Rescuing the Sociology of Educational Knowledge from the Extremes of Voice

Discourse: Towards a New Theoretical Basis for the Sociology of the Curriculum

Michael F. D. Young

British Journal of Sociology of Education, Vol. 21, No. 4. (Dec., 2000), pp. 523-536.

Stable URL:
http://links.jstor.org/sici?sici=0142-5692%28200012%2921%3A4%3C523%3ARTSOEK%3E2.0.CO%3B2-E

British Journal of Sociology of Education is currently published by Taylor & Francis, Ltd..

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/taylorfrancis.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. For
more information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.org
Wed May 16 11:10:37 2007
Britirh Journal of Sociology of Education, Vol. 21, No. 4, 2000

Rescuing the Sociology of Educational Knowledge j-om the

Extremes of Voice Discourse: towards a new theoretical basis
for the sociology of the curn'culurn

MICHAEL F. D. YOUNG, Institute of Education, Unwersip oflondon, U K

ABSTRACT n i s papn is a response to that ofMoore and M u l h '% acourse of Voice and the
F'robh ofKn0wlealp.e and Identi& in the Sociology ofEducation', which appeared in Volume 2 0 ofthis
journal. It starts by summarising and endorsing thdr critictsmf of 'voice discourses' but argues that their
case is weakened by theirfailure to distinguish CAW& between the 'debunking ofknowlealp.e' associated
with the postmodernist theories that undnpin 'voice discourses' and the general propositions of a social
the09 ofkn0wlealp.e. Th idea that knowledge has a social as well as an epistpmological basis is now
wide& a~cq?tedin philosophy as well as sociology. Th paper draws on a paper by Stephen Toulmin
and makes a distinction between anthropological and sociological approaches to the idea of knowledge
having a social basis. It goes on to use some idemfrom the author's recent work on the issue ofknowlealp.e
specialisation to suggest the kind of contribution that a sociological approach to kn0wlealp.e can make to
current cunimlum issues as well as to educational poliy more broad&.

In their recent paper in this journal, Moore & Muller (1999) mounted a powerful attack
on those strands in the sociology of education that have rejected the epistemological
grounding of knowledge in favour of a sociological approach that treats such grounding
as no more than the standpoint or perspective of particular (invariably dominant) social
groups. Their argument is important and much of their critique is justified. However,
whether intentionally or not, what begins as a critique of the extremes of what they refer
to as voice discourse becomes, in effect, a rejection of the claims of any sociological
account of knowledge that sees it as more than a 'field positioning strategy'. The result,
although I am sure not their intention, is to return sociology of education to the problems
and approaches with which it was concerned 30 years ago when it treated education as
little more than an aspect of the distribution of life chances.
In responding to their paper, I seek to accomplish a number of things. I begin by
summarising their argument against voice discourses. I then go on to point to a number
of problems in their argument and suggest that they arise from their over-polemical
stance and are not intrinsic to the case against voice discourses that they want to make.
ISSN 0142-5692 @rint)/ISSN 1465-3346 (online)/00/040523-14 O 2000 Taylor & Francis Ltd

DOI: 10.1080/01425690020017274

 Finally, I build on some points that are hinted at in their paper to suggest the kind of
positive contribution that a sociological approach to knowledge could make to the
sociology of education, and to the study of the curriculum in particular.
In their paper, Moore and Muller trace what they refer to as the 'debunking of
epistemology' in the sociology of education from its early expression in the 1970s, with
its theoretical basis in the phenomenology of Alfred Schutz and the pragmatism of C.
Wright Mills (Young, 1971), to its more recent postmodernist forms that draw on the
writings of Foucault and Lyotard (Usher & Edwards, 1994). They rightly point to the
links between this sociological approach to knowledge and various traditions of radical
politics that, in the 1970s, were largely socialist and populist (Whitty & Young, 1976;
Young & Whitty, 1977). More recently, the sociological critique of epistemology has
taken two paths. One largely disclaims that it has any specific political implications at
all (Usher & Edwards, 1994), and the other has been taken up by feminist and
'post-colonial' writers who use it as a basis for de-legitimising what they see as the
unjustified dominance of Western, white, male knowledge and expertise (Lather, 1997).
'Moore and Muller argue that these approaches, which they refer to as 'voice discourses',
are based on a number of untenable assumptions. The first is the claim that there can
be no epistemology because, fundamentally, it is only experience, not knowledge, science
or expertise, that we can rely on in judging whether something is true. Second, Moore
and Muller argue that 'voice discourse' approaches misuse the writings of historians of
science such as Kuhn (1962) in claiming that they have shown that scientific truths are
no more than what scientists, at a particular time, say is true. Third, Moore and Muller
argue that postmodernist critiques misunderstand science when they equate the assump-
tions made by scientists with the widely discredited philosophy of positivism. As Moore
and Muller point out, such writers appear ignorant of the recent work of philosophers
of science (Papineau, 1996; Toulmin, 1996) who are well aware of the issues raised by
postmodernists but reject their relativist conclusions.
Moore and Muller argue that the result of these errors is that proponents of 'voice
discourses' find themselves in two related kinds of contradiction, one theoretical and one
practical. Both have profoundly negative consequences for research in the sociology of
education and its possible influence on educational policy and practice. They point out
that the theoretical implications of 'voice discourses' is a relativism that leads to the
rejection of all epistemologies; at the same time, voice discourse approaches are far from
relativist about their own position, which denies the possibility of any objective knowl-
edge at all. The practical and political implications of such a rejection of all knowledge
claims is that voice discourses are self-defeating. They deny, to the subordinate groups
with which they claim to identify, the possibility of any knowledge that could be a
resource for overcoming their subordination. There is no knowledge for voice discourse
theorists, only the power of some groups to assert that their experiences should count as
knowledge.
Why, then, do these ideas persist if, as Moore and Muller argue, their flaws are so
obvious and have long been known to be so, and when, in practice, they offer so little
to the groups with which they claim to identifjl? Moore and Muller suggest that the
'debunking of epistemology' does not find support because it has any intellectual or
political merit; for them, it self-evidently does not. The reason it is supported, they claim,
is sociological. Fundamentally flawed ideas persist because they have powerful social
functions in society. Such ideas may, like fascism and racism, have a function for certain
groups in society generally. Equally, in narrower contexts, ideas and methodologies can
persist, as in educational research, because they reflect the professional interests of some
 Educational Knowledge and Voice Discourse 525

