Also on this page: Radio National, Australia intervew with Philip Seymour.
Report by World of Dyslexia newsletter, by James Randerson with thanks to the best source of science research - New Scientist (September 4th 2001).
English is toughest European language to read.Despite being the world's lingua franca, English is the most difficult European language to learn to read. Children learning other languages master the basic elements of literacy within a year, but British kids take two-and-a-half years to reach the same point.
In the most extensive cross-national study ever, Philip Seymour of Dundee University and his team compared the reading abilities of children in 15 European countries. They found that those learning Romance languages such as Italian and French progressed faster than those learning a Germanic language such as German and English. "Children do seem to find English particularly complex and problematic though," says Seymour.
The team focused on the earliest phase of learning to read. They tested the children's ability to match letters to sounds, their capacity to recognise familiar written words, and their ability to work out new words from combinations of familiar syllables.
Seymour's findings might explain why more people are diagnosed as being dyslexic in English-speaking counties than elsewhere.
In languages where sounds simply match letters, some symptoms just would not show up, says Maggie Snowling, a dyslexia expert at the University of York. The condition would be more difficult to diagnose in children who speak these languages, though subtle symptoms such as impaired verbal short-term memory would remain. "People might be struggling, but no one would notice," she says.
Consonant clusters.The Germanic languages are tricky because many words contain clusters of consonants. The word "sprint", for example, is difficult because the letter p is sandwiched between two other consonants, making the p sound difficult to learn.
Another feature of English that makes it difficult is the complex relationship between letters and their sounds.
In Finnish, which Seymour found to be the easiest European language to learn to read, the relationship between a letter and its sound is fixed. However, in English a letter's sound often depends on its context within the word. For example, the letter c can sound soft (as in receive) or hard (as in cat). Many words like "yacht" don't seem to follow any logic at all.
Historical accident.However, the things that make English difficult to read might have contributed to Britain's rich literary tradition. Words like "sign" and "bomb" are difficult because of their silent letters, but these hint at relationships with other words. The connection with words like "signature" and "bombard" is obvious.
Mark Pagel, an expert on language diversity at the University of Reading, acknowledges the irony that despite being the international lingua franca, English is the most difficult to learn. The dominance of English has more to do with historical accident than any innate superiority of the language, he says.
"People who speak English happen to have been the ones that were economically and politically dominant in recent history.
Those forces greatly outweigh any small difficulties in language acquisition.".
Radio National, Australia. lingua franca Saturday 22/09/01.
Philip Seymour on Finnish...
Summary:Why Finnish is easier to learn to read than English... Philip Seymour on his comparative study of the reading abilities of Grade I children in 15 European countries.
Philip Seymour is Professor of Cognitive Psychology at the University of Dundee. He presented the results of his collaborative study of children's foundation literacy skills in 15 European countries at the recent British Association Festival of Science in Glasgow. They show some startling differences, the most glaring being that children who were learning to read English (Grade I children in Scotland), lagged well behind their peers in 14 other countries. Why should English be harder to learn than any other European language?Jill Kitson: Welcome to Lingua Franca, I'm Jill Kitson.
This week: Why Finnish is easier to learn to read than English. Cognitive psychologist Philip Seymour on his comparative study of the reading abilities of Grade 1 children in 15 European countries.
Philip Seymour is Professor of Cognitive Psychology at the University of Dundee. His interest in how children learn to read and why some children have difficulty, and are labelled dyslexic, led him to wonder whether it's easier to learn to read in some languages than others. And first up, whether some European languages are easier than others to learn to read.
Professor Seymour set up a collaborative study of children's foundation literacy skills in 15 European countries and at the British Association Festival of Science in Glasgow earlier this month, he presented the results.
They show some startling differences, the most glaring being that the children who were learning to read English, Grade 1 children in Scotland, lagged well behind their peers in 14 other countries.
All 15 languages the children were learning to read were alphabetic languages. Why should English be harder to learn than any other?
When Philip Seymour talked with me a few days ago about his study, I began by asking him what common tests were given in the different languages?
