What are we really teaching? The implications of including phonics instruction in Group Guided Reading

SACMEQ) 2005, 2010, and 2017, all highlight the inability of young South African learners to Background: In the context of the ongoing crisis in early reading literacy in South Africa, this article gives insight into teacher practices which reduce learners’ opportunities to read continuous text for meaning. Aim: This ethnographic study investigated the microcosm of teacher practices in well-resourced environments. Setting: A purposive sample was drawn of three Grade One teachers in former Model C primary schools for intensive investigation of their literacy teaching practices. A pilot study established that a literacy teaching event resembling Group Guided Reading was attended every day by every learner in these classrooms and therefore provided the focal event in the research. Method: Micro-ethnographic classroom observations, with accompanying video recordings and transcriptions were the primary data sources. Interviews provided narrative and descriptive data. Data were analysed by means of qualitative content analysis and discourse analysis. Results: Findings suggested that the introduction of flashcards and texts designed for phonics practice into a methodology designed to promote reading for meaning has a cumulative impact on the time teachers spend on discussion, modelling comprehension strategies and silent reading. Potentially this limits learners to only one of the four roles of a reader, that is, a reader as code breaker. Conclusion: Findings caution against teachers making changes to the requirements of explicit teaching methodologies. The aims of Group Guided Reading are not achievable when phonics instruction reduces engagement with continuous text. Contribution: This article, by critiquing a practice, encourages teachers and teacher educators to consider the balance between whole-language and phonics-based approaches.


Introduction
Using a whole-language understanding of literacy teaching and learning, this article reports on the activities which were added to and omitted from Group Guided Reading, as prescribed by the Curriculum and Assessment Policy Statement (2011) by three teachers in South African Grade One classrooms.It discusses the effect these additions and omissions may have on young readers' ability to read for meaning.The author uses Castles, Rastle and Nation's understanding that 'the goal of reading development must be to develop a system that allows learners to construct meaning from print ' (2018:7).The discussion of these findings draws on literature to critique the teachers' practices and suggests that constructing meaning from print may be more difficult for emergent readers in classrooms where these practices are followed.read for meaning (Govender & Hugo 2020).The Progress in International Reading Literacy Study (PIRLS) over three cycles has confirmed low reading literacy among Grade 4 learners.
Poor reading progress in junior grades is one of the most significant challenges facing education in South Africa today.However, a decade of interventions focused on improving reading literacy has done little to change the performance of South African learners in the Foundation Phase and Intermediate Phase during national and international assessments.While the Background Report for the 2030 Reading Panel (Spaull 2022) maintains that there has been a steady improvement in the Pre-Progress in International Reading Literacy Study assessment, from 13% of Grade 4 learners achieving the Low benchmark in 2006, 18% achieving the Low benchmark in 2011 and 22% achieving the Low benchmark in 2016, this increase is only at the Low benchmark which assesses learners' ability to retrieve explicitly stated information, a minimum of what can be described as reading for meaning.There was no improvement at the higher benchmarks.The Progress in International Reading Literacy Study 2021 will provide insights into the effect of the COVID-19 pandemic but early indications are that learners in the early grades, already underperforming against international benchmarks, have lost 1.3 years of reading ability (Spaull 2022).
Even before the pandemic disrupted learning there was ongoing cause for concern.In spite of the general acknowledgement that 'the most effective way to bring about socio-economic transformation in South Africa is to improve reading outcomes amongst poor learners' (Taylor 2017:16), change has been slow.
Several interventions were initiated in response to these widespread indications of poor literacy performance.Examples are the Gauteng Primary Language and Mathematics Strategy (2010Strategy ( -2014)), the Early Grade Reading Study (2015Study ( -2016)), the Funda Wande Coaching Intervention (2019), the Primary School Reading Improvement Programme (2017), the Funda Wande Teacher Assistant and Workbook intervention (2021) and the ECDOE Reading Plan (2019)(2020)(2021)(2022)(2023).Since 2011 the Department of Basic Education (DBE) has supplied free workbooks and readers in all home languages to all grades R to 6, together with readers for grades 1-3.
However, the EGRS evaluation report acknowledges that simply distributing resources has not improved reading outcomes significantly and that the effectiveness of the LTMs depends on how they are used by teachers (Taylor et al. 2017:18).The 2022 Reading Panel Background Report suggests that changes to teachers' practices is the key to improved outcomes (Spaull 2022).This report concludes that 'on SA's current trajectory, it will take 80 years to get all learners reading for meaning' (Spaull 2022:8).That is, 80 years until most learners are able, at a grade-appropriate level, to discuss and understand what they are reading: to comprehend.This situation provides both the context and the rationale for the research reported on in this article.
It is important to assert that this article does not argue against systematic, focused teaching of letter, word and sound relationships.This also took place daily in whole-class teaching in the classrooms during the study.Explicit phonics teaching has been shown to benefit early reading literacy in South Africa (Taylor et al. 2017).Instead, this article focuses on the effect of adding phonics and word-recognition drills to methodologies designed to teach reading for meaning, using the example of Group Guided Reading (Fountas & Pinnell 1996, 2017).
Since the EGRS evaluation reports on the relative success of literacy teaching by teachers trained in Group Guided Reading (Taylor et al. 2017), it is likely that there will be an increased focus on training teachers in this methodology which is detailed in Table 2.However, the EGRS evaluation explicitly warns that 'there are some teaching practices such as Group Guided Reading that are difficult to enact and require additional development to be effective' (Taylor et al. 2017:14).This article presents an analysis of teacher practices in Group Guided Reading as an example in support of this warning.Luke and Freebody (1999) provide categories that help us understand the skills that classroom activities might promote among emergent readers.They suggest that readers must develop four roles or resources with which to make meaning of text: as a code breaker, text participant, text user and text analyst.Briefly expressed, code breakers use what they know about sound-letter relationships, sight words and print conventions.Text participants ask questions about meaning and attempt to understand both what is explicitly stated and what is implied.Text users employ their knowledge of written genres to approach a text appropriately.Text critics are able to evaluate a text in terms of the writer's intention.

