 200 years (see Appendix 1). Eventually, a writing system using two main types of characters had developed: (Gardiner 1927:8)

Ideograms - which "signify either the actual object depicted... or else some closely connected notion."

Phonograms - which "are signs used for spelling, which, although originally ideograms and in many cases still employed elsewhere as such , have secondarily acquired sound values." 

Most signs could be used as determinatives (Gardiner 1927:22) a term which referred to an Ideogram which was used following one or more phonograms to end a word. See Table 1:
 

The determinatives in Table 1 are used to indicate different meanings in words which are otherwise homographs. Gardiner defines the determinative "wood" as a generic determinative, that is expressing "the kind of sense borne by these (1929:22). A full list can be found in Gardiner (1927:22f). The determinatives are markers of semantic word classes. Almost all words in Ancient Egyptian were written using such determinatives and where that is not the case an ideogram was usually used on its own. This indicates that the entire corpus of words had been analysed with regard to the semantic class of each item.

2.2 Consonants

In Ancient Egyptian, like the Semitic languages to which it was related, grammatical inflection was probably based on internal vowel variation (Ritner:74). Such a language has less need to clearly indicate vowels. Even today in Arabic and Modern Hebrew, the writing of vowels is an example of a pedography in Mountford's definition (1996:628), that is something which is used in the learning process but not in everyday native-speaker writing.

Daniels & Bright (1996:3) identify three types of systems where single phonemes are represented in script, the alphabet, where both consonants and vowels are on equal footing (as in Latin script), the abjad, where only consonants are written, and vowels, if written at all, are expressed as diacritics (as in Arabic script) and the abugida (as in Ethiopic script), where 

"basic characters denote consonants followed by a particular vowel, and in which diacritics denote the other vowels." (Daniels & Bright 1996:xxxix).

Ancient Egyptian very early developed an abjad, which is usually referred to as the "alphabet" (see Appendix 2). Nevertheless, preference was given to mixed writings which avoided the kind of ambiguity which would follow if the determinatives in Table 1 were omitted. (Ritner:75)

Biliteral (two consonants) and triliteral (three consonants) signs were developed, by a process which is illustrated in Table 2. In the first case, the ideogram for wood has developed phonetic form with the pronunciation /hÊt/. In the second case, it does not appear that the sign was originally used to mean "knee", but as a result of its being used as the determinative in the word for "knee", it developed the phonetic form /pd/.
 

The process of analysis appears to be:

i) Writing begins with the use of simple signs to indicate objects (Appendix 1 figure 1)

then simple events (Appendix 1, figure 3).

ii) Some signs are given sound values - similar to the sound of the thing they represent, or in some cases an arbitrary value. However only consonants are identified, and so the writing of sound signs only would probably lead to ambiguity.

iii) The simple ideograms are categorised and organised into the system of determinatives, which determine word classes. The combination of sound values and determinatives produces a writing system with few remaining ambiguities.

iv) Sometimes the determinative then takes on the phonetic form of a word the meaning of which it has determined.

It is worth observing that the "alphabetic" signs in Appendix 2 cover each one of the phonemes in Ancient Egyptian, and that there is only one sign per phoneme, although after some centuries, owing to historical change, some phonemes had fallen together (like /s/ and /z/), and some redundancy developed. This is somewhat related to the biuniqueness principle in modern phonology - each alphabetic sign was mapped onto one and one only phoneme.
 
 

2.3 Tones

The Thai writing system was based on that in use in the 12th century in the Khmer empire, which itself can be traced back to the Asokan Brahmi used in 3rd century BC India. Asokan Brahmi was an abugida - where the comparatively small number of vowels were marked by small diacritics. The language of India at that time consisted almost entirely of multi-syllabic words, and did not have tones.

Thai, on the other hand, is a language which is monosyllabic and tonal. The application of the Indic alphabet to a language of a completely different type has gone through at least two stages:

i) The development of the Old Khmer alphabet. 

Old Khmer is a polysyllabic language but has a wider number of vowel phonemes than Indic. New vowel shapes were developed which are more substantial in form than the Indic diacritics.

ii) The development of the Old Thai alphabet, which is usually ascribed to King Ramkhamhaeng of Sukothai in 1283 (Diller 1996:458), the main innovation of which was the introduction tone marking.