Philip Seymour: Well, they're very simple tests. One was to be able to identify and give the sounds made by the letters of the alphabet, so the children just had a list of letters and they read down it, giving the sound for each letter. That was the first one.
The second one was reading what we regarded as very familiar words, so these would be common words which occur in the early stages of children's reading schemes, and they're words like 'house', 'school', 'boy' and so forth, very common words. And so we had sets of those in each language, and these again were presented as lists and the children simply read them out loud.
Then the third test was our test of what we call 'de-coding', which is working out a pronunciation for an unfamiliar form on the basis of its letters and sounds. So for this we used very simple nonsense words, which are not real words, but they can be worked out on the basis of knowing how each letter corresponds to a sound in the language. So those were the three tests which we used.
Jill Kitson: Now I suppose one of the things I was amazed at was that the assumption was that there was a common teaching approach used in the different countries, beginning with phonetic mastery of the alphabet, which I think is wonderful, but I have to say that when my own children started at school, there was something they were waiting for called 'reading readiness', and each child was expected really to learn at his or her own pace, and to pick up the alphabet somehow along the way. I take it that that wasn't a method used in any of the schools where you were testing the children?
Philip Seymour: No, I think that's right, that nowhere did we find a complete whole word method without any alphabetic teaching at all. I have seen that occasionally in Scotland, but it's not the usual method which is used, which is normally a combination of teaching a word vocabulary on the one side, and teaching about letters, sounds and decoding at the same time, so it's a kind of parallel approach.
Jill Kitson: Now were there any striking differences in knowledge of the alphabet at the end of Grade 1?
Philip Seymour: Not particularly. Almost all of the groups seem to have mastered the alphabet to a level of 90%-plus accuracy.
The only difference we found was in terms of the speed of reading the letters, which was a little bit slower in the Scottish group who were of course younger than any of the other groups.
Jill Kitson: Yes, now it's worth pointing that up. I'm amazed at the variation in the ages at which children start school. I mean children in Australia start at five, as they do in Scotland. Six was the more common starting age in European countries, but in quite a number, Germany, Austria, Finland, Norway, Sweden and Denmark, it's seven years. So did you find that there was any sort of optimum age for starting school? Was that a discovery you might have made in this study?
Philip Seymour: It's certainly an issue. I mean some people say perhaps we're starting too early at five. On the other hand, people say if it's a difficult language, then you need to start earlier. So that's one argument there. We didn't really find a strong association between age and how well children did on these simple assessments. So the correlation, say between age and the ability to read the nonsense words, was non-significant. There wasn't any association across the European groups - that's excluding the Scottish group.
Jill Kitson: But the variation broke into two parts, didn't it, the ones who had the Romance languages, Italian, Spanish, Portuguese, French, plus Greek and Finnish, did better than the children learning Germanic languages.
Philip Seymour: Yes, that was really just the decoding test, that's reading these nonsense words where it did seem to be easier to learn to decode in the more simple syllable- structure Romance languages than in the more complex syllable-structure of Germanic languages. It didn't actually make any difference to letter reading or word reading.
Jill Kitson: I mean there is that one-to-one correlation isn't there, if you're sounding words out, if you're learning by sounding words out, that you get in a language like Italian, that it seems to me, make it easier to learn than English, where one is running up against all sorts of variations and inconsistencies.
Philip Seymour: Yes, that's right. So there are really sort of two linguistic factors in the study. One is this difference between Romance and Germanic languages which has to do with the structure of syllables. Probably the more important one for our study is what's called orthographic depth, which is the consistency of the relationship between letters and sounds in the way the language is written. So in what's called the shallow orthography, there's a very consistent one-to-one association between each sound of the language and each letter, and that's the case for Finnish, for example, where each sound of phoneme in the language is represented consistently by a single letter. Other languages obviously have a more complex system in which you may need more than one letter to represent a particular sound, and the relationships may be inconsistent and variable.
Jill Kitson: Now tell us just how far the children in Scotland were behind their peers in other countries, which is quite striking. I mean the Scottish education system has a very high reputation, but they were markedly behind and in fact didn't catch up to where their peers were at the end of Grade 1 for another year or so.