Conceptual framework: Roles of the reader
Luke and Freebody contend that if learners do not develop the full range of roles, they will not be able to comprehend a text at all the levels required by formal schooling in a literate society.It is common for early readers first to be offered the role of code breakers, in other words, for there to be an early focus on this role, and for the other roles to be introduced later.This is in line with some thinking about the appropriate focus of early literacy teaching, in which letter-sound relationships and rapid recognition skills are the focus in the first year.The Progress in International Reading Literacy Study 2007, however, maintains that learners in the most highly performing educational systems are introduced to all four roles simultaneously, at an appropriate level (Howie et al. 2007:45-46).The Progress in International Reading Literacy Study report attributes the poor performance of South African learners in that cycle to the trend of introducing comprehension strategies slowly, successively, late or not at all.
Luke and Freebody's four roles usefully bring together in a different terminology two theories of reading pedagogy: the whole-language approach to reading (Dixon & Tuladhar 1996;Reyhner 2020) and the phonics-based approach (Ehri 2020).In summary, the whole-language approach promotes the reading of authentic texts for meaning.Explanations of sound-meaning relationships, punctuation and spelling are embedded in this reading experience through discussion.This approach has been contrasted to the phonics-based approach which emphasises the principles behind spelling and sounding English words in an explicit, disembedded way.This binary, sometimes referred to as the 'reading wars' (Castles et al. 2018;Rhyner 2020) has been recognised as a false one.The hybrid balanced approach blends teaching learners to read for meaning with explicit phonics instruction (Petscher et al. 2020, Scanlon & Anderson 2020).
The balanced approach is a requirement of the curriculum (DoE 2011).Big Books, reading corners, buddy reading and Group Guided Reading exemplify a whole-language approach while flash cards and explicit sound-letter teaching and phonics instruction exemplify a phonics-based approach.The handbook Teaching reading in the early grades: A teacher's handbook (DoE 2008) guides teachers in these pedagogies.The Curriculum and Assessment Policy Statement (CAPS) (DoE 2011) requires Group Guided Reading based on the design of Fountas and Pinnell (1996) and Table 2 shows the contrast between the Fountas and Pinnell design, the elements recommended by the curriculum and the interpretations of the teacher participants in this research.Teacher participants confirmed in interviews that they implemented the balanced approach, and they were observed in a version of Group Guided Reading, Big Book reading, story times, buddy reading and independent reading corners as well as whole class explicit phonics teaching and practice every day.
To take this discussion further, although empirical evidence supports explicit phonics teaching to beginner and struggling readers, there are limitations to this approach (Johnston & Watson 2005;Rose 2006).In particular, a phonics-based approach alone, no matter how fluent learners become, is not enough to guarantee comprehension (Konza 2011).Many learners may have good decoding skills but do not necessarily understand what they are reading, an action called 'barking at print' (Dymock 1993:86).As the balanced approach suggests, phonics instruction should be one of many reading literacy experiences which support young readers finding meaning in what they are reading (Bowers 2020).This article presents an example of a practice that reduces learners' opportunities to read for meaning during Group Guided Reading.It highlights the unintended consequences of adjusting activities in a teaching event intended to promote reading for meaning.

