Stephen van Vlack
Sookmyung Women`s University
Graduate School of TESOL
Teaching Reading
Week 4 - Answers
Gaur - Scripts and writing systems: A historical perspective
1. What are the basic stages in the development of writing?
It has been argued that writing has really only been invented only once in the history of humankind and that all writing systems have come in one shape or form either directly or indirectly from that original system. It is hard to say whether this is actually true or not for there are many writing systems in the world and at least one of them (Mayan) has developed with what is believed to be a distance from the others. There is also a plethora of mysterious and unrelated Filipino scripts about which little is known and which may have been developed (at least for some) in isolation. Taking writing in the Americas and the Philippines aside let's just, for arguments sake, assume that all other writing systems currently in existence, or that have ever been in existence, are based on writing as it was developed first in Egypt and Mesopotamia some 6000 years ago. Looking at this development and the development of writing systems or print writing systems in other parts of the world which might be unrelated, we can see certain parallels.
The development of writing seems to follow certain patterns. Writing systems generally start off as pictographs. A pictographic system of writing is writing in which concrete elements in the physical world are represented by pictures (drawings) of them. The pictures are drawn to show striking physical resemblances to the objects they represent. So, a pictograph of a plant, corn (maize) for example, would look like a stalk of corn leaving of course for stylistic variations. The individual pictographic elements are then ordered and arranged in some sort of sequence in physical space (syntax). Over time the pictographs, which represent concrete objects, become more stylized and abstract sand new symbols are adopted for abstract concepts. This is basically because of the writing instruments employed to draw or create the pictographs are limited in their ability and the idea of efficacy in relation to time. Over time systems are used to make the drawing faster and easier due to expanding uses. After the symbol has been abstracted and with a long period of use there is no longer a direct and simple correlation between the pictograph and the object it signifies. It is at this point when the stylization of the actual object is expanded and eventually enters the realm of the abstract that writing really begins and the symbol comes to mean more than just the concrete physical object it represents. It is at this point that writing really begins. Writing as a system (like language itself) is reliant on a strictly arbitrary relationship between the sign (symbol) and the thing it represents. The development of writing is really all about a systematic abstraction from reality and system building.
Once this abstraction has taken place there are a couple of different directions a writing system can take in its development. As I'm sure you have figured out by now the two directions that the writing system can go would either be to follow more logographic patterns or to follow more phonographic patterns of organization and systematization. If the writing system is to follow more logographic patterns then what will generally happen is that words with similar meanings will be temporarily fused onto the same symbol. If the writing system is to follow more phonographic patterns then words that the same sequence of sounds will be temporarily codified with the same symbol. As the needs for the writing system become more complicated and diverse the system needs to continually adapt to meet those needs and further abstraction and character (symbol) development is necessary.
Another thing which fuels changes in writing systems are the adaptions it needs to make. As we mentioned above, writing systems pass from one society to another with contact. These different societies often have different languages, often involving very different structural configurations (typologies). As the writing system is adapting to a new language, and possibly uses, the speakers of that language, in turn, need to adapt the writing system to fit their language better. This creates different types of writing systems, for example syllabic writing systems or consonantal systems which were the first ones to branch off from Egyptian hieroglyphs. The development of an alphabetic writing system came as a result of the importance of vowels and vowel alterations, typical of Indo-European languages, as a way of marking grammatical. In this way we can see that writing systems change and develop new things based on what they need to better fit in with the structure of a new language. But why all this effort? Certainly developing, spreading, and maintaining a writing system is a lot of work,, even in today`s societies. Thus, we need to think about the gains that are to be made from the use of this technology.