members of particular intellectual fields. Debunking dominant knowledge claims is one

way for members of the younger generation in an intellectual field to assert their claims
for recognition and even to displace existing leaders. Moore and Muller point out that
the weak intellectual basis of sociology of education, together with its relative insulation
from intellectually stronger sub-fields within sociology, has made it particularly vulner-
able to this type of positioning strategy.
There is no doubt that, as Moore and Muller argue, a sociological account
of knowledge did play an important 'positioning' role in the 1970s in the sociology of
education, and that it was related to the location of the emerging generation of
sociologists of education in colleges and university departments of education that were
responsible for training teachers (Young, 1998). The sociological approach to knowledge
that became associated with the 'new sociology of education' challenged the 'old'
sociology of education (Floud et al., 1957) and its association with the political arithmetic
tradition of research established by Booth, Rowntree and the Webbs at the turn of the
century. It also challenged the knowledge base of the liberal academic curriculum that
dominated grammar and public schools and the universities (see Young, 197 1, Chapter
1). The shift of explanatory focus implied by the sociological approach to knowledge was
well expressed in its interpretation of the low educational achievement in schools of
pupils from working-class backgrounds. It was not working-class pupils who were failing
to achieve in terms of the academic cuniculum as was maintained by mainstream
researchers. From the perspective of the sociology of knowledge, it was the academic
curriculum, historically constructed to preserve the status quo of a class society, which
systematically ensured that the majority of working-class pupils were failures. I have
commented elsewhere (Young, 1998) on the oversimplification of such conspiracy
theories and the misguided nature of arguments that equate the legitimacy of knowledge
solely with the social position of those who produce it.
Moore and Muller extend this argument by pointing to the episodic character of the
debunking of epistemology by sociologists. They show that sociological critiques of
knowledge have recurred regularly in recent decades, appearing at different times to use
quite different and often opposing theories. They cite the case of postmodemism, which
served a similar intellectual function in the 1980s and 1990s, as was served by Marxism
and phenomenology in the 1970s, while at the same time mounting a critique of them
both. They draw on the work of Usher & Edwards (1994) in adult education as an
example of a sociological approach to knowledge that has quite different intellectual roots
and political concerns from the earlier arguments of the new sociology of education of
the 1970s. Indeed, from Moore and Muller's position, Usher and Edwards, by claiming
that no particular policy implications follow from their position, are more consistent than
the new sociologists of education of the 1970s. The latter, like present-day feminists and
post-colonialists, saw their attacks on epistemology as having, at least in principle, a
politically emancipatory role. As Ward (1996) points out, some postmodernists such as
Baudrillard take the position of Usher and Edwards on knowledge even further and
argue that there is no point in any social science or philosophy; one might as well write
novels or poetry! However, what such positions never seem to account for is the practical
efficacy of some types of knowledge (scientific knowledge and, in a more limited sense,
knowledge in the social sciences), which makes them materially much different from
knowledge based on experience. Furthermore, they also neglect that writing good novels
or poetry is difficult and arguably far harder than writing postmodernist critiques! In
other words, postmodernism provides no escape from the familiar dilemma of any
relativist position; people continue to make judgements about different explanations of
526 M. F. D.Young

the natural and social world whether they derive from experience or from research in a
specialised discipline They also continue to debate the criteria for making such judge-
ments on the assumption that, for certain purposes, it is worth relying on science.
Moore and Muller's critique of voice discourses, and how they tend to reduce
knowledge to knowing and experience, has a compelling power and a basic common
sense appeal. Many examples would serve to illustrate the force of their main argument
that some kinds of knowledge are, without question, more powerful than others, and that
asserting that all knowledge claims are no more than an expression of the claimant's
experience is absurd. Here are two such examples.

Try designing a domestic lighting system without relying at least implicitly on Ohms
Law or designing an aeroplane without knowing the laws of fluid dynamics. All the
experience in the world will not help you unless it is guided by rules that are not tied
to experience.
How do we account for the problem of the two chemists who, some years ago, claimed
that they had discovered how to create energy by the cold fusion of hydrogen in a test
tube? Their results could not be replicated when the experiments they claimed to have
carried out were repeated and they were exposed as either fraudulent or bad scientists.
We have to treat their claims in some sense as false like that of the alchemists before
them, and the existing scientific view that cold fusion of hydrogen, however desirable,
is impossible, as knowledge.

