Analysing pedagogy: the problem of

framing

Ursula Hoadley

Abstract

The sociologist Basil Bernstein presents a delicate and rigorous conceptual frame for
researching pedagogy, which enables an analysis of transmission and acquisition in relation
to social class. Bernstein’s theoretical project demonstrates how class relations generate
and distribute different forms of communication and ways of making meaning which
differentially position subjects with respect to schooling and its requirements. The purpose
of this article is to interrogate the use of Bernstein’s theory in analysing pedagogy, in
particular in relation to the two key concepts of classification and framing which underpin
his theory. The article considers the application of the theory in the South African context,
and the emergence of empirical texts that ‘fall out’ of the theoretical frame. The
development of the theory in relation to these texts is consequently explored.

The article is located within a broader study addressing the reproduction of social class
differences through pedagogy (Hoadley, 2005). The research was conducted in South
African primary schools in 2004. Drawing on a range of data, including classroom
observation, interview and student task data, the study sought to develop a framework for
the analysis of pedagogic variation across social class school settings, and to show how
inequalities are potentially amplified through the pedagogic practices found in classrooms.

Bernsteinian studies of pedagogy have asserted the centrality of the evaluative

criteria in identifying pedagogic practice optimal to promoting success in
schooling for, especially working class, students. The evaluative criteria
specify the requirements for students’ production of the legitimate text.
Making the evaluative criteria explicit involves “clearly telling children what
is expected of them, of identifying what is missing from their textual
production, of clarifying the concepts, of leading them to make synthesis and
broaden concepts and considering the importance attributed to language as a
mediator of the development of higher mental processes” (Morais, Neves and
Pires, 2004 , p.8). A number of careful empirical studies have emphasized
making the evaluative criteria explicit as a key variable in optimising working
class students’ success in school (for example, Rose, 2004; Fontinhas, Morais
and Neves, 1995; Morais, 2002). It is framing, or control, over the evaluative
criteria that is particularly interrogated in this article.
16 Journal of Education, No. 40, 2006

The article begins with a discussion of Bernstein’s theory, focusing

particularly on classification and framing, and the concepts that are used later
in the discussion of data and theory development. An analysis of classroom
observation data is then presented, and the use of the theory in relation to that
data is considered. The article concludes with some reflections on the
implications of the analysis presented.

Bernstein and pedagogy

For Bernstein, education specializes consciousness with respect to school

ways of organising experience and making meaning, or what has been referred
to elsewhere as context-independent meanings (Holland, 1981), elaborated
code (Bernstein, 1975), or a ‘school code’ (Taylor, Muller and Vinjevold,
2003). Whereas everyone has access to the commonsense knowledge of
everyday life, schooling inducts learners into the ‘uncommonsense’
knowledge of formal education - or, the school code. Bernstein talks about this
process in terms of the specialization of ‘voice’, which refers to the way in
which “subjectivity is created through the socialization of individuals into
categories of agents, knowledge and contexts that are distinguished by the
particularity of their voice” (Dooley, 2001, p.77). ‘Specialization’ then
“reveals differences from, rather than commonality. It means that your
educational identity and specific skills are clearly marked and bounded”
(Bernstein, 1975, p.81).

The purpose of schooling then is to specialize learners’ voice with respect to

the school code. Put another way, pedagogy in this view inducts learners into a
‘school’ way of organizing experience and making meaning. This essentially
entails the transmission and acquisition of context-independent meanings, or
an ‘elaborated code’.

How do we identify the specialization of voice, or more precisely, whether the

transmission is doing the work of specializing? Conventionally, in the
Bernsteinian literature, specialization of voice is adumbrated in terms of
classification and framing values – particular combinations of these
dimensions give rise to variations in the form of the elaborated code.

In summary then, the specializing of consciousness (and this generally will

entail the acquisition of context-independent means of organizing experience
and making meaning) happens through two key mechanisms which are at the
heart of Bernstein’s theory: classification and framing, which refer,
respectively, to power and control.
 Hoadley: Analysing pedagogy: the problem of framing 17

Bernstein’s theory of pedagogy is encapsulated in his theorizing of pedagogic

discourse. Pedagogic discourse describes the specialized form of
communication whereby differential transmission and acquisition is effected
(Bernstein, 1990). Pedagogic discourse describes the relay of pedagogy. It
consists of an instructional discourse embedded in a regulative discourse,
where the instructional discourse is concerned with the transmission/
acquisition of specific competences, skills and knowledge, and the regulative
discourse is about the underlying theory of pedagogy and the expectations of
character, conduct and manner. Put simply, instructional discourse is
concerned with the transmission of knowledge and skills; regulative discourse
is akin to the ‘hidden curriculum’.

Classification and framing

Bernstein provides a language for the description of pedagogic discourse

through the concepts of classification and framing. At a higher level of
abstraction, classification refers to the social division of labour. At the macro
level classification generates categories of agents and discourses: the
categories or insulations are instantiations of power. At the micro level,
classification is about the organizational or structural aspects of pedagogic
practice. Classification is about relations between, and the degree of
maintenance between categories, and these include the boundaries between
agents, spaces and discourses.