Writing was invented as a tool for people to be able to record meaning. It basically started as a way of recoding business transactions to keep people honest (goods bought and sold were marked down). Writing was also used for keeping inventory lists of storehouses. In this way we can see that writing was really developed as an accounting tool initially. Most of the early writing that people have found have been examples of lists, inventories from storehouses, as well as lists that traders would carry with them along with their goods. In the beginning writing was controlled by a very small number of scribes. As societies became more complex and large governmental organizations came into being the need for literacy and easier writing systems became greater, especially in the area of lawmaking. After its original conception as a tool for accounting purposes people soon began to see the power of writing and its uses expanded out to other areas including the recording of history, the solidification of laws, and for religious purposes.
For our purposes as language teachers it is very important that we realize that writing is a tool and has been invented for very specific purposes by humans. Interestingly, it is a tool which continues to develop as we develop new technology to aid in the literacy process, such a computers. Also, at its most basic level, writing is a tool which allows us to make our language visible and therefore our thoughts and perceptions permanent. There are many different reasons for wanting to do this, especially in the complex societies of today. Like most tools, however, writing will change and eventually be supplanted by better and more sophisticated tools. This kind of change is happening rapidly with the advent of computer technology. There is no doubt the computers will substantially alter the tool of writing and possibly render it, if not completely obsolete than definitely of much less value that it is today. In the highly literate society of Korea today it might be hard to imagine that our ancestors will be much more illiterate than us and still be considered highly intelligent and well-educated but that might very well be how it works out.
2. How can this knowledge be applied to teachers of literacy (reading)?
As teachers of language we cannot avoid our job as developers and heighteners of literacy. For the moment anyway literacy is extremely important, so we need to be able to know about how literacy develops and also about how writing developed. The brains of children, or language in general and the development of reading writing skill in children seem to go through the same kinds of stages in their development. Children move from simple associations based on words and meanings to more complex and regularized ones. Think of it this way, writing systems are systems created by human minds. Human minds are constrained by language to a large extent, particularly in this area because writing is a way of making language visible so writing and language really can't be separated easily or at all. Now, language is constrained by systems and patterns of the brain, which in turn is constrained or is modeled on systems in the natural world. Following this idea we see that all the systems have to basically work the same way. Looking at the surface it might seem like all these different systems are operating independently, but in the world basically there is one system which is really patterned over and over again for different purposes, but which are basically the same in their fundamental design. So it is with writing. By looking at the development of writing you really looking at the development of a system which is for the most part very similar to linguistic development. Literacy development, because it must follow the same kind of systematic patterns, should also be very similar to language. The world is a bunch of overlying systems which are more or less the same.
Rogers - Optimal orthographies
3. What are the basic prerequisites for a good writing system?
A good writing system must be efficient to learn and to use.
In thinking about what we would call a good or efficient writing system, there is one simple and abstract prerequisite which we can come up with. It is that the writing system must be easy both to learn and to use. This is how we seek to describe efficiency in this case, but it is not clear what the word really means, nor does it give strict guidelines for how we are to determine such ideas. We, therefore, need to look deeper into the question of what is efficient or optimal.
While doing research into his 1968 publication on phonology (co-authored with Morris Halle) The Sound Pattern of English the syntactician Noam Chomsky made not a new but, nonetheless, a startling discovery for the time. What was found was that words can be seen as being represented by two different levels; their underlying level and their surface level. Now, scholars of the English language had been for at least 200 years prior to Chomsky an Halle`s study been making a lot of complaints and strong comments about the weaknesses of the English writing system based on a strictly phonemic analysis. They were only trying to match spelling with the surface phonological structures of words and English does very badly here. What Chomsky discovered and started to talk about in the late 1960s and early 1970s was that English spelling is not based on surface representations, but rather the underlying representations of words. At this point you must remember that sometimes underlying representations and surface representations are the same form, and sometimes they are different. Following this, linguists came up with some new criteria for what they felt to be the description of an efficient writing system. As is his forte, Chomsky also went too far in his evaluation of the English writing system, claiming that it is optimal or perfect. This is, of course not true. It is also important to remember that the English writing system is NOT ALWAYS based on underlying representations. Here is what has come to be known the Chomskian perspective:
- The underlying level is the best level for orthographic representation.