Gedney (1977) identified that for all languages and dialects of the Tai family, there are 20 groups of words which can be identified for tonal analysis. These are identified according to 4 types of initial consonants and 5 types of tones.
 
 

It was not necessary to provide a tone mark for long or short dead syllables - since those syllables will be obvious if the vowel length and final stop are marked. They are not marked. On the other hand it was necessary to mark the 3 tones on live syllables, and this is the main innovation of King Ramkhamhaeng's system. This is a different approach to modern linguistic method, such as the IPA system, which would usually marks all tones - or none.

With the exception of a very few irregular words, all the words in each of Gedney's boxes will have the same tone, and several boxes will have the same tone as each other. Therefore any writing system which identifies the box in which the word belongs, rather than simply indicating the tone quality, will survive changes in the tonal system. This is what the Thai system of King Ramkhamhaeng does.

In modern Thai, the originally voiced sounds have fallen together with the originally voiceless friction sounds. It is theorised that in the time of King Ramkhamhaeng, there were three tones, A, B and C for live syllables, and that there was later a split, so that today a word with an originally voiceless friction initial of the originally A tone has a different tone from a word with an originally voiced initial of same tone.

The tones of each box have changed, without the writing system needing to change. Today, although each of the unmarked A proto-tone, and the marked B and C proto-tones now have two or even three different tonemes, depending on the initial consonant, the writing system has not needed to change. This is because in effect the writing system tells us which box a word belongs in, and hence which tone, rather than characterising or describing the tone.

Could it be that the early Thai tones were not as regular as has been reconstructed, and that a writing system was worked out which would not need to change as tones change?

Even if this were not so, the Thai writing system shows that the inventors have fully analysed the citation tones of words, and it is, according to Diller (1996:458) "arguably the first time that phonemic tone was regularly indicated in a writing system intended for common use."
 
 

2.4 Some Modern Examples:
 
 

2.4.1 Shong Lue Yang

The development of the Pahawh Hmong, which was supposedly "revealed" to Shong Lue Yang, began in the late 1950s, is an example of the development of a script by a completely untrained person.

Hmong is a language with mostly monosyllabic words, and eight tones. Shong Lue Yang's system involves syllables following each other from left to right, but where for each word the vowel and tone mark (written together as a sign with a diacritic) are placed before the consonant. As Smalley, Vang & Yang (1990:60) comment: "In effect, the vowel is the nucleus, the heart of the written syllable, with the tone written as a diacritic on the vowel."

Consonantal and vowel signs are usually arbitrary and yet tone marks for newly written languages in South East Asia are often based on the IPA system, a kind of pictorial representation of tone shape. It is not clear that modern linguistic method has yet established what the distinguishing features of tones are. Pitch, contour, quality, glottalisation, breathiness and length are but a few of the factors, one or more of which might be the key distinguishing feature. Given this, more arbitrary signs, such as the diacritics developed by Shong Lue Yang, would seem more appropriate, because they do not mislead as to the important features of the tones. 

Development of the Script

The script underwent four stages, which are described as follows (Smalley, Vang & Vang 1990)

• The source version, in which each vowel/tone has a different sign, and each consonant has a different sign.
• The second stage, in which the tones are divided into two groups, with diacritics marking the tones. The 60 initial consonants are also grouped into three groups, with diacritics used to indicate change in consonants (see Appendix 3)
• The third stage, a further refinement.
• The final stage, in which there is only one sign for each vowel, and more diacritics are used. It is not used because "although more linguistically advanced, it is not as important culturally..." (Ratcliff 1996:620)

i) Identification of the sounds.

ii) Refinement to increase economy - the use of diacritics on the vowel to mark tones, and to set up the key vowel-tone axis as the nucleus of the syllable.

iii) Eventual move to a script where "each vowel quality is associated with one symbol, and each tone with one diacritic" (Ratcliff 1996:620). This is the point reached by the final stage, but due to the early death of Shong Lue Yang, the third stage is the one in use.
 
 

2.4.2 Matthew McDaniel

Matthew McDaniel is an American jewelry importer who has lived for several years in Mae Sai, Northern Thailand and has become involved with efforts to assist the Akha, one of the Hilltribes of Northern Thailand and Burma. McDaniel has no formal training in linguistics. When commencing the process of learning Akha, he attempted to make use of Lewis (1989), but found it baffling. He concluded that it would be necessary to use a different writing system altogether. He recorded how he learned the language in steps and then patterned his language system along those lines.