Classification is expressed as being strong (where boundaries are explicit and

categories are insulated from one another), or weak (where there is integration,
or where the boundary is weak or blurred). In terms of discourses, the relations
between different subject areas (inter-disciplinary relations), and between
school knowledge and everyday knowledge (inter-discursive relations) are
considered, as well as the relation between knowledge within a particular
subject area (intra-discursive). With respect to agents, the theory identifies
how teachers and learners’ pedagogic identities are demarcated.

Where classification at the macro level is related to the social division of

labour, framing refers to social relations within this social division. That is,
specific social relations in production/reproduction generate particular
practices which we can talk about in terms of framing, or control relations.
Framing, therefore, refers to relations within (within boundaries). Framing, in
a sense, supports classification, it produces ‘the animation of the power grid’
(Hasan, 2002), but it also opens up the potential for the change of boundaries,
the contesting of power relations. It is through interaction (framing) that
boundaries between discourses, spaces and subjects are defined, maintained
and changed.
18 Journal of Education, No. 40, 2006

There is, therefore, a crucial relation between classification and framing. It is

framing (control) which contains within it the making and the unmaking of the
classification (power). It is in the distinction between power and control that
Bernstein allows the intentional and structural aspects of power (Atkinson,
1985) of conventional sociological theories to co-exist, and operate
dialectically.

At the micro level of pedagogic practice, framing refers to the location of

control over the rules of communication. “Framing refers to the degree of
control teacher and pupil possess over the selection, sequencing, pacing and
evaluation of the knowledge transmitted and received in the pedagogical
relationship” (Bernstein 1975, p.88). Conventionally, framing has to do with
the way in which the relationship between the teacher and the learner is set up,
where strong framing refers to a limited degree of options for students, and
weak framing implies more ‘apparent’1 control by learners. Again, framing is
expressed in terms of its strength or degree of control. Strong framing would
imply that students have limited control over the ‘relations within’ and a
limited degree of control over the sequencing, pacing, selection and evaluation
of the knowledge transmitted.

There are, however, difficulties in working with the concepts of classification

and framing empirically. Because they are dialectically linked, to ‘see’ them
separately poses a challenge for the researcher. Classification cannot maintain
itself without framing. Thus, instances of the classification relation are evident
only through the framing relations, the interactional. The interactional and the
organizational are dialectically linked, and empirical instances of one always
imply the other. Framing is, after all, defined as pedagogic discourse, which
Bernstein (1996, p.28) sets out as:

instructional discourse ID
Framing = ––––––––––––––––––
regulative discourse RD

Classification is in a hierarchical relation to framing, it is prior, but it is empty

without the mechanisms to achieve the boundary – that is, framing. Hence the
dialectic. The difficulty is also alluded to in Bernstein (1996, p.19), where he
argues that “power and control are analytically distinguished and operate at

1
Because Bernstein privileges a particular definition of pedagogy which is hierarchical, and
where the transmitter is in possession of the rules for evaluation, learner control over the
discursive rules of pedagogic practice must be ‘apparent’. This would also explain why the
rule for regulating the conduct of transmitters and acquirers is the ‘hierarchical’ rule
(Dooley, 2001, p.61).
 Hoadley: Analysing pedagogy: the problem of framing 19

different levels of analysis. Empirically, we shall find that they are embedded
in each other.”

The relationship between everyday knowledge and school knowledge is a case

in point. Bernstein would describe these differences in terms of framing:
“Thus we can consider the variations in the strengths of frames as these refer
to the strength of the boundary between educational knowledge and everyday
community knowledge of teacher and taught” (Bernstein, 1975, p.206).
Atkinson (1985), referring to the same quotation, has argued that, in practice,
“this latter aspect of boundary seems equally a matter of classification and
frame, since it is often related directly to the relative purity and strength of the
membrane of curriculum contents. Empirical research tends to reflect this
overlap and ambiguity” (Atkinson, 1985, p.136).2

In relation to framing, Bernstein asserts that “control is double faced for it

carries both the power of reproduction and the potential for its change”
(Bernstein, 1996, p.19). The distinction between power and control, unique in
the discipline of sociology, thus allows for the description of the making and
potential unmaking of the social reproduction of inequality, one of the key
concerns of Bernstein and those who draw on his work.

Bernstein (2000, p.100) provides a taut formula for classifying codes in terms
of the different dimensions and values outlined above:

E
+-C ie/+-F ie

Here E stands for orientation to meaning – elaborated, or what was referred to

earlier as the school code. The line stands for the embedding of the orientation
in classification and framing values. Variation in these classification and
framing values gives rise to different modalities of pedagogic practice
(Bernstein, 2000).

Classification and framing describe the structural and interactional aspects of

pedagogic practice, exposing the power and control relations that inhere in

2
Thus, for example, classification can be used to describe the way in which knowledge is
organized and the relationship between educational knowledge and everyday knowledge.
When talking about the extent to which teacher and learner control selection of content (i.e.
the recruitment of everyday narratives, for example), this could constitute framing over
selection, not classification. When framing over selection is weak, this has an effect on
classification in that, at least temporarily, it reduces the specialty of the pedagogic
discourse.
20 Journal of Education, No. 40, 2006

pedagogic practice. These concepts are connected at both macro and micro
levels to a set of related concepts which allow for the analysis of the workings
of power and control, in particular in relation to transmission and acquisition
processes.