Research method and design
The sites of the research The three sites of research, former Model C schools, exemplified those which best support literacy learning (PIRLS 2007).In well-resourced classrooms, small classes received early reading literacy instruction in an immersive environment with English as the language of learning and teaching (see Table 1).
The three teacher participants were English-speaking women who had added further qualifications to their Foundation Phase teaching diplomas.They attended teacher-training programmes annually and had training in remedial teaching.They had taught Grade One in their schools for between 5 and 19 years and also other Foundation Phase grades.They had well-articulated notions on teaching reading literacy; they identified experience as the most important guide to their practice.Within these broad similarities, Mrs Samuel had the least experience but was most highly qualified, having taught Foundation Phase for 16 years and Grade 1 for 5 years.She had a recent B.Ed. Honours degree.Mrs Michaels had the most experience, having taught in the Foundation Phase for 32 years and Grade 1 for 19.Mrs Danes had taught for 17 years in the Foundation Phase and Grade 1 for 4 years.These teachers prepared thoroughly for lessons and made many teaching materials themselves.They also used the same commercial-graded reading series, Ginn 360 (Ginn 1978), in their reading instruction.These classrooms therefore provide credible examples of government-funded environments most likely to exemplify excellent teaching practice in South Africa.This makes them appropriate sites to investigate teacher practices.
In these classrooms, the central literacy teaching event was called Reading on the Mat.It resembled Group Guided Reading, a pedagogy described in the curriculum (Department of Education [DoE] 2011) and outlined in the guidelines Teaching reading in the early grades: A teacher's handbook (DoE 2008).In the classrooms of the research every learner came to the mat every day in five or six groups of between six and eight learners, and the teacher worked with each group for periods lasting between 20 min and 40 min.Whole-class literacy learning pedagogies took place on a looser rotation during the week and with less pedagogic intensity.Reading on the Mat is a seating arrangement that promotes strongperhaps the strongest -normative work in these classrooms, because of the close contact teachers have with young readers in small groups.It seemed that the roles and priorities maintained on the mat would therefore strongly impact learners' notions of what it means to be a reader, whether as On the mat, the three teachers deviated significantly from the guidelines Teaching reading in the early grades: A teacher's handbook (DoE 2008) and the changes they made, and their potential impact, are the focus of this article.

Features of Group Guided Reading
Table 2 shows the recommendations in the original design of Group Guided Reading, comparing it to the requirements of the curriculum and to the digressions of the teachers in the study (bold type in the right-hand column).It is worth noting that the curriculum requirements are already an interpretation of Group Guided Reading but also that the teachers of the study were aware of the curriculum requirements.Their practice on the mat had not changed with the implementation of the Revised National Curriculum Statement three years before the study commenced (Interviews).
Although theoretically learners were in ability groups, instruction was largely undifferentiated.Teachers repeated the day's teaching five or six times with minor variations; learners were on the same level of graded readers.Weaker or stronger readers were given additional opportunities to read.They could be sent to the reading corner or to mats or tables in the corridor, paired with 'reading buddies' or given additional coaching by the teacher at break.

Discussion before and during reading (4 & 5 at the table)
Literature on whole-language reading theory recommends that teachers should promote learners' engagement with texts through discussion.This kind of interaction offers young readers a model of how to engage with and how to think about texts: It is thus through talk about texts that learners construct and develop facility in the mental activities that are involved in the literate thinking that makes possible the construction of 'scientific knowledge'.(Chang-Wells & Wells 1993:64) Similarly, Palincsar, Brown and Campione (1993) recommend structured dialogue or dialogic reading (Folsom 2017) in which the teacher provides explanations, modelling, support and feedback.Like text talk, this dialogue reveals the mental processes of comprehension to learners, and models strategies used by successful readers (Clark et al. 2003).There are also parallels with the teacher talking aloud or thinking aloud (Lipson 2007) which takes learners through strategies for comprehension (Traga-Philippakos 2021).The Progress in International Reading Literacy Study highlights the social benefits of discussion and the role it plays in drawing an emergent reader into a community of literate practice (Prinsloo 2013).At the same time discussion promotes 'intellectual depth' (Howie et al. 2007:19).

Questions before and after reading
Research upholds the importance of teacher questions in emergent readers' understanding of text (Blything, Hardie & Cain 2019;Dillon 1988;Thompson 1997;Wragg & Brown 2001).Questions have been called 'the teacher's most potent tools' (Petty 1993:139).However, there is also concern about the cognitive level of teacher's questions, with Applegate, Quinn and Applegate warning against a preponderance of lower-order literal and retrieval questions: 'Literal comprehenders may function effectively and may even be judged as competent readers while they cultivate a negative attitude towards reading ' (2002:175).Retrieval questions dominated reading activities both on and off the mat in the classrooms of this study.
With specific reference to Group Guided Reading, Fountas and Pinnell (1996) assert that: [T]he overall purpose of guided reading is to enable learners to read for meaning at all times.The instruction may involve brief detours to focus learners' attention on detail, but the construction of meaning overrides.(p.4) The Curriculum and Assessment Policy Statement confirms this, and even supplies lists of questions which 'will help develop both lower and higher order comprehension skills' (DoE 2011:20).
As well as questions which probe the text during reading, the pedagogy driving Group Guided Reading recommends a closing discussion of points raised earlier.This strategy confirms emergent understanding and strengthens perceptions of meaning, providing an opportunity to give opinions, to form judgements and to evaluate the whole text.These are higherorder thinking skills essential to developing learners as text participants, users and analysts.The concluding discussion may suggest comparisons with other texts, or invite comments on characters, their motivations or moral issues.It is important to note that these activities are also suitable for emergent readers at a simple level, appropriate to the text.
The Progress in International Reading Literacy Study (2007) recommends strongly that all levels of questions should be introduced in Grade One, maintaining that 'the reading achievement for learners for whom the skill was introduced in Grade 1 achieved higher than for those learners for whom the strategies were introduced in later grades' (Howie et al. 2007:46).The Curriculum and Assessment Policy Statement (DoE 2011) recommendations for discussion and questioning suggest that learners should be encouraged to participate, to use and to analyse text from their earliest encounters with reading.