Using his own English dialect as the basis, he has built up from observation a new Akha alphabet which he claims is easier for Akha people to use, and he is now producing children's materials in it.

The key differences between the McDaniel and Lewis scripts are:

i) Lewis has several aspirated phonemes which have unaspirated allophones in certain environments. McDaniel has marked these as different consonants.

ii) McDaniel identifies several rather rare sounds which Lewis does not mention. Native speakers report that they intuit these sounds to be different from others.

iii) McDaniel opts for writing tones with unused letters of the English alphabet, making the writing of the script easier.

iv) Hmong tones have two aspects - laryngealisation and pitch. McDanielís system marks laryngealisation first, and pitch second, which probably represents the way native speakers think of the tones.

McDaniel said: (pers. comm.:1998)

"You could call it an environmental script ... it isn't artificial or academic. You will be able to spell quickly, you will be able to spell exactly as you hear and then as you become familiar with the tone distinctions you can add that on. But it requires no knowledge of rules about patterns. A non linguist would never know that when one word is said with a short tone it changes the pronunciation and the spelling. They would just assume that it was a different word with a different tone and a different spelling."

"The alphabet is considered unfinished since there are minor items that the Akha have pointed out as phonetic irregularities in the language that they don't agree as to their cause. There appear to be enough individual cases where the language has unique distinctions to keep it both interesting and the alphabet "unfinished".

McDaniel's experience raises the question of whether the modern linguistic approach is indeed the best for the production of workable writing systems, or whether non-specialist analysis actually leads to better systems. This may be due to the rather theory-dependent nature of academic endeavour. It is beyond the scope of this essay to do more than just put the question:

Is the analysis revealed in the development of writing systems more appropriate for their development than modern theory-laden linguistic analysis?
 
 

3. Conclusion

The Egyptians analysed their language to the point of identifying all the consonant phonemes, and many semantic word classes, and produced a fully integrated writing system. They knew that it was necessary to write words in a way which was not ambiguous - and used word class markers (the determinatives) to do it. They did not analyse the writing of vowels as important - any more than we require the writing of tones in modern English, though in both cases the meaning can be affected.

In Thai, where tone is more important, it is marked, and done so in a way which allows the language to change without the writing system having to change with it. Whether this has arisen as the result of careful analysis or whether it came about by accident cannot yet be established.

Shong Lue Yang showed that for Hmong the vowel-tone axis has primacy. That in itself is a significant piece of analysis, given that the opposite conclusion was reached by perhaps analogous processes in Ancient Egypt - and perhaps an indication that universalist attempts at language notation are likely to be grounded in the tradition from which they come and may not cover all languages equally well.

The information about the McDaniel script for Akha is included because it has been possible to ask a non-linguist developer of a writing system questions about it. The analysis is not from a theory-driven perspective, but it may prove to be more successful.

What is success? Successful writing systems might be defined as those which have lasted for a long time, and those that are in use. They might mark different allophones with different signs; they might not mark some aspect which linguistically seems essential - like tones or even vowels. And they might mark the tones in the Thai way - identifying the word group rather than the tone quality.

And they might contain all sorts of redundancies and irregularities and things which seem to be theoretically implausible, but still work, like the Latin script as adapted for English.
 
 

The development of writing systems is evidence for thorough examination of the languages concerned. In this sense it can be called linguistic analysis. What the development of writing systems does not seem to be, is the use of modern linguistic theoretical models. 
 
 

about 2600 words.

Bibliography

Berry, Jack. 1977. "'The Making of Alphabets' Revisited". Advances in the Creation and Revision of Writing Systems. ed. by Joshua A. Fishman. The Hague: Mouton

Daniels, Peter T. & W. Bright. 1996. The World's Writing Systems. New York: Oxford University Press

Davies, W.V. 1990. "Egyptian Hieroglyphs" Reading the Past - Ancient Writing from Cuneiform to the Alphabet. Int. by J.T. Hooker. Berkeley: University of California Press.

Diller, Anthony. 1996. "Thai and Lao Writing" The World's Writing Systems. ed. by P. T. Daniels & W. Bright. New York: Oxford University Press

Diringer, David. 1968. The Alphabet 2 vols. Oxford: Alden & Mowbray Ltd.

