First half of the Intro to my PhD. Any thoughts will be most welcome. I will begin by exploring the spelling of just one word, television, in all its forms, derivatives and related words: telly, TV, televise televises, telethon, Teletubbies… A close analysis of this kind will give us an idea of the number of issues that need to be understood and explained in a full theory of English spelling - and also the impossibility of doing that. Letters and phonemes Television begins with a simple mapping from letters to sounds. The corresponds to /t/ and the to /ɛ/. These are like default spellings, the kind of correspondences people have traditionally expected from an alphabetic spelling. One symbol for one sound. Things get a bit more complicated after that. The /l/ is represented by just one l. While this still might seem normal, it is regularised in the clipped form telly. There is something odd about a system that prefers to use two ls to represent just one sound. Our theory of spelling is somehow going to have to account for this kind of positionally determined variation. let has one l but tell and telly have two. Next we have an unstressed vowel sound, represented by the same letter as the the stressed /ɛ/. This is also strange. Why does this letter represent two different sounds? And what exactly is that sound? Is it the same unstressed vowel that ends telly? For some speakers it is, others not. Assuming the former means several things. The first is that the sequence /tɛlɪ/ can be spelt either or , so the system allows for more than one way of spelling the same set of sounds. The second thing is that a spelling determined by the words pronunciation necessarily prioritises one accent over others. Yet people speaking with those other accents can still read the word, even if there remains a certain degree of inconsistency between the spelling and their pronunciation. A spelling must represent something more than just the sound of a word, if it even does that. It will be crucial to explore what a spelling is - or might be - before we can try to understand exactly how it functions. Syllables, Feet and Stress The appears fairly unremarkable, other than to say that /v/ is the most consistently represented phoneme of English. There is scarcely a in the language. Only in modern slang words like navvy does occur. Indeed /v/ is scarcely ever represented any other way, except for some speakers nephew. Yet this apparent consistency comes at a price. The spelling doesnt demonstrate the difference between never and fever. What we have here is consistency at the phoneme level but not at the syllabic level. There is no relationship between the and the . Doubling consonant letters marks vowel length and, secondarily, breaks the word into syllables. Compare hopping and hoping. Furthermore, the pronunciation of the word vision is dependent on the entire foot. Analyses dependent on mapping single phonemes to letters and back are not able to explain how /ʒ/ is represented. Evidence is beginning to suggest that the orthographic - or graphematic - foot may well have a major role in spelling (Evertz and Primus, 2013). The understanding of these higher units of representation will help to fill a massive gap between two very different traditional approaches to spelling. One approach has been the traditional and ultimately failed effort to simply map phonemes to letters and back. The other, more anecdotal approach, has been to consider English to be an etymological system, where the spellings are largely the same as the language from which they derive. Etymological spelling Preserving the etymology of the spelling means that knowing the pronunciation of a word is not always enough information to know how to spell it. This is why television has one but telly has two. Tele derives from Greek and Vision is from Latin. The spelling may even have been formed in French first, as , borrowed into English and amended slightly to suit the local system. Accounts differ. (Compare OED and etymonline) Clipped or slang forms like telly are often spelt using the conventions that have developed for base, ancient vocabulary like belly. Subsystems English, it seems, is not one big morass of complicated spellings. Instead it is a system of interlocking subsystems that are yet to be fully comprehended. Different spelling patterns can occur in different sectors of the vocabulary. Scientific terms - and television was once one of those! - are often formed by combining the spellings of §Greek or §Latinate morphemes (not usually both) and amending them somewhat. Slang or abbreviated forms like telly are formed within the §Basic subsystem. Compare also §Latinate lunatic and §Basic looney. Thus are born complications that we must be able to account for. Words like shoot and chute sound identical but have different origins and thus different spellings. This results in many, many complications. Compare the variant pronunciations of : there is §Basic chancer, §French chauffeur and §Greek chaos. This is not ideal, to say the least, although recognizing the patterns is clearly the first step to understanding them. A little work has been done on the topic, and Carney (1994) introduces the notation and categories used here. The system makes a lot of sense in a world where people are literate in more than one language, especially French, Latin and perhaps Greek too. It will be