Moore and Muller are correct to point out the flaws in the postmodernist arguments and
how writers such as Usher and Edwards construct a straw-man account of science that
they label positivist in order to show its untenability. The fact that science does not
operate, and indeed could not operate, in the way positivists claim is widely recognised
and does not detract from the explanatory power of scientific concepts. Postmodernist
attacks on positivist concepts of science say little about science. They say more about the
attackers themselves and their objection to those forms of social science that take over
the mathematical form of the natural sciences but lack its conceptual basis.
The other powerful argument that Moore and Muller make is against the way that
'voice discourses' invoke experience as the foundation of all knowledge and therefore the
basis for claiming that all knowledge or truth claims are equivalent, whether they derive
from common sense, folk tradition, laboratory-based scientific research or systematic,
disciplinary knowledge. One does not need to denigrate the knowledge that people gain
from experience, or even to deny that there is an experiential element in all knowledge,
however abstract, to recognise that experience is often an extremely unreliable basis for
deciding whether something is true.
Moore and Muller's argument is that voice discourses, by invoking the experience of
subordinate groups, whether wrapped in the jargon of postmodernism or in the language
of political correctness, have been a cul de sac for the sociology of education. However,
in providing a constructive basis for a way ahead for the sociology of education, their
paper is ultimately disappointing. In the next section of this paper, I want to try and
explain why this is so and then to develop an alternative approach that rescues the
sociology of knowledge both from the relativist excesses of voice discourses and the
limitations of the kind of asocial epistemological realism that Moore and Muller are in
danger of slipping into. I shall develop this alternative by way of commenting on three
issues that arise from Moore and Muller's paper: (i) the tendency for the sociology of
knowledge to be treated as a 'field positioning strategy'; (ii) the insulation of the sociology
 Educational Knowledge and Voice Discourse 527

of education from other 'stronger' intellectual fields; and (iii) the issue of theory and
practice in educational studies.
The first issue relates to the sociological argument they draw from Bernstein (1973,
1999) that seeks to explain the persistence of what they refer to as the sociology of
knowledge approach in the sociology of education in terms of its role as a 'field
positioning strategy' in a weak intellectual field that is highly insulated from other bodies
of related knowledge. Of course, they have a point. First, it is true that sociology of
education, like educational studies generally, is and has long been a weak intellectual field
with what Bernstein (1999) calls a weak 'grammar' or set of rules for describing the
phenomena it seeks to explain. Furthermore, at least in the 1970s when a sociological
critique of epistemology was first applied to the curriculum, the sociology of education
was based largely in University Schools of Education and was relatively insulated from
mainstream sociology and the other social sciences. It is also true, as Bernstein pointed
out many years ago (Bernstein, 1973) that, in the late 1960s and early 1970s, there were
elements of a generation struggle in the debates between the 'new' and 'old' sociologies
of education in the UK. No doubt a similar point can be made about feminist,
post-colonial and social constructivist attempts to set experience against knowledge in
other fields within the social sciences the 1990s. However, this sociological point is not
the end of the matter. Critiques of the possibility of grounding truth claims in
epistemology have gained plausibility and even acceptance in many fields far closer to
what might be defined as the intellectual mainstream than the sociology of education,
and this has been going on since Nietzche's writings became influential from the turn of
the century.
What Moore and Muller refer to as the 'debunking of knowledge' has had a significant
influence in philosophy. Toulmin (1996) refers to the crisis in epistemology, and in
literary and cultural studies, none of which could be described as insulated from other
intellectual fields. It follows that the insulation argument in relation to the sociology of
education fails in as much as it cannot explain the pervasiveness of the same ideas in
other intellectual fields that are positioned very differently. If the insulation argument is
rejected, we have a bigger question about knowledge for both sociology and philosophy
that Moore and Muller do not address, but which continues to haunt philosophers and
sociologists, as is indicated in Ward's (1996) recent study and in a brief but characteris-
tically elegant paper by Toulmin (1996). Ward and Toulmin both state that the
argument for an epistemological grounding of knowledge, at least in its strong form, no
longer has any intellectual credibility in philosophy or the social sciences. In his paper
'Knowledge as Shared Procedures', Toulmin (1996) argues that the 'crisis in epistemol-
ogy' is as old as the twentieth century. Philosophers from Husserl to Rorty, he states:
... have destroyed the programme for finding the unshakable foundations for
human knowledge dreamed up by Descartes and his successors from 1630
onwards.
Toulmin's point is that traditional epistemological arguments were flawed because of
their obsessive concern with how individuals can arrive at any certain knowledge about
the world, as if this is the 'knowledge' problem. The rationality of scientific procedures
is not, he argues, a matter of clarity and distinctness or logical coherence alone and
therefore just a question of epistemology. It depends on the way in which these
procedures develop in the historical evolution of any given discipline; in other words, it
is also a sociological question. Moore and Muller appear to understand Toulmin's point
at a latter stage of their paper when they refer to 'post empirical epistemology'. However,
528 M. F. D.Young