Classification provides the key to distinguishing contexts. It is classification

which orients the speaker to what is expected and what is legitimate given the
context. It provides the recognition rules for what is required. Framing
regulates how legitimate meanings may be put together and made public,
crucially through the evaluative criteria. These are the realization rules,
referring to how an appropriate text may be realized. The related concepts are
summarized in diagrammatic form below.

Figure 1: Classification and framing and related concepts (adapted from

Dowling, 1999, p.9; Bernstein, 1990, p.31 and Bernstein 1990, p.38)

Power Control
Social division of labour (structural) Social relations (interactional)
Classification Framing
Relations between Relations within
Recognition rules Realization rules

The point is that there is a differential distribution of power and control

relations across different social classes, and these produce different practices
and forms of consciousness. It is through the codes that we see the differential
positioning of subjects of different social class groupings, dominant and
dominated. Bernstein poses the question in this way:

What we are asking here is how the distribution of power and the principles of control are
transformed, at the level of the subject, into different, invidiously related, organizing
principles, in such a way as to create the possibility of change in such positioning (1990,
p.13).

And he answers his own question like this:

The broad answer given by this thesis is that class relations generate, distribute, reproduce,
and legitimate distinctive forms of communication, which transmit dominant and dominated
codes, and that the subjects are differently positioned by these codes in the process of
acquiring them (1990, p.13).
 Hoadley: Analysing pedagogy: the problem of framing 21

Applying the theory: the social reproduction of

inequalities through pedagogy

The broader study referred to earlier (Hoadley, 2005) sought to develop a

model that showed the mechanisms whereby social differences are reproduced
through pedagogic practice. The main data source in this regard was classroom
observation data, collected across three days in eight Grade 3 teachers’
classrooms, teaching in different social class school settings in the greater
Cape Town area. Literacy and mathematics lessons were extracted from the
data for analysis.

The analytic frame consisted of a number of dimensions, and considered the

instructional form (drawing on Pedro, 1981), and the strategies deployed by
teachers (drawing on the work of Dowling, 1998), and the structure of the
pedagogy (using the work of Bernstein, 1975; 1990; 2000). It is the latter
aspect of the analysis, and the use of the concepts of classification and
especially framing, that are considered in this article.

In relation to the analysis of the classroom observation data, classification and

framing addressed two dimensions of variation across the different
classrooms: classification identified the structuring of discourses, spaces and
agents (the ‘what’ and ‘who’ of pedagogy), and framing, described the relative
control teachers and learners had over selection, sequencing, pacing,
evaluation and hierarchical rules (the ‘how’ of pedagogy). These conceptual
dimensions are summarized in Figure 2 on the next page:
22 Journal of Education, No. 40, 2006

Figure 2: Conceptual categories for characterising pedagogy

Discursive rules Extent to which teacher controls selection of content

Extent to which teacher controls sequencing of content
Extent to which teacher controls pacing of content
Framing
Extent to which teacher makes explicit the rules for
evaluation of learners’ performances
Hierarchical rules Extent to which teacher makes formal or informal the social
relations between teacher and learners

Relations Inter-disciplinary (strength of boundary between the subject

between area and other subject areas)
discourses
Inter-discursive (strength of boundary between the subject
area and everyday knowledge)
Relations between Teacher–learner (strength of demarcation between spaces
Classification
spaces used by teachers and learners)
Space for learning (strength of boundary between space,
internal and external, to the classroom and learning)
Relations Teacher–learner (strength of demarcation of pedagogic
between agents identities)

For the broader study (Hoadley, 2005), a coding instrument broadly following
the work of Morais and Neves (2001), was designed to assign values, in terms
of framing, to the discursive rules of pedagogic practice: the selection,
sequencing, pacing and evaluative criteria of educational knowledge. The
coding instrument also assigned values to the hierarchical rules (the extent to
which teacher and learner have control over the order, character and manner of
the conduct of learners). The instrument also considered discourse relations in
terms of the strength of classification (or boundedness) between different
subject areas (inter-discursive) and between school knowledge and everyday
knowledge (inter-discursive). The instrument further looked at the
classification of spaces and agents.

In the coding instrument, the high-level concepts of classification and framing

were translated into a coding scheme to read the data. The indicators, or
theoretical constructs, named empirical instances of particular abstract
 Hoadley: Analysing pedagogy: the problem of framing 23

concepts. The coding scheme comprised 19 indicators, providing a means for

making the conceptual categories shown in Figure 2 ‘observable’. An example
of one of the indicators (No.5), for the framing of the evaluation criteria, is
given in Figure 3 below.