important to remember this when we come to explore the development of English spelling in the multi-lingual Anglo-Norman period and later on during the gradual standardisation process in the Early Modern period where Latin and Greek play major roles. In the modern era, connections can easily be made between English and other languages. The words look alike even if the pronunciation is quite different. This is especially useful for learned or scientific vocabulary - a category that television once fell into. Morphemic spelling Another major advantage of etymological spelling is that it allows connections to be kept between morphologically related words in English. Many words begin with the sequence even though the stress and vowel qualities change. Compare telepath, telepathy and telepathic. Notice also how television and televise differ, using and for related alternating sounds. The same pattern occurs in revise - revision and circumcise - circumcision. This is a basic principle of English spelling but it is not without its glitches. deride - derision does not follow the same pattern. In this case, and in many others, the words were formed in other languages (usually Latin or French) and borrowed with alternate spellings that reflect the morphophonological patterns of those languages, rather than English. Thus etymology and morphology interact and even contradict one another. This is a problem. An awareness of morphemic spelling emerged in the late 1960s when it was found that English tends to prioritise consistency at the morphemic level rather than phonemic, or indeed any other level (Venezky 1967, 1970; Chomsky and Halle 1968, henceforth SPE). English spelling can represent morphophonemic alternations optimally (SPE 184n), at least given the limitation of using only twenty-six lower-case letters arrayed linearly. Thus we have, inter alia: • profane - profanity • serene - serenity • divine - divinity This approach accounts for a huge number of apparent problems and silent letters. Venezky notices the orthographic similarity between sign - signature and paradigm - paradigmatic. Other examples include the consistent use of plural in cats and dogs, despite the surface phonetic variation. Getting into the system Unravelling how words get into the system in the first place will be a major part of this thesis. In all of the examples so far the spellings have been borrowed from another language and then adapted to suit English. Television isnt identical to its Greek or Latin constituents. Firstly, a transliteration process has turned the Greek combining form into Latin . This in itself is a process biased by analysis. The same Latin represents the two different Greek letters. So some information is lost here. As for the vision part, it is an Anglo-Norman word which, out of existing variants such as , and , was finally modelled on Latin visio, visionis. (oed.elib.tcd.ie/view/Entry/223943#eid15366983). Importantly, there is no Latin form spelt . It is a stem. The formation process involves an interaction between at least: • the existing pronunciation (if it exists), • the available etymological spelling, • the existing spelling patterns in the language, and • the morphophonemic alternations of the grammar. The interaction between these competing forces is clearly manifest in different ways in different languages. French has télévision, employing accents to indicate the pronunciation. Italian televisione has an extra , while Portuguese represents nasality with a diacritic. Polish, a Slavic language employing a very different set of sound-spelling correspondences, is less able to respect the etymology, rendering it as telewizja. Irish goes further with teilifís, employing a diacritic to indicate the quality of the lateral consonant. Unlike English, Irish cannot easily reconcile the words etymology with the languages rather different consonantal system, characteristic of the Goidelic languages. Such reconciliation is a major challenge in English spelling formation. Being in the system What words actually do after they enter the system is a rather different matter to how they got there in the first place. This is why there is a conflict between the concept of etymological and morphological spelling, one which has not been resolved at all in the literature. The derived form televise and its inflected forms televises, televised are formed within English and subject to slightly different formation forces to television. The morpheme is now more important than the etymology. The importance of this distinction will become clearer once we come to blend words like Teletubby and abbreviations such as TV. Amendments and graphotactics What the two kinds of formation do share is the constant process of amendment. Just as the Greek form was adapted to Latin and the Latin form was adapted (or merged with the Anglo-Norman) to suit English, so English continues to amend and adapt its own formations. Televise is clearly not spelt the same as television. An extra is added as a kind of pronunciation guide to the reader, helping us to work out that the word rhymes with rise not viz. This also makes it a remarkably English-looking word. This would not occur in French or Latin, for example. It also prepares the word for its inflected forms, televised and televises. These amendments fall into the category of graphotactics, a slippery concept that has never