they do not, I think, grasp its full implications and, in particular, the importance of
distinguishing between a number of arguments. These can be summarised as follows.
The 'postmodernist' argument that there are only at best pragmatic grounds for
distinguishing knowledge from experience; hence knowledge, in the sense that the
word is normally used, is impossible.
The 'voice discourse' argument that knowledge claims are always the political claims
of dominant groups. The voice discourse position follows from the postmodernist
argument. It asserts the claims of experience and the equal validity of the perspectives
of all groups, whether expert or not, on the grounds that claims for knowledge, to be
in some objective sense independent of the social position of the knower, are
untenable.
The 'multi-dimensional' argument that the objectivity of truth claims always depends
on their external validity-that they do explain something in a convincing way, on the
support they invoke from a particular community of experts and on the legitimacy of
the particular community involved.
In the case of the example of the cold fusion of hydrogen referred to earlier, the claims
of the two scientists failed on the second criterion. In many cases, the issue will be less
clear-cut. There may be disagreements within particular communities or the legitimacy
of a particular community may be challenged, as in the case of challenges to scientific
experts on increasing numbers of environmental and health issues.
The third position, which is the one adopted in this paper, is not relativist. However,
it allows and, indeed, requires sociological and historical research into the application of
knowledge in different contexts, into changes in knowledge communities, whether of
experts or non-experts, specialists associated with disciplines or those who transcend
disciplines. It is concerned with how communities construct, challenge and modify
knowledge, and how they are challenged from within and from without. In the process,
we can reconstruct the origins and histories of subjects, disciplines and curricula without
denying that subjects and disciplines are not totally constructed by specialist communi-
ties, whether this refers to history, geography, chemistry or even literature. A totalising
social constructionism is flawed as was so vividly apparent in the case of Lysenko's social
environmentalist view of biology (Lecort, 1977). In claiming that nature was endlessly
modifiable, he ended up being able to say nothing useful or new about nature.
Furthermore, Lysenko laid his work open to abuse by Stalin, who was able to use his
views to assert what in fact he already believed, that nothing was beyond the control of
the Party, even nature!
Moore and Muller raise another important issue but do not develop it very far in their
paper; it concerns the relations between theory and practice. At least in the short term,
and certainly from a practical or policy point of view, whether philosophy or literary or
cultural studies adopt a postmodernist perspective on truth and knowledge may not
matter much. However, there could be serious consequences if, during their training,
intending teachers learn not only that all curriculum knowledge is socially constructed,
but that it inevitably reflects the values and interests of the dominant class. For example,
it could appear to follow that the same claims to objectivity could be made for the
everyday mathematics of the street market as are made for the mathematics of the text
book. From a postmodernist perspective, the only difference between the two types of
mathematics would be that the latter reflects the dominant perspective shared by
professional teachers and mathematicians. The limited scope, beyond a very specific set
of contexts, of street or what has become known as ethnomathematics, can be easily
 Educational Knowledge and Voice Dircourse 529

forgotten. This kind of sociological approach to mathematics can undermine the

curriculum rationale for teaching formal mathematics at all. The issue is, of course, not
just concerned with mathematics; it applies to all curriculum subjects. If, in literary
studies, students learn that there are no criteria for claiming the superiority of Jane
Austen over Neighbours or Home and Away, then the only basis for selection of texts for the
English literature cumculum that we are left with is the white middle-class biases and
prejudices of teachers or curriculum policy-makers, on the one hand, or what students
want, a kind of consumer approach, on the other. It is the possibility that a sociological
view of knowledge might actually influence the ways that teachers think about how and
what they teach that makes the knowledge question so much more problematic in the
sociology of education than in intellectual fields such as cultural and literary studies,
which are far more insulated from any practice external to themselves. This possible link
between a sociological view of knowledge and practice is, however, only a problem if it
leads to the relativistic view of knowledge as already outlined. I shall argue later in this
paper (see also Young, 1998, Chapter 2) that this link between the sociology of
knowledge and the classroom practice of teachers can be understood quite differently,
and in a way that enables it to contribute to a more enhanced view of the teacher as a
professional in an increasingly knowledge-based economy.
The problem with Moore and Muller's paper is that, whereas at some points they
appear to challenge the sociological approach itself as reducing knowledge to relation-
ships or processes, at other times they take a subtler and more nuanced position. They
expose the flaws in Usher and Edward's postmodernist arguments in order to argue
against crude versions of sociology that polarise social and epistemological positions on
knowledge. However, the overall thrust of their argument seems to imply that all
sociological accounts tend towards the extremes of voice discourse approaches.
As they put it, voice discourses do not allow science to claim less than absolute truth.
From the perspective of voice discourse, if science and rationality cannot be absolute
then they cannot be objective at all. Obviously, a middle way that combines the sociology
of knowledge and epistemology in making truth claims is a much more defensible
position, although how in particular cases one decides the balance between the two
remains an issue. However, such a 'middle way' would suggest that there is far more
common ground between Moore and Muller and a non-postmodern, non-voice discourse
sociological approach to knowledge than would be apparent from the sharpness of their
attack. My argument, therefore, is that Moore and Muller, in not making clear
distinctions between what we might call 'voice' and 'non-voice' sociologies of knowledge,
are in danger of laying themselves open to the same criticism that they make of those
they attack, such as Usher and Edwards. As they rightly claim, Usher and Edwards fail
to distinguish between science as practiced by scientists and as theorised by philosophers
of science, and the straw-man account of a positivist version of science that they criticise.
What Moore and Muller do is to choose an extreme and sociologically unsophisticated
approach to knowledge (in their case Usher and Edwards, although others might have
been chosen), point out its flaws and link these flaws to a whole theoretical tradition in
the sociology of knowledge as it has been applied in educational research. In a parallel
way, they equate the voice discourse approaches of certain feminist, post-feminist and
post-colonial views of knowledge with all attempts to explore the practical or political
implications of social theories of knowledge and in what ways different concepts of
knowledge may be related to social interests. The outcome is that they leave us with a
distinctly impoverished view of the sociology of education and, in particular, the
sociology of the curriculum.
530 M. F. D. Young