Sustained silent reading
Silent independent reading or sustained silent reading requires learners to have a daily opportunity for voluntary free reading during class time (Garan & Devoogd 2018).This recommendation expresses the understanding that learners learn to read by reading frequently, and also that sustained silent reading, not reading aloud, is the goal of reading instruction.Garan and Devoogd (2018)   Teacher chooses a book for the session from a stock of 'real' books (not graded readers).
Teacher selects a graded reader at a level lower than texts for whole-class shared reading.Selects a language feature for the day's teaching.
Teacher selects a graded reader from a set or sets available, usually the next in the series.Language focus omitted.

2.
Teacher calls group members together and gives each a copy.
Inferred: Teacher calls group members together and gives each a copy.
Teacher calls group members together and gives each a copy.Does book and homework admin.

3.
Teacher sits with the group on the rug or around a small table.
Inferred: Teacher sits with the group.No mention of mat / rug.
Teacher sits with the group on the mat or on a small chair.

4.
Teacher introduces the book and some of the language in the book.
Teacher introduces the type of book.Makes links between the topic of the book and their own experiences.
Teacher leads phonics or word-recognition practice using cards or other text.

5.
Teacher points out aspects of the pictures and print.Learners ask questions or make comments.
Teacher talks through the illustrations, pointing out details and asking learners what might be happening.Points out organisational features.Introduces difficult words.
Teacher sometimes talks through pictures.Almost never talks through print, although may point out organisational features.Learners do not ask questions.

6.
Teacher asks learners to repeat the language of the book.Conversation as they notice and point out things.

No recommendation
Teachers seldom engage learners in discussion or encourage questions.May ask each learner a single retrieval question.

Teacher asks learners to locate individual words in the text on several pages
No recommendation Omitted

8.
Each learner reads the whole book softly while the teacher observes.She may interact briefly to solve difficulties but she tries not to interrupt.At the same time she assesses their ability to use reading strategies, and whether the book is at the right level.
Learners read silently until the teacher asks them to read aloud.The teacher observes their reading and may teach additional strategies based on what she sees.The teacher moves from learner to learner and listens to a small section of the text read aloud, prompting learners with questions on the text or making suggestions for strategies.
Mrs Danes: All learners read the book silently and choose a double-page spread to read to her before leaving.She observes their reading closely.There is no explicit teaching of reading strategies.She asks a retrieval question of each learner.

9.
Using a small whiteboard, the teacher teaches a decoding strategy to aid comprehension.

No recommendation
Mrs Danes uses word cards to teach decoding strategies.Mrs Michaels teaches strategies verbally in response to learners' reading.
Mrs Samuel teaches strategies verbally in response to learners' reading.

10.
Learners locate the focus word of the decoding teaching in the text.

No recommendation Omitted
Teacher returns to questions raised in the text talk earlier.May ask questions on phonics or grammar, or to develop comprehension.
No whole-group final discussion; no questions from learners.

11.
Learners practise unison reading with fluency and phrasing.
No recommendation Mrs Michaels requires regular unison reading for fluency and phrasing.
Mrs Danes and Mrs Samuel ask for unison reading at the end of the year to move through the text more quickly. 12.
Teacher adds the book to the browsing box of previously read texts so that learners can re-read it.
Learners re-read the text alone or in pairs on following days.
Learners choose Box books for homework reading and also re-read books from the graded readers, but move steadily up the levels.Teachers supply books from additional series of graded readers for consolidation.

13.
The teacher may do an extension activity with the group.

No recommendation
Extension takes place outside reading on the Mat.

14.
One learner remains for assessment of a 'known text'.No recommendation Formal assessment takes place separately from reading on the mat.
Source: Fountas, I.C. & Pinnell, G.S., 1996, Guided reading: Good first teaching for all learners, Heinemann, Portsmouth, NH; and Department of Education (DoE), 2011, Curriculum and assessment policy statement.English home language.Foundation Phase, Government Printer, Pretoria reading pleasure which has been shown to be powerful driver of a life-long reading habit.