really been pinned down in the literature (e.g. Carney, 1994: 66-69). We might initially think of it as the set of devices used in spelling to help the reader work out how to read a word. For pronunciation, we have consonant gemination (hopping v hoping), the use of to mark vowel length (hop v hope) and the use of to mark consonant quality (age v ague). But the term is also described to describe some grammatical and even historical trends too. There is the three-letter rule, whereby non-grammatical words must have at least three letters (Albrow, 1972). This is useful for distinguishing homophonous pairs like to - too and by - bye. Other strange patterns include never ending a word with , or (e.g. happy, blue, love). The concept was constructed as a parallel to phonotactics, which deals with permissable strings of phonemes. Carney observes that any definition of graphotactics is therefore contingent upon the definition of the grapheme. Doing this will be a pre-requisite to a more advanced and wide-ranging definition of graphotactics to be developed later. This will include all reading devices used across the entire writing system, be they consistent and systematic or not. This view will allow for a much more radical view of how writing systems operate, arguing that visual devices such as italics, bold, capitals and even colour, can all be put into the same functional bracket as pronunciation aids such as the last in televise. Graphemes The grapheme is a termed designed as a parallel to the phoneme, used to describe the fundamental building blocks of alphabetic writing systems. Like phonemes, graphemes are expected to have allographs and be subject to graphetic and graphemic variation. It is a very structuralist concept, working off the principle that is not is not . These are graphemes of English. But is fundamentally the same as or , or any other visibly recognisable allograph. This intuitive concept quickly runs into major problems once it is used to describe writing systems. If and are graphemes, what about ? How many graphemes are there in thorn? Five - one for each letter? Four - one for each phoneme /θɔrn/? What about non-rhotic /θɔ:n/? Other problems include capital letters? Are and both allographs of the same grapheme, despite looking so different? With these complications, there still exist two very different and contradictory definitions in the literature. One is the referential view of the grapheme, the other is the analogical view. The referential view deals with the correspondences between phonemes and letters or letter clusters. In this view is a grapheme, as are the and of bath, Bach and batch. The analogical view is concerned with graphemes as individual visible units, rather than how they combine to produce units of linguistic representation. The reason why the two definitions can still exist is because there does not exist a theory of spelling or even writing which takes a broad enough view of the subject that it has been necessary to explore the limits of the concept and distinguish between them fully. Put another way, nobody has been interested enough in how writing can be used to create meaning on the page independently of the spoken forms. The viewpoint that will be developed in this thesis is that the grapheme is an abstract form whose visible realisations are primarily used to render linguistic forms on the page. Videlicet, they are used for spelling words: . However, they can have a host of other uses. Their second use is graphic. The letters might be rendered in such a way as to give them an extra iconic or pictorial meaning. Logos are designed on this principle. The letters have to have adhere to the basic shape required. But that shape can be distorted, coloured and resized to create further meaning. In the Zippo logo, the dot on the is also a flame. The colour draws attention to this, and the choice of colour is iconic, representing fire. Insert Zippo logo Such formations have not traditionally interested linguists. Surface graphic variation is the stuff of calligraphy and art history (Daniels, 1994). It has nothing to do with the representation of linguistic structures. Yet these two distinct categories seem to be merging more and more, especially in the professional world of graphic design, where writing meets art. The graphic dimensions of colour, shape, size and position can all be used to help us read and interpret a text. Words might be coloured differently instead of spaced. This is particularly useful for websites. Sportingbet does not have a space in its URL. This can go further. In RBYES campaign, colour and position are combined, allowing the to be used twice, once as part of RBS, once as part of YES. E.Gs Colour used to break up words (sportingbet.cm) RBYES (colour and position combined) Boylesports (all of them) Essentially, this is an extension of the traditional use of italics, bold and capital letters. All of them are graphic devices which can be used with differing degrees of conventionality. It will be argued in depth that these are an essential part of a broad theory of spelling, subsumed under the concept of graphohacktics, whereby writing is seen as an ongoing series of hacks, used to help the reader makes sense of the text. Next up • Constructed homophones • Blends, puns, iconicity
Posted on: Thu, 28 Nov 2013 13:14:22 +0000
Trending Topics
Recently Viewed Topics
© 2015