Moore and Muller are right to worry that 'structural level concerns with social
stratification' are being neglected in educational research. However, the neglect of
structural questions by educational researchers is more likely to be the result of the
current obsession by government and by those who fund research, with a peculiarly
narrow view of 'evidence-based' education policy. At least implicitly, Moore and Muller's
programme for the sociology of education seems to hark back to the 1960s, when
sociologists knew that their place in the academic division of labour was to map the
unequal distribution of educational opportunities. Explaining why such inequalities
persisted despite decades of educational reform was, like fundamental questions of
knowledge and purpose of education, left to philosophers.
Let me summarise my argument so far. Moore and Muller are correct to criticise the
excesses of voice discourses and, in particular, the dangers of invoking experience against
the knowledge claims of expertise, science or other bodies of specialist knowledge.
Despite the fact that, in practice, sociological approaches to knowledge have adopted
many of the features of what Moore and Muller describe as 'voice discourses', and that
this has happened in the peculiar episodic way they describe, this is not an inevitable
process. Furthermore, it does not follow that sociological analysis cannot be applied to
questions of knowledge and truth.
The sociology of knowledge reminds us that history and experience inevitably enter
into all forms of knowledge, whether or not it is produced by specialists within particular
disciplinary traditions. The fact that the inescapable role of experience in the production
of new knowledge is often denied is itself a problem for both sociology and epistemology,
and may or may not have practical implications. However, to use an argument about the
role of experience in even the most abstract forms of knowledge as a way of dismissing
the degree and form of objectivity that such knowledge can provide, which is what voice
discourses do, is quite another matter. In relation to education, it follows that we have
to accept the implications of the sociology of knowledge that social class, gender and race
inequalities, as well as a variety of professional interests, are embedded in the organiza-
tion of the school curriculum as well as in the academic disciplines and other structures
within which new knowledge is produced. However, we also have to recognise that how
this embedding takes place in particular cases, and with what consequences, is far from
clear. Furthermore, such general conclusions from the sociology of knowledge give no
necessary grounds for rejecting particular forms of knowledge organisation in the
curriculum. Arguments about the embeddedness of social interests in different forms of
knowledge are at best, hypotheses to be explored; they are never, on their own, grounds
for supporting (or opposing) particular curriculum proposals.
It is philosophy rather than sociology (with the notable exception of Durkheim), and
in particular Wittgenstein's argument that concepts and judgements obtain their meaning
collectively through 'forms of life', which took the sociology of knowledge beyond the
rather narrow debates about ideology initiated by Mannheim. Wittgenstein also put an
end to epistemology in its traditional sense. However, this way of thinking about
knowledge says nothing about the nature of the 'forms of life' that underpin disciplines,
fields, school subjects, or laboratory, workshop and craft practices. Nor, despite having
himself been a school teacher, did Witgenstein discuss the practical implications for
teachers of his ideas. However, the logical implication of his ideas is that curriculum
subjects, like any other source of meaning, are not given but shaped by society and a
product of the practice of subject specialists themselves.
Sociologists, especially those who see their sociology as having a political role in either
preserving or overcoming social inequalities, have been too quick to jump from
 Educational Knowkdge and Voice L?iscourse 53 1

Wittgenstein's points of principle to drawing concrete conclusions about the relationship

between knowledge and particular 'forms of life'. They have been slow to recognise that
Wittgenstein's idea is only a premise for the sociology of the cumculum; it is not itself
a sociological approach to the curriculum. It cannot be assumed, because knowledge
takes its meaning from the 'forms of life' within which it is located, that any particular
social organisation-whether it is a group of specialists, a social class, a gender category
or a group defined by its ethnicity or a particular disability-is a 'form of life' that gives
meaning to knowledge in the sense Wittgenstein meant. The process of embedding or
giving meaning to knowledge is subtle and difficult to describe concretely. Enquiries have
to start from specific cases of actual 'communities7. Sociologists who have argued for an
ethnographic approach to knowledge are therefore partly correct and are supported by
Toulmin (1996) as a philosopher. If we accept that knowledge can be defined as, in some
sense, the shared procedures of particular professional or other communities, it is the
particular processes of knowledge production in such communities that we need to
describe and that will form the basis for any broader or more universalistic statements
about truth and objectivity. Toulmin warns against the tendency in the social sciences to
'move on to theory too soon when we would be better focusing on the humbler task of
giving accurate descriptions of human activities ... (which might help us to judge when)
formal theories can do us any good' (Toulmin, 1996).
The recognition that knowledge and truth claims are inescapably rooted in social
activities, forms of life or practices has a number of implications for research in the
sociology of education. I shall discuss each briefly. First, it suggests that, contrary to
the arguments of Bernstein (1999) and Karabel & Halsey (1977) that the location of the
sociology of education in institutions dominated by the practical activity of training
teachers is a weakness, this is or could be a strength. Following Toulmin's argument
discussed earlier, it is a mistake to imagine that there is something called 'theory' or
research in the social sciences that is best developed independently of the exigencies
of policy and practice. If sensitive and accurate descriptions of practice are the bedrock
of our knowledge of any human activities, such as learning and teaching, then re-
searchers such as sociologists of education, have, as Toulmin puts it, to 'Lgettheir hands
dirty and work with and alongside the (educational) professionals whose enterprises they
study'. Second, the social theory of knowledge that can be derived from Wittgenstein's
propositions involves two interdependent stages (although not in a literal sequence).
Toulmin puts the point clearly when he states that:
... meanings are created in the public domain in the context of collective
situations and activities.
Furthermore, Toulmin argues that these meanings are internalised by individuals:
... in the learning processes by which we master everyday language and
subsequently the conceptual content of more sophisticated enterprises (such as
science, poetry or management).
The key issue for the sociology of education is the relationship between these two types
or stages of learning. Essentially, they parallel the distinction between the tacit knowledge
that is embedded in every practice (and fundamentally in the language of the community
into which one is born) and the codified knowledge that is abstracted from practice for
different purposes, such as the subject knowledge of the school curriculum and the
academic disciplines within which most research is undertaken. It follows not only that
the sociology of knowledge is inseparable from the sociology of learning, but that the
study of the cunculum is inseparable from the study of learning and pedagogy.
532 M. F. D.Young