This would seem to be true, of course bearing in mind that in some languages there are very close correspondences between underlying representations and surface representations in sound. In English there are often differences. Think, for example, about the word radiator in English. Different dialects of English pronounce this word in vastly different ways, but everybody is operating from the same underlying representation. All English speakers are using different phonological rules to come up with the different surface representations, but underneath its all the same and so is the writing. Differences in surface and underlying forms become heightened particularly when words undergo morphological operations.
photograph ---- photographer (Derivational morphology)
leaf ---- leaves (Inflectional morphology)
A problem with this is that underlying representations (URs) may not actually exist in the way Chomsky or what has become the field of generative phonology claim they exist. They are a theory with little experimental support. Likewise, the notion that linguistic units are stored as word units is also just a theory and one which had come under attack recently.
Here is another idea of optimality for writing systems
- Each morpheme should have only one representation without showing allophonic variation.
One example of where this actually works in English is the plural morpheme -s. The form of the morpheme in writing is always the same, yet there are three different allomorphs which emerge in the pronunciation. Think about the words courses, dogs, and cats. For each of these words the plural morpheme has a different form. They are allomorphs. And here the system works because they all have the same representation in the writing. This does not hold however for other morphemes, for example the prefix in-. This prefix has two different allomorphs /In/ and /Im/ as well as two different writing representations for them, so in this case the writing does not follow our prerequisite for efficiency because there are indeed different representations for allomorphs, which is a problem.
- Different morphemes should be represented differently.
To understand this would need to remember that a morpheme is a unit with meaning. Please do not confuse the structural notion of syllable with the semantic notion of morpheme although they are often one in the same. So, for example in an efficient writing system different morphemes like sun and son need to be encoded differently. In the above example they are, but this is not always the case in English with other morphemes. Look at the example of table (which you sit at) and table (which you read). They seem different, but look at the morpheme bag in these two sentences.
(1) Would you like a bag with that.
(2) I`m sorry but that`s just not my bag.
Somehow we maybe get the impression that both these renditions of bag are somehow related but we are not sure how or how to verbalize their relation, so are they the same or different morphemes? Since a morpheme is a semantic concept and semantics is a fuzzy issue at best, it is hard to quantify this argument.
We ourselves can make a few observations about all this. One observation is that despite its overt leanings in one direction or another a good system needs to be a hybrid in that it uses both sound (phonographic principles) and eye (logographic principles) to encode the language in writing. Another important idea that we need to acknowledge is that writing optimizes itself over time, or at lest it did before the development of print technology. That is, languages and the societies that speak them force writing systems to change, generally for the better (that is, there is usually a movement toward more efficient systems) over time. This is understandable since writing systems are generally (the American and Philippine systems mentioned above are possible exceptions here) imported cross-linguistically and cross-culturally. But the question we need to ask ourselves is does this occur naturally or does it need to be pushed by standardized reforms? English has NEVER had an official language-wide spelling reform. Do we need one? Has English passed beyond the brink of efficiency?
4. What are the variables that go into studying a writing system?
To analyze a writing system we need to look at the following variables
Synthesis
This is the amount of morphemes per word.
English has a tendency to be an isolating language where there is a low number of morphemes per word.
Fusion
This is about how easy it is to tell morpheme boundaries and begs the question do morphemes have a tendency to blend?
Usually it is quite easy to tell morpheme boundaries in English (at least in isolation), but there are exceptions like some irregular forms and archaic forms like went or took.
Variation
This is the amount of allophonic variation the language might have.
English has a fairly high level of allophonic variation, both at the phoneme and morpheme level.
Homonomy
This is about how many separate lexical units sound alike.
English renders about a mediocre score on this criterion, which means that there is a fair number of homophones, but this is clouded by the fact that as words are blended in regular English speech they have a tendency to sound more alike. (Think about the stress systems of English) So, in English being able to distinguish words in isolation or careful speech, which is what writing is supposed to encode, does not mean that they can be distinguished in chains of speech.