Philip Seymour: Yes, that's quite right. The children were good readers so far as the norms in the United Kingdom are concerned, with reading tests. So they were actually reading ahead of their age. But on the tests which we were giving, they were far, far behind the other groups. So the score for, say, the familiar words was over 90% for the majority of languages, but was no more than about 30% correct for the Scottish group at the end of their first year, and then it had risen to 60% to 70% at the end of the second year. So it's really a very massive and striking difference.
Jill Kitson: So you then were able to make some sort of projected idea of how long it actually takes to learn English, and you put it down at something like 2-1/2 years as against what is achieved in that first year in Grade 1 in other countries.
Philip Seymour: Yes, that's right. The other countries are reaching the sort of mastery level within their first year of learning.
What we found was that the reading age needed to match the same levels in English was about 7 and a half to 8 years, and if the reading age scale starts at five years, as it does when children begin to learn to read, then that is implying that about 2 and a half years of reading experience are needed to establish this very basic foundation level of reading.
Jill Kitson: Was it common across all 15 countries for there to be some children who at the end of Grade 1 or Grade 2 you would label dyslexic, or is there some language there in amongst all those languages, that any child can learn to read?
Philip Seymour: Well that's an interesting question, obviously. If one looks at the distribution for the easier languages, then you find that the great majority of children are bunched together at the top end of the scale, but there is a tail of children who are doing less well. No real examples of children who couldn't read at all, but a few who are reading less well than the majority. So you could define dyslexia in that way for those languages. I suppose the point is that in English it would be a much more extreme failure which would count as dyslexia - in terms of being outside the distribution - than you would find in these other languages.
Jill Kitson: So what are the problems of English? Danish is the next hardest it seems, but English is strikingly harder to learn. It has this wonderful ability to take in words from many, many different languages, but is this actually creating the problem that we have with children learning to read?
Philip Seymour: It may be what's creating the problem at this early stage of reading, in that as you say, you have words which have come in from many different sources, and they've all got slightly different bases for their spelling, and that's what produces this great amount of inconsistency in English spelling. So that's producing problems for the early stage of learning to read. Of course there may be other aspects which come in later on: it may almost be an advantage to have a language in which the spelling signals the meanings of words or their grammatical structure and so forth, which is what happens in English.
Jill Kitson: Now of the Romance languages, French and Portuguese are harder to learn, and so what are the characteristics there that make them less easy to learn than other Romance languages?
Philip Seymour: Well I think they have a certain degree of what are called orthographic depth in that in French you may have multiple letters to represent phonemes and there's inconsistency, the spelling is influenced by the meaning or morphology of words; there are silent letters which have to be there for grammatical reasons, but which may not be pronounced and so forth.
So it is quite complex. Not as complex as English but not so different.
Jill Kitson: One of the things that I found myself considering was the fact that with English being the global lingua franca of commerce, popular entertainment, of the Internet, international relations and so on, this is rather paradoxical. But I wondered whether it means that global English has to be more like the English of, say, Grade 3 readers, a vocabulary that has simple syllables and simple phonetic correspondence; that from global English, the more difficult and complex words will simply be screened out.
Philip Seymour: I suppose that might happen; that would be a kind of world wide revolution in English spelling which could happen. I mean as you say, it is strange in a way that the language which is very widely used is the one in which it is perhaps most difficult to learn to read. Whether people learning English as a second language do find our spelling system a problem, I'm just not quite sure. Most of the people I know who have learned English and are using it in academic work and so forth, don't seem to me to keep complaining about the spelling system.
Jill Kitson: And just lastly: why did children learning Finnish, which is a sort of completely exotic one-off language, why is that easier to learn?
Philip Seymour: Well it has a simple syllable structure and it has a very, very consistent spelling system, so that's probably the main point. The language itself seems to me to be tremendously complex, if one was wanting to learn to speak it, because each word has numerous grammatical components attached to it in many, many different combinations, so it looks very, very complex to us to understand. But from the point of view of the reader it is straightforward and simple.
Jill Kitson: Philip Seymour, Professor of Cognitive Psychology at the University of Dundee.
And that's all for this week's Lingua Franca.
Back to the top.