Research design and analysis
This article reports on one aspect of findings from microethnographic research into the core literacy learning event in three Grade One classrooms (van der Mescht 2013).
Ethnography is well suited to investigating literacy practices, as well as classroom interactions (Bourne 2002;Day & Park 2005;Dyson 1996;Gee 2004;Hawkins 2005;Heath 1983;Hoadley & Ensor 2005;Jones 1989;Prinsloo & Steyn 2004;Street 1995Street , 2001;;Toohey 1998Toohey , 2000;;Willett 1995).Street affirms that research into literacy practice 'necessarily entails an ethnographic approach which provides closely detailed accounts of the whole cultural context in which those practices have meaning ' (1995:29).The benefit of using an ethnographic approach to investigate teachers' practice is that it demands a detailed description of the event before analysis.
Linguistic ethnographers approach the speech event (Hymes 1974) with the open question 'What is going on here?'To answer it, they observe, interview participants, make notes from informal conversations, videotape or record activities, photograph and collect artifacts used in the event.
Ethnographers seek access to anything which might shed light on the practices they are observing.The research reported on here was a linguistic micro-ethnography (Garcez 2017) and included all the activities named previously, but in a restricted time scale and with a focus on a single representative event, Reading on the Mat.
The narrower time scale and reduced size of the event in turn enabled micro-analysis.Data were subjected to reiterations of analysis, allowing them to generate their own categories and thereby to present their own significance.In addition, early interpretations were presented to participants and their responses were further analysed.These cycles tested the authenticity of interpretations made of participants' intentions during the event.Choosing an ethnographic approach therefore also impacts analysis and interpretation.This inductive approach is a characteristic of interaction analysis, which has been used since the 1960s to investigate interactions in educational environments.Jordan and Henderson (1995) write: Verifiable observation provides the best foundation for analytic knowledge of the world.This view implies a commitment to grounding theories of knowledge and action in empirical evidence, that is, to building generalizations from records of particular, naturally occurring activities, and steadfastly holding our theories accountable to that evidence.Underlying this attitude is the assumption that the world is accessible and sensible not only to participants in daily human interaction but also to analysts when they observe such interaction on videotape.Analytic work, then, draws, at least in part, on our experience and expertise as competent members of ongoing social systems and functioning communities of practice.(p.4) In line with these principles, data were collected in a week in each term, over the course of a year from three classrooms.The researcher observed all literacy learning events in these classes, but the primary data for analysis were the video recordings of Reading on the Mat of each successive group during the day.These were supported by audio recordings, observation notes, classroom materials, formal and informal interviews with teachers and the texts and artefacts used in teaching on the mat.
Hymes ' (1974) eight categories for analysing a speech event in linguistic ethnography were used at the first and broadest level of analysis.When a category revealed richer, or conflicting data, then additional analytic tools were applied to the same data for sharper focus.As a result the study brought together findings from analyses of setting, participants, goals and outcomes, sequences of acts, norms, teacher and learners' questions, the key or mood of the event, analysis of the verbal and visual texts of books, a discourse analysis of teacher's speech, a concordance search for teacher's questions tags and the non-verbal communication of gesture and posture.The study (van der Mescht 2013) from which this article is drawn, showed how, in different modes and at different stages of the interactions, the most positive reinforcement was given to learners during decoding practice; that is, their most valued role was as code breakers.
Ethical clearance for this research was granted by Rhodes University Education Department Higher Degrees Committee; informed consent was obtained from all participants.Schools and teachers have been given pseudonyms.

Findings Comparing recommendations in curricula to practice
The entry point for the analysis of teaching activities on the mat was Teaching reading in the early grades: A teacher's handbook (DoE 2008) where clear guidelines are given for the interactions in Group Guided Reading, closely based on Fountas andPinnell (1996, 2017).Table 2 shows that while teachers' practice aligned with some features of Group Guided Reading in the curriculum other features were omitted.The mechanisms of this are examined in further detail below and possible reasons for the changes are the focus of the final discussion.

Reading on the Mat
Table 2 shows not only where the participating teachers had added activities, but also what they had omitted in order to make the additions.The additions and the omissions both change the learning experience, and foreground particular roles for emergent readers.The mechanisms of this are examined in detail below and provide the focus in the final discussion.
Examining the changes to recommended practice, it is appropriate to ask why the three teacher participants, independently of each other, changed Group Guided Reading by adding the teaching of letter-sound relationships, phonemic and phonological awareness and word-recognition skills.They all identified the changes as innovations of their own and each change involved considerable additional work in preparing booklets, notes and flashcards.

What was introduced? Activities with vocabulary cards
At the beginning of each session on the mat, the three participating teachers used word and letter cards to practise phonics skills in letter-sound relationships and word recognition.Large flashcards were usually shown to all the learners in the circle one after another, with each learner saying the word as they were shown it.
Mrs Danes spent on average 71% of her daily time with each group on the mat with cards, and Mrs Michaels spent 66% of her time with each group on the mat with cards, Ladder books or the self published Reading is fun phonics primer (Smook 2008) which is referred to again later.When Mrs Samuel used the graded reader, Ginn 360, 80% of the time was spent engaging with the narrative text.However, she did this only once or twice a week.On other days she focused exclusively on phonics, poetry, or word games on cards when she taught on the mat.
The small cards were about 1.5 cm square with letters (early in the year) or words (later in the year) printed on them.Learners had their own sets for homework and word or sentence construction at their desks.The small cards were used daily on the mat to sound out letters, to construct words or sentences, and in simple letter or word identification drills.In this regard, the practice of the three teachers was identical.
Teachers' practice with larger flashcards was also identical.However, the cards themselves varied: Mrs Michaels and Mrs Samuel used commercially printed cards, as well as cards they had made themselves.The words seemed to be chosen at random; that is, they did not relate to other texts.Mrs Danes' large cards replicated the small cards and were used every day in game-like interactions throughout the year.The basic activity was that the teacher presented the card to each learner in turn, who then said the word.
As mentioned above, Mrs Danes' practice was to use cards 71% of the time on the mat, and Ginn 360 graded readers as the only other text.Although activities with the cards were a decontextualised word drill, the vocabulary on the cards was that of the Ginn 360 series, but a level above the one at which learners were reading.Her word-recognition drill, therefore, prepared learners to recognise the words in the next level of readers, promoting fluency and supporting comprehension.Mrs Danes' practice of blending card drill and the vocabulary of the Ginn 360 readers, was the most systematic of the three teachers, as she explained: 'if you are not 95% fluent you don't have sufficient comprehension because you're so busy struggling to read you lose meaning' (van der Mescht 2013).