Emery, W.B. 1961. Archaic Egypt. Harmondsworth, Middlesex: Penguin Books

Gardiner, Sir Alan. 1927. Egyptian Grammar. London: Oxford University Press

Gedney, William J. 1972. "A Checklist for Determining Tones in Tai Dialects" Studies in Linguistics in honor of George L. Trager. ed. by M. Estelle Smith, pp423-37, The Hague: Mouton

Haas, Mary R. 1964. Thai-English Student's Dictionary. Stanford: Stanford University Press

Lewis, Paul. 1989. A% Ka% Daw% - Ga% La% Pyu$ Daw% - Tai% Daw% Di$ Sha% Na% Li%

(Akha-English-Thai Dictionary). Chiang Rai: Development & Agricultural Project for Akha.

Mountford, John. 1996. "A Functional Classification" The World's Writing Systems. ed. by P. T. Daniels & W. Bright. New York: Oxford University Press

Oates, Lynette. 1990. "Aboriginal Records of Aboriginal Languages" Language and History: essays in Honour of Luise A. Hercus. ed. by P. Austin, T. Dutton, R.M.W. Dixon and I. White, Pacific Linguistics C:116, Canberra: Australian National University

Petrie, Sir W.M. Flinders Kt. 1904 Hierakonopolis I. London: British School of Egyptian Archaeology.

Ratcliff, Martha. 1996. "The Pahawh Hmong Script" The World's Writing Systems. ed. by P. T. Daniels & W. Bright. New York: Oxford University Press

Ritner, Robert K. 1996. "Egyptian Writing." The World's Writing Systems. ed. by P. T. Daniels & W. Bright. New York: Oxford University Press

Royal Academy of Thailand. 1993. Dictionary of Literary Words of the Sukhothai Period - based on the Inscription of King Ramkhamhaeng the Great. Royal Academy of Thailand.

Smalley, William A., Chia Koua Vang & Gnia Yee Yang. 1990. Mother of Writing - The Origin and Development of the Hmong Messianic Script. Chicago: The University of Chicago Press

In addition, I had personal communication with Mathew McDaniel, which included a visit to Mae Sai, Thailand in February 1998 and several e-mail communications.

Much of what I have to say about the Thai writing system is based on personal knowledge which has been gathered over many years from a multiplicity of sources.

APPENDICES

Appendix 1: Some examples of the development of Egyptian Writing
 
 

Fig 1 A Predynastic Pot marking from Nagadeh (Petrie 1904)

Fig 2 A Protodynastic Seal from King Ka-Ap, found at Abydos (Petrie 1904)

Fig 3 The Palette of Narmer, (Emery 1961) 

Fig 4 The Ebony Label of King Hor Aha (Emery 1961)

Fig 5 Ebony Label from the tomb of Udimu (Den), Abydos (Emery 1961)
 
 

Figure 1 can be regarded as simply pictures - although it may tell a story.

Figure 2 contains the name of a King, surrounded by the ideogram for the kings name, and topped by a representation of the god Horus, in the same way as one of the royal names of all the Egyptian kings was for 3000 years after. It is one of the very earliest examples of a personal name in any script.

Figure 3 tells a story (presumed to be that of the unification of Egypt in about 3000 BC), and as well contains some names (that of the King in the top centre, and that of the chieftain he is presumably about to slay).

Figure 4 contains even more signs which are not simply pictures, but clearly writing. However it is not yet fully developed and some parts of the story are still told by pictures.

Figure 5 is an early example of what appears to be fully integrated writing. It also contains the royal name, variously read as Den or Udimu. The first reading would assume that the 

"alphabetic" signs were already fixed, and the second reading suggests that the signs still carry some association with the phonetic form of the words for the meanings of "hand" and 

"water". As far as I know this matter is still controversial.
 
 
 
 
 
 

Appendix 2: The Ancient Egyptian "Alphabet"
 

Appendix: 3

Hmong Writing System (from Smalley, Vang & Yang 1990)

Appendix 4 Akha Writing systems

Comparison of the McDaniel and Lewis Systems

Lewis (1989:9) identifies "seven consonants whose quality is determined by the quality of the following vowel. As a general rule, the consonant is aspirated when followed by an oral vowel, and unaspirated when followed by a laryngealized (or glottalized) vowel." The largyngealised vowels are marked with the tone marks ^ or & .