In raising a third issue, I part company from Toulmin and agree with Moore and
Muller's argument, although not with the precise implications they draw. Moore and
Muller argue that the misplaced invoking of experience by 'voice discourse' theorists has
led them to reduce sociological research to ethnography, thus avoiding the major
structural issues of inequality that dominate mass education systems as much today as
when they were first established in the nineteenth century. From a more sophisticated
theoretical position, Toulrnin argues that understanding-human activity mist eschew
generalisations and 'theory' and concentrate on the careful description of practice. The
outcome of following Toulmin's recommendations, however, is essentially the same as
that from the 'voice discourse' theorists. As Moore and Muller point out, the latter
endorse ethnography as the preferred method for revealing the experience that, for them,
is hidden by 'scientific methods.' However, ethnographic methods, although giving an
authentic sense of lived experience, neglect the contexts that shape different kinds of
experience and leave us with descriptions of practice as if they had no context. Toulmin
is right, in my view, to warn of the dangers of premature moves to theory and the
absurdity of trying to achieve the kind of abstract theoretical structures in the social
sciences that characterised Newtonian physics. Despite their postmodernist rejection of
science, this ironically is what the 'voice discourse' theorists do in applying their
dominant/subordinate group distinction in a universal way; they use it as an 'abstract
theoretical' structure to undermine the legitimacy of science and other forms of expertise.
Where Toulmin's argument can be taken too far, at least from a sociological perspective,
is if it leads to the rejection of attempts to develop concepts that go beyond specific
examples of practice. We must heed Toulmin's warnings and be acutely aware of giving
such concepts an objectivity and generalisability that they do not warrant. However, it
is equally mistaken to deny that concepts such as social class, alienation, ideology,
bureaucracy and the division of labour do not have a conceptual power that enables us
to use them to go beyond our own experience.
In making clear the difference between Toulmin's approach and the one I am arguing
for, it is useful to distinguish between two separate but related ways in which knowledge
can be understood as being socially constructed. I shall call the two ways anthropological
(in the broader rather than the disciplinary sense of the term) and sociological. The
anthropological version of the social basis of knowledge refers to the view adopted by
many modem philosophers that now complements or, some would say, has replaced
epistemology. It is the position proposed by Wittgenstein and developed by Toulmin but
has much in common with Durkheim's early writings in which he argued that even the
most basic categories of thought, such as cause and time, were social in origin. Such a
view of knowledge is a conceptual statement about its social nature; it is not a statement
about how different social realities shape the nature and content of knowledge. In
referring to it as anthropological, I mean that it applies universally to all people and all
cultures; it might also be described as a social epistemology. It asserts that any knowledge
is premised on some idea of society. The second or sociological sense of a social theory
of knowledge is premised on such a social epistemology. However, because it has
different purposes, it is both more ambitious and more modest. It is more ambitious in
that it goes beyond the general principle that knowledge has social origins and argues
that it makes sense to identify the particular ways in which social interests shape the form
and content of knowledge. The continuing task of sociologists is to explore how, in
particular cases such as those involving environmental and health risks or the content of
the curriculum, social relations of power and interest are expressed in the organisation
and content of knowledge. The sociology of knowledge is (or should be) modest in
 Educational Knowledge and Voice Discourse 533