Figure 3: Extract from the coding instrument for analysing the classroom
observation data in terms of the classification and framing of
pedagogic discourse

Discursive rule EVALUATION CRITERIA (F +-)

The extent to which teacher and learner have control over the evaluative criteria of the
instructional knowledge pertaining to the meaning of concepts and principles and their
appropriate realisation

5. In the F ++ F+ F! F !! F0
course of
learners Evaluative Evaluative Evaluative Evaluative Transmission
conducting criteria very criteria quite criteria quite criteria very of evaluative
clear and clear and unclear and unclear and criteria not
an activity
explicit explicit implicit implicit observable
or task
The teacher The teacher The teacher The teacher The teacher
constantly makes some makes a few looks at a engages in
moves points, either comments few learners’ other work in
around and to the whole during the work when it her space and
monitors class or to course of the is brought to is not seen to
what learners individual task and her attention. look at what
are doing learners, so looks at She rarely or the learners
and makes as to clarify some of the never listens are doing.
comments. what is learners to them read. She makes
To the whole expected of work or She seldom no comment
class, and to them in the listens to makes a on the work
individuals, task. them read. comment to as it
she However, the learner. proceeds. No
repeatedly this is not Comments action is
goes over sustained are not taken to
what and the extended to ascertain
constitutes criteria for a the whole what the
an successful class. learners are
appropriate production doing.
performance. are not made
explicit to
all.
24 Journal of Education, No. 40, 2006

In Figure 3 framing is expressed in terms of its strength or weakness using

standard Bernsteinian notation – F ++ representing the strongest framing (or
teacher control) over the evaluation criteria and F -- representing very weak
framing. F 0 will be explained below. A sample of lessons was then coded
using the coding scheme. Each lesson was coded for 19 indicators (such as the
one above) for the different dimensions of pedagogy as offered by the theory.
In the course of this initial analysis changes were made to the coding scheme.
A similar process had already been undertaken with the pilot data and the
coding scheme, and this was repeated. The coding scheme in the end went
through at least six iterations. The ‘external language of description’
(Bernstein, 2000) – the indicators, or theoretical constructs – was thus
developed through iterations between the theory and the empirical data.

0
An analysis of text and the F value

Criticisms of the use of Bernstein’s work in South Africa have often referred
to the mechanical deployment of the categories, where “It would appear, from
the criticisms made, that we enter the field with categories shaped rather like
containers, into which we scoop our data!” (Ensor and Hoadley, 2004). Here, I
want to show, at some length, how the potential for theory development opens
up through analysis. This is a modest exemplar, but it does illustrate the
generative effects of the process of analysis in terms of the theoretical
concepts informing the study. In its specificity, Moore (2001) explains how
Bernstein’s theory holds the potential to avoid the circularity referred to
earlier:

By rigorously specifying in advance what we should expect to see if the theory holds, we
can measure the limitations of the theory if we fail to find it or encounter other things
unexpected and unspecified. In this way the theory can avoid the circularity that so often
characterises (and vitiates) research applications of theories lacking such methodological
depth (p.368).

Bernstein’s sociological theory of pedagogy presents the researcher with a

highly systematic account of how pedagogy works. The theory is worked out
with a rigour and precision that gives rise to an array of inter-related concepts
that have a delicacy and ‘methodological depth’ which is extremely useful to
the researcher.

However, Bernstein’s theoretical categories do not allow for a direct reading

of the empirical: a language of description is needed, and a significant amount
of work needs to be undertaken in order to bring the concepts closer to the
data for its reading. Here I illustrate an instance of where such work was done,
in relation to the conceptual category framing. I also reflect on some of the
 Hoadley: Analysing pedagogy: the problem of framing 25

difficulties raised earlier in working with the concept. I take an example from
classroom observation data generated in the broader study (Hoadley, 2005) to
show how the theoretical range F ++ to F -- for the evaluative rules failed to
capture certain pedagogical forms in the data. The example comes from a
Grade 3 literacy lesson in a classroom in a school in Khayelitsha. The children
in the classroom come from the surrounding working class community. The
lesson is one of a number focusing on the theme of trees.

The teacher stands at the front of the class and pages through the textbook. All the learners
have a copy of the textbook in front of them.

1 Teacher: ‘Here people, I like this section on leaves. We were learning about trees, neh? And
2 then went on to leaves.

The teacher goes on to explain what the book says about colours, that there are shades of
colours, for example, blue-green. She copies a set of leaves, which shows these colour
variations, from the textbook onto the board. However, she copies only the set of leaves, not the
colours. What the teacher has encountered in the textbook is the end of a previous section on
colours, which precedes a section on trees. The iconographic indicator – leaves – has led her
to select this page as leaves relate to the more general theme in use, trees. But the topic of trees
is only addressed halfway down the page. The lesson continues.

The teacher numbers the leaves she has drawn on the board and the learners shout out the
numbers as she writes them. The teacher then moves directly onto the next section in the book
on trees.

3 Teacher: He says here there are parts of the tree, that’s what I like, but then he says we don’t
4 tell colours as they are. So here are the parts of the tree. He says write them in their order from
5 the biggest to the smallest. Read these as I write them on the board.

The teacher writes on the board: tree, leaf, branch, bush, and addresses the learners.