Activities with Ladder books, Reading is fun, Yellow books and Flip files
In addition to the cards, Mrs Michaels and Mrs Samuel used Ladder books, Reading is fun (Smook 2008), Yellow books and Flip files for consolidating phonemic awareness and in word-recognition drills.The Ladder books were lists of words and the Yellow books were texts focusing on a particular sound.These were developed by Mrs Michaels.Mrs Samuel had developed Flip files of her own texts.Reading is fun (Smook 2008) is a home published phonics primer.Mrs Danes did not use any comparable text in her classroom but worked with cards for the majority (71%) of her time on the mat.These texts were not used in whole-class teaching, or as part of the daily phonics instruction, but instead were the main focus in Group Guided Reading.
These texts have been grouped for discussion as they shared two features.Firstly, they were not illustrated, thereby removing pictures as a cue for meaning.I return to the significance of this later.Secondly, their language was disjointed and the texts lacked a storyline, because they had been written to teach sounds.Examples and how they affected practices on the mat are to be discussed.These texts had a tongue-twister quality derived from the closely repeated, similar sounds.This distracted readers from meaning and presented instead a demanding decoding challenge.While this might indeed have been the point of the activity, the consequence was to replace discussion and silent reading with texts which made predictive, interpretive or evaluative questions either inappropriate or impossible.They were never used for sustained silent reading.Huey criticised the 'unnatural, boring, and meaningless sentences found in phonics primers', referring to them as 'sentencehash' (Snow & Juel 2004:504), yet it seems that this kind of text is still created by teachers and valued for phonics drill.
Here are further details of teachers' practice with examples.
Early in the year, Mrs Michaels used Ladder books to open teaching on the mat.These A4 booklets contained word lists ('ladders').Each learner in the group read a list, repeating each word three times.In this way, the day's lists were read through several times in each group, alternating individual and unison reading.Later in the year, Mrs Michaels added Yellow books, a collection of paragraphs each developed around a phonics sound: The bewildering illogic of the text removes any possibility that learners might read for meaning.Only simple retrieval questions could be asked of these texts and after reading the example above, Mrs Michaels asked learners about the colour of the bags, how many rats were in it and who held the bag.
Mrs Michaels and Mrs Samuel also used Reading is fun (Smook 2008), together with the Ladder books, Yellow books or Flip files.Mrs Samuel asked learners to read the sounds on a page and then to read the sentences in unison.Mrs Michaels used the same sequence as she used in the Ladder books: each word repeated three times and then the sentences read either in unison or individually.Examples of sentences on the sound -ng: • 'Mom sent me to the king.' • 'I can sing a lot of songs.' • 'I will sing to the king.' • 'The king will ring his bell.' • 'Ding dong come along.' • 'The king is fond of songs.' • 'I wish I had a ring like the king.' • 'The king has a big ring.'(Smook 2008:6) After learners had read these sentences Mrs Michaels and Mrs Samuel did not ask questions but asked the learners to find words that rhymed.
In addition, Reading is Fun (Smook 2008) has two features which are problematic.Firstly, the words exemplify sounds alone and many will not be understood by six-year-olds.Book 1 includes 'sap, tod, tag, gag, rut, max, vex, tax, jig, fig, wit, quit, quill, zep, tuff, muff, ruff and yen' (tod and zep are not English words).Secondly, some sound differences, like the i for tin and i for ink, will be difficult for adult English home-language speakers to hear.I observed a lesson full of frustration when Mrs Michaels attempted to teach this difference.
Mrs Samuel used Flip files made of A4 pages with paragraphs constructed around a sound.These were similar to Mrs Michaels' Yellow books although she had greater success at creating a narrative whole: Sleepy the tall tree stood in the middle of the farmer's beetroot field.All the creepy crawlies would creep up to Sleepy to get out of the hot sun.Sleepy the tree hated feeling the creepy crawlies creep up his thick rough tree trunk.Day after day the creepy crawlies tickled his bark.This made him feel rather sleepy.One day the beetroot farmer pushed his wheelbarrow all the way across the beetroot field, only to stop right in front of Sleepy.The beetroot farmer grinned at Sleepy with rotten brown teeth.The beetroot farmer grabbed his axe out of his green wheelbarrow.This made Sleepy the tree very frightened.The creepy crawlies wanted to help poor Sleepy.They all jumped into the beetroot farmer's sheep skin hat and began to creep into his dirty brown hair.The farmer took off at an incredible speed.Sleepy the tree never saw the beetroot farmer again.(van der Mescht 2013) After they had read this passage, Mrs Samuel showed the learners cards with the ee sound, and then led them in unison reading.She explained the meaning of rotten, axe and incredible speed.A learner asked what a creepy crawly was.She ended by asking why the insects wanted to help Sleepy.
Mrs Samuel also used poetry books on the mat.These contained poems on weekly themes.They were used for unison and individual Reading on the Mat but were not discussed for effect or meaning.
In the last quarter of the year, Mrs Danes and Mrs Michaels omitted the word-recognition practice and went straight into reading.This suggests that for them the early emphasis on word recognition and phonics was temporary, although there was not a concurrent increase in discussion, teaching comprehension skills or silent reading and their questions remained at a retrieval level.
An analysis of activities on the mat, therefore, shows that teachers reduced reading for meaning in four intersecting ways.Firstly, decoding practice was added to a seating arrangement and pedagogy designed to model and promote comprehension.Secondly, adding decoding reduced the time learners spent reading continuous text, both silently and aloud.Thirdly, adding decoding as an introductory activity meant that teachers left out the introductory context building, high-level questioning and consolidating final discussion.Fourthly, activities with cards, Ladder Books, Yellow books and Reading is fun (Smook 2008) did not allow for interactions beyond retrieval questions.The combined result of these changes was that learners' engagement with continuous narrative text was significantly reduced, every day of Grade One, as were teachers' opportunities for modelling comprehension.The consequences are discussed in further detail below.