recognising that while it is a useful hypothesis to consider how knowledge claims reflect
attempts to preserve particular social interests, the counter hypothesis that they fre-
quently go beyond such interests is also important; even ruling groups need reality tests
if they are to remain in power. Sociology also has to be modest in recognising that the
ways that interests shape knowledge has to be explored empirically in specific cases.
There is no general law of the sociology of knowledge beyond what I have referred to
as its anthropological basis.
In my own work on the sociology of the cumculum (Young, 1998),I have focused on
the changing form of specialisation,both in relation to the production of knowledge and
the division of labour generally. I suggest that changing forms of specialisation are among
the ways that the wider context shapes the production of knowledge (Gibbons et al., 1994)
and the organisation of the curriculum. I argue that the distinction between insular and
connective specialisation captures some of the underlying tensions and contradictions in
contemporary curriculum debates, and can provide a way of linking these debates with
wider social changes. However, this is not to propose either that connective specialisation
is some kind of general theory that can be imposed mechanically on current changes in
the cumculum or that it is a cumculum principle that teachers can apply. It has a
number of more modest purposes. First, the distinction between two forms of specialis-
ation offer a way of making some sense of current cumculum debates, especially but not
only those in the post-compulsory phase of schooling. Second, in the specific case of the
English post-1 6 cumculum, it leads to a critical analysis of the conflict over A-levels and
the reformulation of future possibilities for the advanced-level curriculum. Whereas
insular specialisation is an accurate description of the traditional A-level curriculum and
its basis in groups of separate subjects, connective specialisation does not propose the
abolition of subjects, but a re-definition of their role in relation to the purposes of the
post-compulsory curriculum as a whole.
Elsewhere (Young, 1998, 1999), I have suggested that the distinction between insular
and connective can be applied to a number of aspects of the curriculum. These are the
relationships between:
the everyday or common-sense knowledge that learners bring to the curriculum and
the organisation of knowledge in the curriculum itself;
the tacit knowledge that learners acquire in any context (including schools) and the
conceptual or codified knowledge embodied in cumculum subjects;
the component elements of the curriculum (which may be subjects or vocational fields
of study) and the curriculum as a whole; and
school knowledge (both tacit and codified) and the workplace knowledge
- that students
may acquire through work experience, part time jobs or placements (Griffiths & Guile,
1999).
In the case of each relationship, it is possible to trace a tension in the history of mass
education between the trend to greater (and more insular) specialisation in the cumcu-
lum, as well as in the organisation of work and knowledge generally, and recent attempts
to connect different types of knowledge and learning.

S u m m a r y and Conclusions
In responding to Moore and Muller's paper, I have, first, argued that in the sociology
of education we need to take seriously their criticism of certain trends that have been
devoted to debunking specialist or disciplinary knowledge, whether in the curriculum or
534 M. F. D. Young

in research. Second, I have suggested that, while their criticisms are well merited, they
undermine their own case by broadening their attack beyond where it can claim validity.
Third, I have attempted to rescue a constructive social theory of knowledge - from the
extremes to which it was taken by the relativism of voice discourse approaches and the
postmodernist theories on which they are based. Fourth, I have drawn on Wittgenstein's
ideas, as developed by Toulmin, in favour of a social theory of knowledge and suggested
that it is useful to make a distinction between anthropological and sociological ap-
proaches to the idea that knowledge has a social basis; I present the former as the basis
for the latter, not an alternative or a critique of it. Finally, I have indicated one strand
of the kind of research programme such a sociological approach points to by focusing on
the issue of knowledge specialisation and its relationship with the distribution of power
and the role of institutions.
Finally, it is legitimate to ask whether all this esoteric discussion about knowledge
matters when inner-city schools are failing and good head teachers are resigning in
exhaustion? Are the real issues not the perennial ones of governments being reluctant to
spend the necessary money to improve the curriculum? Who bothers any more about
what counts as knowledge when everyone is chasing after grants to do more research on
the literacy hour and other examples of what claim to be evidence-based practice? I have
written this paper because I think the issues raised by Moore and Muller do matter even
when I disagree with them, not only for sociologists but for anyone interested in
education. This is for a number of reasons. First, the intellectual dishonesty and potential
harm that can be caused by voice discourse theorists needs to be pointed out loudly and
clearly, and Moore and Muller do that. 'Voice discourse' theorists are often clever and,
in appearing to be democratic both in their deference to the experience of so-called 'real
people' and in their critique of expertise, they are also seductive. In rejecting the claims
for any kind of objective knowledge, the logic of their position is nihilist and leads to the
cynicism of social scientists who reject the grounds of their own practice or, as in the case
of Baudrillard, give up social science altogether. If taken seriously, as some in Germany
and elsewhere took Nietzche, the rejection of any grounds for knowledge can lead to the
view that the only question is who has the power to impose their view of the world. It
is deeply depressing that such arguments can be found among those identifying with the
real subordination of minorities or excluded groups. Second, Moore and Muller's
critique of social theories of knowledge itself needs to be challenged. It denies sociologists
some of the tools they need, both in exposing the hypocrisies and contradictions in public
policies and in suggesting realistic criteria for alternatives. For example, a social theory
of knowledge is important for exposing the current rhetoric of standards, targets,
partnerships and improvement (all good ideas in principle of course, but easily misused)
and why they may be resisted in the reality of the schools that they are supposed to
influence. The sociology of knowledge takes the debate about such ideas back to the
educational purposes that might underpin alternative approaches to policy.
This is not to argue that sociology should just be a critique of whatever policies are
proposed by government and therefore always in opposition; far from it. It is rather to
argue that, if sociologists and other educational researchers are to contribute to
educational policy, they need to start from a position that many, if not all, of the
assumptions that are unquestioned in policy are anything but given. The lessons from the
past 30 years of the sociology of education are not that the sociology of knowledge should
not have been applied to the curriculum or other aspects of educational policy; they are
that its role needs to be more embedded in the realities and constraints of policy and
practice. The sociology of knowledge has to be radical in the theoretical challenges it
 Educational Knowledge and Voice Discourse 535

poses, but more aware of the constraints within which those who make and implement
educational policy find themselves. Educational policies and practices are like science,
human constructs developed consciously and sometimes unconsciously for certain pur-
poses. The sociology of knowledge provides an important set of conceptual tools for not
letting us become immersed in the minutiae of specific policies and for reminding us that
the fundamental questions about purpose and inequality in education have to continue
to be asked. As Moore and Muller show clearly in their paper, sociological approaches
to knowledge can easily be used irresponsibly. My argument is that this irresponsibility
is not intrinsic to the theories, but is a reflection of how theorists sometimes see their role.
An over-politicised view of theory and its radical potential, as in the case of 'voice
discourse' approaches and the early versions of the new sociology of education, is one
response to the possibilities of the sociology of knowledge; a denial of responsibility for
the consequences of one's theory is another. Both responses can be found in the writings
of educational theorists in recent decades. Neither possibility is a reason for rejecting the
necessary role of the sociology of knowledge in any serious programme of educational
research, however uncomfortable this may be for government and whatever intellectual
and moral difficulties are involved for researchers.