6 Teacher: Which are found at the bottom of a tree?

7 Learner: Roots are found at the bottom of the tree.
8 Teacher: No no. Don’t tell me things you haven’t seen. I’m not asking for what you’ve thought
9 about, I’m asking for what you’ve seen. Okay, from the tree, bush, leaf and branch, which one
10 do you get from the bottom of the tree? Things that you get at the bottom. Bottom, bottom.’

The teacher underlines ‘bottom’ on the board. Another learner says roots. After a while the
teacher looks back at the text book and realises that she has made a mistake, reading ‘tree,
bush, branches, leaf’, instead of ‘stem, roots, branches, leaves’. She moves directly on to the
next question, which requires writing from biggest to smallest, tree, branch, leaf, bush.
Learners respond and the teacher writes each word on the board. She then returns to the
question of what is found at the bottom of the tree. As she writes, learners repeat the words
over and over again. The following exchange occurs as she moves onto the next section in the
book:
26 Journal of Education, No. 40, 2006

11 Learners: leaves, leaves, leaves, leaves, leaves, leaves leaves, leaves, leaves leaves, leaves,
12 leaves leaves, leaves, leaves
13 Teacher: Hey stop. The reason why we are repeating this is because you do your own thing
14 when I turn my back on you. Now the writer says the same words rhyme at the end. Now
15 we’ve done a tree. Haven’t we done a tree?
16 Learners: Yes Miss. Yes we’ve done it.
17 Teacher: Now we know how a tree is formed. Now the writer says there are certain words that
18 rhyme at the end. This is what I like. And he also says write those that rhyme in the box.
19 [Teacher looks at the book for a while] Ja, here’s work. Close your books. I’ll give you work on
20 the board. Don’t talk Grade 3. Don’t talk, don’t talk. Sleep on your desks. Lower your heads.

The teacher writes 6 words on the board: tree, fruits, home, flowers, smoke, bushes, roots.

21 Teacher: Listen, listen. I did not say shout on the top of the roof. Now write the rhyming words.
22 He [the author of the textbook] says some words are rhyming at the end so he wants you to
23 write those that rhyme at the end. Here’s the correct date, the thirty-first. Let’s write. Let’s
24 work. No talking. I want rhyming words. I want rhyming words. I want rhyming words.

Later the teacher bangs on her table with a ruler and shouts at the learners to be quiet.

25 Teacher: Write, write, even though you don’t know.

The teacher sits at her desk for the remaining 23 minutes of the lesson. At no point does she see
what learners are writing. The bell rings for break and learners close their books and go out.

A consideration of the coding of this lesson will be illustrative of the coding

procedure in general, and what I define as the F 0 category in particular.

In terms of the regulative discourse, the hierarchical rules would be coded F ++.
The control relation is generally about constraint and is based on the
teacher/pupil hierarchy, rather than an explication of rules or principles
underlying the control. In this imperative form (F ++) the acquirer is given no
options in responding to the control of the teacher, apart from an explicit
challenge to authority.

Selection and sequence in this instance would both be coded F ++. The reason
for this is that the teacher decides what knowledge will be transmitted and in
what order transmission will take place. Learners are not given opportunities
to alter the selection and the sequence of the knowledge, even where, at one
point (line 7) their interjections potentially are a corrective to the teacher’s
 Hoadley: Analysing pedagogy: the problem of framing 27

misreading of the text.3

The coding of the framing of pacing is slightly more complex. Although it is

clear that the pace of the lesson is extremely slow, it is the teacher who
controls this. Learners do not intervene when they have completed work well
ahead of time. It is the teacher who asserts the pacing or expected rate of
transmission. She decides that the lesson will continue until the bell rings, and
learners do not have control over the stipulated pacing. Pacing is therefore also
coded F ++.

A problem arises, however, in the coding of the evaluative criteria. The

required performance of learners ultimately is to copy down words that rhyme,
but no concept of rhyming is transmitted, and its recognition is potentially
opaque to the learners. Because the evaluative criteria have not been
transmitted, the teacher can only elicit the legitimate text from the learners on
the basis of assertion: ‘Write, write, even though you don’t know’ (line 25),
and that legitimate text appears devoid of instructional content. The learners
are required to write; what they write does not seem to matter.

Similarly, the earlier exchange in the same lesson unfolds as:

Teacher: Now we’ve done a tree. Haven’t we done a tree?

Learners: Yes Miss. Yes we’ve done it.

Teacher: Now we know how a tree is formed (lines 14–17).

Again, the legitimate text is extracted on the basis of assertion. The learners
had merely named parts of a tree up to this point; they had not addressed ‘how
a tree is formed’.