What was left out? Text talk
The analysis showed that, in order to add phonics practice to reading instruction on the mat, the teachers left out most of the opportunities for text talk, the discussion designed to help young readers to construct meaning from text (Beck & McKeown 2001).This goes against the assertion in the curriculum that 'text talk between teacher and learners (and learners with each other) is central to this approach' (DoE 2011:12).Fountas and Pinnell explicitly disagree with including phonics instruction in Group Guided Reading and with any approach that 'requires teaching one item (a word or a letter for example) at a time in a tightly controlled sequence.Such tight control reduces children's opportunities to put together the process ' (1996:157).The Department of Education (DoE 2011) and its accompanying handbook (DoE 2008) do not mention phonics instruction as part of Group Guided Reading.
Not only did the introduction of texts for word recognition and phonics practice reduce the time teachers could spend on the meaning of any whole text, but the texts they introduced also precluded discussion.This is most clearly the case with flash cards.The various other texts were difficult to understand as a coherent whole, as the examples above show.
In addition, picture talk, which uses the semiotic of images as cognitive preparation for or support of reading was impossible during word recognition with cards and could not be included when texts like the Ladder books or Reading is fun (Smook 2008) were used.Drawing attention to visual cues enables learners to bring meaning into the text from the images, as well as to gain meaning from it through decoding.The same benefits apply to opening a discussion of the title and predicting what the text might be about and these must also be omitted when the text has no narrative coherence.Collins asserts that 'an instructional process that consists primarily of learners reading in a word-by-word fashion and teachers providing isolated decoding cues will leave the beginning readers without much practice in applying their knowledge of spoken language to the task of reading ' (2006:136).

Questions during and after reading
Although each teacher was recorded using questions at all levels of cognitive challenge when teaching the whole class, retrieval questions dominated all other devices for engaging with text on the mat (van der Mescht 2013).Teachers in the study also omitted the concluding discussion required by the curriculum (DoE 2011).The purpose of this final discussion is to return to points that were raised in the introductory discussion.The concluding discussion confirms learners' developing understanding and strengthens their perceptions of how to arrive at meaning.It is an opportunity for readers to evaluate the whole text, make genre-related comments, give opinions, form judgements and develop their tastes.These are all higherorder thinking skills essential to their early development as text participants, users and analysts.In the concluding discussion teachers may introduce a critical perspective, suggest comparisons or ask for comments on characters, their motivations and any moral issues.At this point, learners should be invited to ask questions for clarification and start developing their taste in genres.However, during the year of this research I recorded only two questions from learners related to the text, suggesting that learners in these classrooms do not see themselves as users or critics of text.
The consequence of these omissions on the mat was to accentuate reading as a decoding task, offering learners on the mat the role of code breakers alone, rather than presenting reading as a project in meaning making or individual pleasure.In this regard, the Progress in International Reading Literacy Study suggests that South African learners may not be given adequate opportunities to answer higher-order questions (Howie et al. 2007;Howie et al. 2017).It appears from this analysis of teacher practice that this may be the case.

Sustained silent reading
Perhaps the most significant feature of practices on the mat is that silent reading is omitted at the end of the sessions.
Only Mrs Danes actively promoted sustained silent reading on the mat, and she encouraged learners to use the reading corner and to read at their desks.Mrs Samuel promoted silent reading at other sites, but not on the mat.Mrs Michaels did not promote silent reading either on the mat or at other sites in the classroom during the year of the research.