5. Based on this article, how optimal are the writing systems of English and Korean and how might this be expected to affect literacy development of English here in Korea?
Having looked briefly at some of the considerations for what we would determine to be an optimal type of writing system we find that English is probably not as horrible as we had originally anticipated having mostly looked at surface level forms. At the same time, however, we still find that English is far from optimal. There is still a wide range of irregularities in the writing system which really cannot be explained in any easy manner and would therefore make it quite difficult for learners to deal with the orthography and for experienced users to use the spelling system with ease and effect.
Venezky - How English is read: Grapheme-phoneme regularity and orthographic structure in word recognition
6.What are some of the variables in the word recognition process?
According to the article by Venezky, there are several different variables which play a role in the word recognition process. Among these variables are; spelling patterns, Grapheme-phoneme patterns, orthographic structure, statistical structure, and rule-governed structure. Let's just take a minute to try to go over what all of these might actually mean. When we talk about spelling patterns what we're really referring to are the patterns that are present in the language as reflected by the spelling. This is a general term which incorporates many of the others which are mentioned below it. For example, English words like hate, lake, and fate all represent a particular type of spelling pattern. This spelling pattern has to do with graphemic-phonemic regularity is well as different types of structures. In this way we can see the that spelling patterns are a general phenomenon in writing systems and need to be broken down into more specific areas.
The first of these areas is the area of grapheme-phoneme regularity. What we're really talking about here are letters and the sounds that they are supposed to represent. In a perfectly regular system one grapheme or letter would be used to represent one sound. Based on this, sounds are then easily predictable from the letters that represent them. Now, we know that grapheme-phoneme regularity in English is more of an illusion than anything else, especially when we are dealing with vowel sounds. Since that does not seem to be a high correlation in grapheme-phoneme regularity then many researchers to try to appraise the situation by trying to come up with approximations. That is, is the relationship between a grapheme and a phoneme invariant, variant-predictable, or variant-unpredictable? We need to try to figure this out within the context of the spelling system and patterns that are inherent in that system.
The second thing we need to look at is the area of orthographic structure. Orthographic structure is different from grapheme-phoneme mapping because it deals exclusively with the symbols (in the case of English letters) and is seen as being separate from sound. One of the areas of orthographic structure we need to look at is statistical redundancy. Statistical redundancy refers to how often particular letters, letter sequences, and even words occur in the writing of the language (orthography). For example, the letter x occurs in the front position of words in English only very rarely. Upon further analysis we find out that this is basically because x needs to follow a letter which represents a vowel. Part of being a good speller of the English language, for example, is knowing the statistical frequencies of certain letters in certain locations. This allows readers to make certain predictions when they read and they can thus read faster. Another thing that people look at when they're looking at statistical redundancy would be clusters of letters and how they occur. Reading letters is really not all that different than the general type of reading that we talked about in the very beginning of this course. In order to read faster people need to make advanced predictions about what is going to come. They do this by looking at the different letters which surround others. As is mentioned in this article it is not a strictly left to right process. Our eyes actually dart around the page as we read, hitting both the front and end of words before going to the middle. The other side of this orthographic structure issue is rule governed regularity, or the specific rules which the language has come up with in arranging the letters. Again this has nothing to do with sound, it really just has to do with the sequences of the letters themselves and the occasional rules which the language has deemed to apply to this. In looking at we find a simple example in the use of, for example, geminites or double letters as in the word `common`. The English language has rules which revolve around how geminites are used. Such phenomenon, as this is part of the rule governed regularity of the language, and such phenomena actually cause confusion because they are not related to sound. Under such considerations even what seem to be the simplest of terms are not so simple.