The framing of the evaluative criteria is difficult to categorise as either weak

or strong. Thus the category F 0 in the coding scheme has been developed in
order to capture such instances of transmission, which appear devoid of
evaluative criteria relating to the instructional discourse, or where these are

3
However, it could be argued that, in this case, the teacher in fact substitutes the textbook for
herself; or she recruits a proxy voice – the sequence and selection of the textbook – because
her voice isn’t able to do the pedagogic work. Neither student nor teacher here appears to be
controlling the knowledge but rather the textbook is followed to the word, strongly dictating
the sequence and selection. So an initial (iconic) selection in terms of the theme ‘trees’ is
made, but from there the sequencing follows that of the textbook from the top of the page to
the bottom.
28 Journal of Education, No. 40, 2006

obscured by regulative criteria.4 That is, the transmission of principles relating

to the subject knowledge are obscured. All is about comportment, form, or
behaviour. The descriptor for one of the indicators for F 0 for the evaluative
criteria reads:

Evaluative criteria: In the introduction/explanation/exposition to a topic/task

F 0 – It appears as if no attempt is made to transmit the concepts and principles in the
instructional practice. What counts as a successful production in terms of instructional
knowledge is therefore totally unclear. The purpose of the task/activity/discussion is
unclear. Learners are unclear as to how to proceed, or they are only given criteria relating to
how they should behave.

It is also evident from this example that the coding of the data can at times be
derived only in conjunction with an assessment of what learners recognize and
realize in their performances. This is because there must be certainty (using
this instance as an example) that the learners have not spent several prior
lessons focusing on rhyme, and that the absence of an explicit articulation of
the evaluative criteria could therefore be considered redundant. In such cases
reference can be made to learner productions and learner notebooks, or to
observation of learners carrying out the tasks, or to a consideration of a
‘teaching unit’: a series of lessons with a particular trajectory.

The evaluative criteria can also not be coded as extremely weak (F --), as this
would imply that it is the learners who have control over what the legitimate
text would constitute. The strong framing over the hierarchical rules and rules
of selection mitigate against this. It must be emphasized that F 0 does not lie on
the continuum (i.e. it is not extremely weak framing), but rather is a rupture. F 0
represents an inability to observe the code. It may also point to a breakdown in
pedagogic discourse, or the absence of (a particular dimension of) pedagogy.
Bernstein (2000) at one point does suggest the possibility of the F 0. In the
following quotation we recall that realization rules are transmitted and
acquired through framing relations and the recognition rules through
classification:

Many children of the marginal classes may indeed have a recognition rule, that is, they can
recognise the power relations in which they are involved, and their positioning in them, but
they may not possess the realisation rule. If they do not possess the realisation rule, they
cannot speak the expected legitimate text. These children in school, then, will not have

4
There is not an absence of control. The teacher asserts control here, but there is no evidence
of the teachers’ specialized voice. Without the evaluative criteria – that which specifies the
legitimate text, students are unable to recognise what is required, and the teacher does not
evaluate their productions according to instructional criteria. Consequently the evaluative
rules are coded F 0.
 Hoadley: Analysing pedagogy: the problem of framing 29

acquired the legitimate pedagogic code, but they will have acquired their place in the
classificatory system. For these children, the experience of school is essentially an
experience of the classificatory system and their place in it (Bernstein, 2000, p.17, my
emphasis).

Bernstein here acknowledges the possibility that one can be positioned within
the school (that is, one always gets sorted into the social relation) but have no
access to the realization rule. The question arises as to how one codes framing
values in such a pedagogy. Here I propose the necessity of including an F 0
value for the framing over the evaluative criteria where the absence of being
able to observe the pedagogic code for instructional discourse is confronted. In
a sense, it represents a collapse of the instructional discourse into the
regulative discourse.

Dowling (1999) develops a detailed critique of framing. His response to the

difficulties alluded to above is to dispense with framing. He does this because
he does not operate with a notion of boundary. Classification for him refers to
degrees of specialization rather than strength of insularities, and his project is
to analyze the contents of the classification rather than elucidate its structure. I
would argue, however, that the distinction between classification and framing
in the analysis of the reproduction of differences through pedagogy and the
specializing of consciousness is crucial. This is because the theoretical
underpinnings of these concepts in power and control, and the social division
of labour and social relations, allow for a consideration of the making and the
unmaking of power.

This brings us back to the power (classification) and control (framing)

dialectic. If the framing (over instructional content in the evaluative rules) is
absent, the potential for the unmaking of the classification is compromised. In
the extract we see a weak classification of school knowledge – the labeled
parts of the tree are not related to the formation of the tree, the words are not
interrogated in terms of the concept of rhyming. There is a predominance of
everyday activity in naming, and conceptual elements are subordinated to the
theme of trees. It is through the evaluative rules that we would see the
strengthening (or weakening) of these boundaries. This would entail a
specification of what it is that learners should know, do or understand with
respect to the instructional knowledge. Without this there can be no change in
the classification. One is merely sorted. One is stuck.

When there is an apparent absence of evaluative criteria, what then is

sequenced, selected and paced? Is it possible that F 0 in fact represents an
absence of pedagogy, and should therefore be bracketed off from an analysis
of classification and framing? The reluctance to do this is based on what we
know from classroom research – that this is a form of pedagogy (or non-
30 Journal of Education, No. 40, 2006

pedagogy) not uncommon in schools in South Africa (see especially Ensor et

al, 2004; Hoadley, 2003; Jacklin, 2004). Within the same analytic framework,
therefore, I have attempted to capture more conventional understandings of
what pedagogy entails, and that which appears to fall outside these
understandings, or ‘pedagogic breakdown’.

