The teacher stands in front of her Grade 4 class. The 45 nine and ten-year olds are crammed together at desks, huddled over shared books. Some are sitting on the floor. "Now, class, read from the top of the page," the teacher says. They comply in a slow sing-song drawl.
"Stop," says the teacher. "It is not 'Wed-nes-day', you say it 'Wensday'. It is what?" "Wensday," the class responds. "Again." "Wensday." The reading resumes, the teacher frequently stopping to correct her pupils' pronunciation.
Sometimes the children read aloud in groups. At other times, she calls a child to come to the front and read aloud. Not once does she ask a question about what the story means. Nor do the children discuss or write about what they have read.
This is the typical approach to how teaching is read in most South African primary schools. Reading is largely understood as an oral performance. In our research, my colleague Sandra Land and I describe this as "oratorical reading". The emphasis is on reading aloud, fluency, accuracy and correct pronunciation. There is very little emphasis on reading comprehension and actually making sense of the written word. If you were to stop the children and ask them what the story is about, many would look at you blankly.
Pronunciation, accuracy and fluency are important in reading. But they have no value without comprehension. Countries around the world are paying increasing attention to reading comprehension, as indicated by improving results in international literacy tests.
The problem with the oratorical reading approach is evident in the results of the recent Progress in International Reading Literacy Study (PIRLS) 2016 tests. PIRLS' purpose is to assess reading comprehension and to monitor trends in literacy at five-year intervals. Countries participate voluntarily. Learners write the test in the language of learning and teaching used in Grades 1 to 3 in their school.
The tests revealed that 78% of grade 4 pupils in South Africa fell below the lowest level on the PIRLS scale: meaning, in effect, that they cannot understanding what they're reading. There was some improvement from learners writing in Sesotho, isiNdebele, Xitsonga, Tshivenda and Sepedi from a very low base in 2011, but no overall improvement in South Africa's performance.
South Africa was last out of 50 countries surveyed. It came in just behind Egypt and Morocco. The Russian Federation came first followed by Singapore, Hong Kong and Ireland.
South Africa also performs poorly in the Southern and Eastern Africa Consortium for Monitoring Educational Quality surveys. These show that in reading and numeracy South Africa is lagging behind much poorer African countries such as Tanzania and Zimbabwe.
Our research on reading at a rural primary school and an adult centre in the KwaZulu-Natal province showed that the oratorical approach to teaching reading was dominant both in the school and adult classes. Both adults and children were not learning to read with meaning, and so were not achieving literacy despite attending classes. Our findings confirmed the results of other South African studies.
So where does the problem lie and how can South Africa address it?
To understand the situation more deeply we interviewed teachers and explored how they had learned to read. We found that they teach as they were taught; an indication that oratorical reading is a cycle repeated from one generation to the next unless it is broken.
Teachers told us they assessed pupils' reading ability just as they were assessed by their teachers: by having them read aloud. Marks were allocated for individual oral reading performance. This was based not on understanding the passage, but on fluency and pronunciation. There was no written assessment of reading comprehension. Reading was about memorising sounds and decoding words.
This suggests that the problem in learners' performance lies in how reading is taught in most South African schools. Learners are taught to read aloud and pronounce correctly, but not to understand the written word and make sense of it for themselves. Another consequence is that the pleasure and joy of discovery and meaning-making are divorced from school reading.
There are no quick fixes, but there certainly are slow and sure ones. The first is to get reading education in pre-service teacher training right. A report by JET Education Services, an independent non-profit organisation that works to improve education, found that universities don't give enough attention to reading pedagogies.
Universities need to teach reading as a process that involves decoding and understanding text in its context, not just as a "mechanical skill". Countries such as India, with its great diversity and disadvantaged populations, have begun to address the need for this change in how reading is taught.
The second "fix" concerns in-service training. The Department of Basic Education has a crucial role to play here. Teachers need to reflect on how they themselves were taught to read and to understand the shortcomings of an oratorical approach.
Effective reading instruction, such as the "Read to Learn" and "scaffolding" approaches, should be modelled and reinforced. In a multi-lingual African context, strategies that allow teachers and learners to use all their language resources in making meaning should be encouraged. Teachers' own reading is vital, and can be developed through book clubs and reading groups.
The school environment is also crucial. According to the PIRLS interviews with principals, 62% of South African primary schools do not have school libraries. These are central to promoting a reading culture, as work in New Zealand shows.
Schools should develop strategies such as Drop Everything and Read slots in the timetable, library corners in classrooms, prizes for reading a target number of books and writing about them, and creating learners' reading clubs. Learners can draw on local oral traditions by gathering stories from elders, writing them and reading them to others.
Finally, the home environment is vital. The PIRLS research showed that children with parents who read, and especially read to them, do better at reading. Our research found that children with parents who attended adult classes were highly motivated to learn and read with their parents. Even if parents are illiterate, older siblings can read to younger children. The Family Literacy Project, a non-profit organisation in KwaZulu-Natal, has done excellent work in creating literate family and community environments in deep rural areas, showing what is possible.
Developing families as reading assets rather than viewing them as deficits can help to strengthen schools and build a reading nation.