The picture looked even darker when it emerged that out of eight countries taking part in the study, only Poland has a higher percentage of people at this basic literacy level.
However, the study scotches the myth that there was ever a golden age of literacy; there was a higher proportion performing at the lowest level among the over-45s than younger age-groups. Further inspection makes the international comparisons less grim. Britain does have 22 per cent performing at the lowest level of the five-band scale, but 17 per cent of the population fall into the top two levels (defined as the ability to comprehend and manipulate complex text and specialist language and draw inferences from it). The country has more people at this level than Switzerland, Germany or the Netherlands.
The study, carried out by the Office for National Statistics, measured three types of literacy: prose literacy (extracting and comprehending information from fiction or newspaper articles); document literacy (understanding and using timetables, graphs, charts and forms); and quantitative literacy (the ability to understand and use numerical information). The survey covered a random sample of 3,800 16 to 65-year-olds in Britain. Other countries which have contributed are Canada, the United States, Germany, the Netherlands, Switzerland, Poland and Sweden.
Poland had by far the highest number of young people in the semi-literate category (43 per cent) and only 3 per cent at the highest level, while Sweden came out top, with 32 per cent of the population highly literate and only 8 per cent at basic level. The English-speaking countries tend to fall between these two extremes and show similar patterns, although there were more younger people at the lowest levels in the US, and both the US and Canada had a slightly bigger proportion of highly literate citizens.
Within the UK, there was no difference between the Scots and the English, but the Welsh scored lower on all three scales. On prose comprehension, 17 per cent of English and Scots perform at level 5, compared with 9 per cent of the Welsh.
There were predictable correlations between poor literacy and low-skilled, low-paid work or unemployment, and high levels of literacy with the top jobs. Most of the respondents had a high opinion of their literary abilities despite all evidence to the contrary - 46 per cent rated their literacy as "excellent" and a further 40 per cent as "good" even though over half the respondents fell into the bottom two levels on the prose scale. People were only slightly less happy with their writing and numerical ability.
The study also threw up some insights into contemporary British life; 18 per cent of us never read a book (although 31 per cent read one every day); 93 per cent of households contain a dictionary, 83 per cent had more than 25 books, 70 per cent take a daily paper. Those who used computers regularly tended to be highly literate.
Alan Wells, director of the Basic Skills Agency, said the survey revealed a pressing need for a national literacy campaign covering all age groups. "The Government's drive on literacy in schools is vital, but it is important not to write off older generations who are going to be in the workforce for some time."
He also suggested that the inclusion of numeracy as a branch of literacy "almost certainly skewed the survey" because people who score highly on literacy are not necessarily equally confident with numbers.
Having difficulty filling out forms and "understanding information from Government departments" might say as much about the authors of the documents as the form-filler.
The survey showed that literate parents produce literate children, so helping parents would also help schools.
"We need to look outside education, at workplace schemes and mentoring schemes, rather than just telling people they need to go to college," said Alan Wells. The Government would, he admitted, first have the delicate task of disillusioning all those who exist blithely under the impression that their literacy is "excellent".
Adult Literacy in Britain, Office of National Statistics Survey. Pounds 30 from the Stationery Office.