But there is not an absence of transmission activity. It is still necessary to

elucidate what it is that is transmitted, if not the evaluative rules. This is where
the theory needs to be either extended or considered in relation to alternative
explanatory frameworks. In the broader project (Hoadley, 2005) I drew on the
work of Dowling (1998) to look at the contents of the meanings transmitted.
His notions of localizing and specializing strategies in particular allowed for a
consideration of the relationship between what was transmitted and the
domain to which it referred. Knowledge in this analysis draws its authority
either from the public domain (broadly everyday meanings), or the esoteric
domain (disciplinary understandings). Unlike Dowling, however, I do not
eschew framing. In this particular case, framing allows for the analysis of what
it is that is absent in the structure of the pedagogy, and what the implications
are with respect to power and control relations.

To return now to the extract and its analysis, it is not possible to draw
conclusions on the basis of a single lesson, but were we to find this patterning
of pedagogy across a series of lessons, weeks, a year, we could then say that
the potential for the specialization of the students’ voice is seriously
undermined in the pedagogy. The students’ potential for acquiring the school
code is compromised.

The general methodology for operationalizing the concepts of classification

and framing broadly follows the work of Morais and Neves (2001); Morais,
Neves and Pires (2004); and more generally the work of the Sociological
Studies of the Classroom project at the University of Lisbon (ESSA).
However, the original theory of classification and framing was developed in
contexts of schooling that were possibly far more functional, in conventional
notions of the working of schools, than many schools found in South Africa.
Further, the theory and its application to classroom observation data was
extended and developed largely in Portugal with respect to science, two
contexts which are in all probability far more strongly classified than
phenomena that arise in other contexts (for example, literacy learning in South
Africa, shown in the example above).

Thus, in order to capture pedagogy of very different types on the same scale, it
was necessary to extend that scale to include a greater range of pedagogic
forms, and this emerged in relation to the framing of the evaluative rules. In
this way the original theory begins to open up both the possibilities and the
 Hoadley: Analysing pedagogy: the problem of framing 31

exigencies for further theorizing. However, it was necessary to start with a

sound theory in order to make this visible. “Without a model, the researcher
can never know what could have been and was not” (Bernstein, 2000, p.135).

Discussion

What does classification and framing offer? It shows the inner logic of
pedagogy and reveals the structuring of inequality with respect to different
groups of pupils. We have known for a long time that schooling reproduces
inequalities, or at the least interrupts these in very limited measure. We need to
know exactly how. Why, beyond academic interest? A number of studies have
attempted to make explicit pedagogic modalities which optimise learning for
working class children.

The on-going work of ESSA (for example, Morais and Miranda, 1996; Morais
and Neves, 2001; Morais, Neves and Pires, 2004) has focused on the micro
processes in the classroom to explore the “relations present in the context of
reproduction of the pedagogic discourse” (Neves, Morais and Afonso, 2004,
p.280). The various authors show that specific aspects of pedagogic practice
are favourable to the development of the elaborated coding orientation
required by the school (Fontinhas, et al., 1995). The work of ESSA comprises
action research, and more effective pedagogic modalities, derived from the
research, are designed and tested with learners from different social class
backgrounds. Teachers are thus explicitly trained to teach particular modalities
of elaborated code.

Morais (2002) summarizes some of the results of the empirical work of the
ESSA, explicitly defining what values of classification and framing, along
which dimensions, proved optimal for enhanced achievement of working-class
students. Consistent with all of the ESSA work, Morais (2002) again stresses
“explicating the evaluative criteria as the most crucial aspect of a pedagogic
practice to promote higher levels of learning of all students” (p.568).

In particular, for students from lower social groups, the explication of the
evaluative rules, and weak framing over pacing, creating the opportunity for
students to intervene in the expected rate of their acquisition, are those aspects
identified as being most crucial in facilitating their access to school learning.
Likewise Rose (2004), in his research into literacy pedagogy for ‘indigenous
learners’, specifies precisely the dimensions facilitating a weakening of the
negative relation between social class and educational achievement: a
weakening of the framing of pacing and sequencing rules, and a weakening of
“the framing regulating the flow of communication between the school
classroom and the community the school draws on” (p.106).
32 Journal of Education, No. 40, 2006

In the light of these findings, F 0 for the evaluative criteria is a particularly

devastating indictment on what opportunities, especially working class,
children have for learning in some South African schools. There is an opening
to explore further, in the South African context, what the optimal variation
along the pedagogic dimensions offered by the theory, are for working class
children here.

Further, and particularly in relation to the South African context, an analysis

of pedagogy which specifies its intricate dimensions allows us to think more
carefully about the external regulation of teaching. If we are aware that there
are many teachers whose voice is inadequately specialized for the task of
transmitting school knowledge, then questions are raised as to what form of
teacher training we advance, and what type of curricula and textbooks we
construct. In the latter case, if teachers recruit the textbook as a proxy voice, it
is necessary to consider the level of specificity in particular in relation to
conceptual knowledge contained in textbooks. We need to question how it is
that we can design curricula most likely to advantage those that most need
teachers to make explicit what the criteria are for their success at school.
Bernstein’s theory allows us to pay close attention to how power structures
and how control makes and unmakes the categories into which learners are
sorted.