The notions of regular and irregular, for example, are extremely difficult notions for us to try to pin down as we are studying, or at least trying to study the writing systems of English. A major part of the problem is that we can never have the whole picture. To do so would simply require too much information. For example one of the reasons why some spellings in English seem to be very irregular is because the words which contain the spelling were borrowed from different languages at different times. English has always been a language that tries to preserve, and particularly through spelling, the respective origins of the words it has borrowed. The example Venezky cites in the article are the words chief and chef. They were both borrowed from French and even from the same word in French, but the borrowings occurred at different times and as a result there are different spellings as well as different pronunciations and meanings associated with each word. What first appears to be extremely irregular and problematic is actually based on some sort of regularity once we know about it, and the same kind of thing applies to this system is a whole. There is a rhyme and reason behind the system, and even behind the irregularity, but much of the time we do not know what it is. The question we need to ask ourselves is whether we really should need to know this as users of English and this forces us to ask whether English need to preserve this type of information in the spelling anymore. In this way, we can say that irregularity really is regularity which we simply don't understand fully.
7. How does the concept of phonological mediation work?
The basic idea behind phonological mediation is that we use phonological information, as stored in the lexical entry for different words in the mental lexicon, to help us figure out what word is on the page in front of us. The basic underlying idea is that along with the lexical entry for each word that we have there is also information about the phonological structure of that word. For example, along with the meaning and grammatical information and everything else we have about the word cat we also have information about how the word cat sounds stored in our brains. The idea behind phonological mediation is that we use this information to try to identify or match the word that were looking at and the word that is stored in our mind. Many researchers feel that this is the key to be able to figure out reading, yet others feel that is not particularly important. What clouds the issue somewhat is the belief that we also have to have an orthographic representation (or orthographic representations) in our mental lexicon. This would mean that in addition to the sounds that make up the word we would also have to have an orthographic representation stored along with the rest of the information in the lexical entries of words. The idea behind both of these kind of retrieval systems is that we go segment by segment matching what we see with what we have stored in our mind. It is basically an orthographic version of the cohort model, which is one of the more accepted theories in psycholinguistic research.
Scholes - Orthography, vision, phonemic awareness
8. What are the results of the Bruce (1964) study?
The basic results of the Bruce study conducted in 1964 reveal that literate users of English employee both phonemic as well as graphemic make strategies in trying to read words. What strategy they employ really depends on the word itself and the relationship between the word and its actual spelling, or lease the regularity in its spelling. Simply put, when the graphemes in a given word have close and regular phonological correspondences then the reader has a tendency to use more phonemic strategies in assessing, or locating the word itself. When the structure of the word graphemically is more opaque in relation to its phonological sound then the reader is going to use more graphemic based strategies for deciphering the word. Based on this we can see that English really does employ a hybrid system in relation to writing, as does every language. The difference here, however, between English and many other languages is this distinction between graphemic and phonemic is not border based. The two of them encroached entirely on each other's territory and not really with any patterning. This visiting the speaker when they're approaching reading don't know really what to expect and need to be able to employ both at a moment's notice. In this reading, therefore, requires the availability of using both systems in a rather irregular way depending on the words themselves.
9. What are the results of the study carried out on orthography and allophones?
The results of this study have drawn attention to a three way system of how literates deal with the written word which is referred to as the litera. In the litera triumvirate in addition to the phonemic (potestas) and the graphemic (figura), there is also the concept of the nomen. The nomen is the concept of the word itself as a whole. The study show that for people to be truly literate they need to be able to deal with att three of these concepts.
10. How do orthography and grammar correlate?
This is somewhat similar to what we did above although it is reappraised through Worfian ideas about language. Basically it claims that the way in which a language is structured will play a large role in the way that the speakers of that language will react to the writing system of that language. Speakers interpret, whether they perceive it or not, all language-related data through the confines of their own language. This relates to both sound and spelling. The sound system of a language is part of its grammar as are its structural elements. People are restricted in how they can perceive spelling based on the details of their language.