Discussion
The clear consequence of the activities introduced to and omitted from Reading on the Mat is to reduce reading for meaning by elevating decoding practice.There must be reasons for these changes, as well as benefits for teachers, otherwise they would not find it worthwhile to develop and print the additional text resources.I offer the following interpretations of teachers' practice.
A first possible reason is that skills-based literacy teaching which focused on rapid decoding was entrenched in the nineties (Bloch 1999;Flanagan 1995), and adding phonics to Group Guided Reading may be evidence of the tenacity of a particular teaching practice.
A second possible reason is that Outcomes-Based Education, implemented in South Africa in 1998 as Curriculum 2005, emphasised group work.This may have encouraged teachers to use the seating arrangement for Group Guided Reading in a more general way, as Mrs Samuel did.Mrs Samuel used all the texts available in her classroom for both whole-class instruction and on the mat.In addition, groups also come to the mat for numeracy teaching and other instruction.It is possible that Group Guided Reading may be viewed by teachers as just another seating arrangement for which they have developed their own methods in preference to following curriculum requirements.
Thirdly, there appear to be benefits for the teachers.A first benefit is that focusing on decoding enables teachers to manage multilingualism in a way that promotes their learners as effective readers, as they do in interviews, staff meetings and personal conversations, and in consequence to present themselves as effective teachers.Sounding out letters and words and putting words in sentences provides teachers with a visible performance of their learners' skills, while fluent decoding suggests that learners are reading well.On the other hand, the processes of comprehension cannot be assessed through simple performance, and success is therefore harder to monitor than reading aloud (DHET 2021).A serious consequence in later grades is that proficient decoding is disconnected from comprehension.For example, at Greenbanks a Grade Two reader could read hydrogen and aeroplane but could not guess what either word meant from their context.
For a related reason, teachers may avoid sustained silent reading and emphasise decoding activities to regulate young learners' behaviour.Garan and Devoogd (2018:341) observed that teachers find sustained silent reading in classrooms unnerving because it means giving up control.
They go on to comment that 'loss of control can be daunting, particularly in the present climate of standards, mandated curricula, and accountability' (Garan & Devoogd 2018:341).
There is no doubt that decoding practice is an efficient way of keeping learners busy on the mat and of maintaining discipline there.
A third but related benefit is that reducing discussion and sustained silent reading allows teachers to finish Group Guided Reading relatively quickly.Reading on the Mat took place throughout the day.With five or six groups in each class, and each session on the mat separated from the next by whole-class activities, there was considerable time pressure to complete each session.Although the literature and curriculum suggests 15 min -20 min for each group, Reading on the Mat usually took more than 30 min for each group and would have taken longer if the focus had been on higher-order questions and discussion.The advantage of decoding practice is that it is relatively fast; the disadvantage of discussion, questions at a higher cognitive level and sustained silent reading is that it must progress at the pace of the learners and in Grade One this will be relatively slow.

Conclusion
The analysis presented in this article demonstrates that the choice of learning and teaching materials plays crucial role in promoting or limiting the opportunities offered to emergent readers.The roles of a reader which promote understanding, analysis, or critique of text may not be possible during Reading on the Mat as it was observed in these classrooms, even though they exemplify the most advantaged learning environments in South Africa.Moreover, teaching comprehension skills or reading for meaning did not play a significant role in any of the other methodologies used for engagement with text in these three classrooms.
Reading, as it is presented in Group Guided Reading, requires a pedagogy focused on narrative text, as well as richly resourced classrooms.By introducing texts that were not a flowing narrative whole, the teachers were unable to involve learners in discussion, or to ask higher-order questions.
Jordan and Henderson assert that 'the basic premise is that artefacts and technologies set up a social field within which certain activities become very likely, others possible, and still others very improbable or impossible ' (1995:44).Even when lavishly supplied with narrative texts, teachers may change a methodology and disrupt its intended learning value.Books or cards are not neutral materials as they both allow and disallow interaction, as the analysis of Group Guided Reading shows.
If the model of reading supplied to learners is one that reduces comprehension and values decoding above it, it is possible that these learners will lose early opportunities to learn that text is for understanding, or that they will learn it too late.In the teaching practices observed in this study, the introduced texts cannot be queried, compared, considered or discussed.This and similar practices, if ubiquitous in South African schools, or if unintentionally promoted in teacher training, may well promote readers who cannot perform well in national and international assessments of reading which require high-order insights into narrative and informational texts.Dixon asserts that in the Gauteng classrooms of her study: 'The emphasis on skills like decoding and encoding texts rather than meaning-making constructs a limited literate subject ' (2007:ii).Observations of teachers' practices in a different province suggests a similar finding.
Michaels: Each learner reads a section of text while others follow in their books.Individual and unison reading alternate.She observes their reading closely.She asks one comprehension or retrieval questions of each learner.No silent reading recorded.Mrs Samuel: Each learner reads a section of text while others follow in their books.She observes their reading closely.Learners may read in unison.No silent reading recorded.
Come mom come.See the rat.Dad has one too.Dad has the red bag.Come mom come.Dad has ten cats.Come bring the dog for mom to pat.The red peg is for Meg.Come mom come see my pet rat.Dad has one too.Bring a bag for the rat.Have you got the bag?I see two bags.One rat for Pat too.(van der Mescht 2013)

TABLE 1 :
Language profiles of learners.

TABLE 2 :
Comparison of Group Guided Reading, curriculum recommendations and teachers' practice.
The most frequently used question word was what … ?which asked each learner to remember a single fact from the text.Mrs Danes used this form in 38% of all questions, Mrs Michaels in 35% of her questions and Mrs Samuel in 47%.The question tag does … ?followed simple retrieval questions in frequency.Teachers only required a Yes/No answer to these retrieval questions.Mrs Danes used does … ? in 31% of all her questions, Mrs Michaels in 24% of her questions and Mrs Samuel 43%.Possible reasons for this deliberate move to lower-order questions on the mat are presented in the conclusion to this article.