References

Atkinson, P. 1985. Language, structure and reproduction: an introduction to

the sociology of Basil Bernstein. London: Methuen.

Bernstein, B. 1975. Class, codes and control volume 3: towards a theory of

educational transmissions. London: Routledge and Kegan Paul.

Bernstein, B. 1990. Class, codes and control, volume 4: the structuring of

pedagogic discourse. London: Routledge.

Bernstein, B. 1996. Pedagogy symbolic control and identity: theory, research,

critique. London: Taylor & Francis.

Bernstein, B. 2000. Pedagogy, symbolic control and identity: theory, research

and critique. Revised edition. Oxford: Rowman & Littlefield.

Dooley, K.T. 2001. Adapting to diversity: pedagogy for Taiwanese students in

mainstream Australian secondary school classes. Unpublished PhD thesis.
Brisbane: Griffith University. Available at: http://www4.gu.edu.au:8080/adt-
root/uploads/approved/adt-QGU20030102.105906/public/02Whole.pdf
 Hoadley: Analysing pedagogy: the problem of framing 33

Dowling, P. 1998. The sociology of mathematics education: mathematical

myths/pedagogic texts. Falmer: London.

Dowling, P. 1999. Basil Bernstein in frame: ‘Oh dear, is this a structuralist

analysis?’. Available at:
http://www.ioe.ac.uk/ccs/dowling/kings1999/index.html

Ensor, P. and Hoadley, U.K. 2004. Developing languages of description to

research pedagogy. Journal of Education, 32, pp.81–104.

Ensor, P., Dunne, T., Galant, J., Gumedze, F., Jaffer, S., Reeves, C. and
Tawodzera, G. 2002. Textbooks, teaching and learning in primary
mathematics classrooms. In Lubisi, C. and Malcolm, C. (Eds), Proceedings of
the 10 th SAARMSTE Conference. Pietermaritzburg: University of Natal.

Fontinhas, F., Morais, A.M. & Neves, I.P. 1995. Students’ coding orientation
and school socializing context in their relation with students’ scientific
achievement. Journal of Research in Science Teaching, 32(5): pp.445–462.

Hasan, R. 2002. Semiotic mediation, language and society: three exotropic

theories – Vygotsky, Halliday and Bernstein. Presentation to the Second
International Basil Bernstein Symposium: Knowledges, pedagogy and society.
Cape Town, July.

Hoadley, U.K. 2003. Time to learn: pacing and the external regulation of
teachers’ work. Journal of Education for Teaching, 299(3): pp.47–62.

Hoadley, U.K. 2005 Social class, pedagogy and the specialization of voice in
four South African primary schools. Unpublished PhD thesis. Cape Town:
University of Cape Town.

Holland, J. 1981. Social class and changes in orientations to meaning.

Sociology, 15: pp.1–18.

Jacklin, H.J. 2004. Repetition and difference: a rythmanalysis of pedagogic

practice. Unpublished PhD thesis. Johannesburg: University of the
Witwatersrand.

Moore, R. 2001. Obituary: Basil Bernstein 1924–2000. British Educational

Research Journal, 27(3): pp.367–370.

Morais, A. 2002. Basil Bernstein at the micro level of the classroom. British
Journal of Sociology of Education, 23(4): pp.559–569.
34 Journal of Education, No. 40, 2006

Morais, A.M. and Miranda, C. 1996. Understanding teachers’ evaluation

criteria: a condition for success in science classes. Journal of Research in
Science Teaching, 33(6): pp.601–624.

Morais, A. and Neves, I. 2001. Pedagogical social contexts: studies for a

sociology of learning. In Morais, A., Neves, I., Davies, B. and Daniels, H.
(Eds). Towards a sociology of pedagogy: the contribution of Basil Bernstein
to research. New York: Peter Lang.

Morais, A., Neves, I. and Pires, D. 2004. The what and the how of teaching
and learning. In Muller, J., Davies, B. and Morais, A. (Eds). Reading
Bernstein, researching Bernstein. London: RoutledgeFalmer.

Neves, I., Morais, A. and Afonso, M. 2004. Teacher training contexts: study of
specific sociological characteristics. In Muller, J., Davies. B. and Morais, A.
(Eds). Reading Bernstein, researching Bernstein. London: RoutledgeFalmer.

Pedro, E.R. 1981. Social stratification and classroom discourse: a

sociolinguistic analysis of classroom practice. Stockholm: Stockholm Institute
of Education, Department of Educational Research.

Rose, D. 2004. Sequencing and pacing of the hidden curriculum: how

indigenous children are left out of the chain. In Muller, J., Davies, B. and
Morais, A. (Eds). Reading Bernstein, researching Bernstein. London:
RoutledgeFalmer.

Taylor, N., Muller, J. and Vinjevold, P. 2003. Getting schools working. Cape
Town: Pearson Education South Africa.

Ursula Hoadley
University of Cape Town and Human Sciences Research Council

[email protected]