Acknowledgements
The author would like to thank Michael Barnett, Toni Griffiths, David Guile and Joe
Muller for their helpful comments on earlier drafts of this paper.

Correspondence: Michael Young, Life Long Learning Group, Institute of Education,

University of London, 55-59 Gordon Square, London WClH ONT, UK. E-mail:
m-young@ioe.ac.uk

REFERENCES
BERNSTEIN, B. (1973) C h s , Codw and Conh.01, Volume 3 (London, Routledge).
BERNSTEIN, B. (1999) Vertical and horizontal discourse, BnbhJoumal of Sociobgv ofEducation, 20, pp. 157-173.
ROUD, J., HALSEY,A.H. & MARTIN,F.M. (1957) Social C h s and Educational Opportunib (London, Heineman).
GIBBONS,M., LIMOGES, C., NowoTNY, H., SCHWARLZMAN, S., Scorn, P. & TROW, M. (1994) 7hc New Mo& of
Knowhdge Production (London, Sage).
G R I ~ H ST., & GUILE,D. (1999) Learning and pedagogy in work based contexts, in: P. MORTIMORE (Ed.)
Understunding Pedagogv and its Impact on Lenmitzg (London, Sage).
IC\RABE~ J. & HALSEY,A.H. (Eds) (1977) Power, Ideologv and Education (Oxford Oxford University Press).
KUHN,T.S. (1962) 7 7 S~tnuhrrc of Scimtjfc Revolutions (Chicago, IL, Chicago University Press)
LATHER,P. (1997) Getting Smart (London, Routledge).
LECOURT, D. (1977) ProletRion Science? 7hc cacc oflysmko (London, New Left Books).
MOORE, R. & MULLER, J.
(1999) The discourse of 'voice' and the problem of knowledge and identity in the
sociology of education, Bribh J'ournal of Socwbgv ofEducation, 20(2), pp. 189-206.

P A P ~ E A UD., (Ed.) (1996) irhe Phifosophy of Science (Oxford, Oxford University Press).
TOULMIN, S. (1996) Knowledge as shared procedures, in Y. ENGESTROM, R. MIET~ISEN & R.L. PUNAMAKI
(Eds) Persspcrives on Actiuib irheoy (Cambridge, Cambridge University Press).
USHER,R. & EDWARDS, R. (1994) Post M o d e m k and Education (London, Routledge).
WARD,S. (1996) Reco&ring Tmth:poshodemirm, science studies and the searchfor a new model ofhowleke (Boston,
MA, Rowman & Littlefield)
WH*, G. & YOUNG,M. (Eds) (1976) Explorations in the Potirits ofSchool Knowledge (Driffield, Nafferton Books).
536 M. F. D.Young

YOUNG, M.F.D. (Ed.) (1971) Knowhdgc and Conhob new directions for the sociology of education (London, Collier
Macmillan).
YOUNG, M.F.D. (1998) irhc Cuniculwn of thc Fufure (London, Falmer Press).
Yo%, M.F.D. (1999) Knowledge, learning and the cuniculum of the future, British Educational RrrearchJoumal,
25,(4) pp. 463-477.
YOUNG, M. & W H ~G., (Eds) (1977) So&&, Sh& and Schooling (Brighton, Falmer Press).
http://www.jstor.org

LINKED CITATIONS
- Page 1 of 1 -

You have printed the following article:

Rescuing the Sociology of Educational Knowledge from the Extremes of Voice Discourse:
Towards a New Theoretical Basis for the Sociology of the Curriculum
Michael F. D. Young
British Journal of Sociology of Education, Vol. 21, No. 4. (Dec., 2000), pp. 523-536.
Stable URL:
http://links.jstor.org/sici?sici=0142-5692%28200012%2921%3A4%3C523%3ARTSOEK%3E2.0.CO%3B2-E

This article references the following linked citations. If you are trying to access articles from an
off-campus location, you may be required to first logon via your library web site to access JSTOR. Please
visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

References

Vertical and Horizontal Discourse: An Essay

Basil Bernstein
British Journal of Sociology of Education, Vol. 20, No. 2. (Jun., 1999), pp. 157-173.
Stable URL:
http://links.jstor.org/sici?sici=0142-5692%28199906%2920%3A2%3C157%3AVAHDAE%3E2.0.CO%3B2-H

The Discourse of 'Voice' and the Problem of Knowledge and Identity in the Sociology of
Education
Rob Moore; Johan Muller!y
British Journal of Sociology of Education, Vol. 20, No. 2. (Jun., 1999), pp. 189-206.
Stable URL:
http://links.jstor.org/sici?sici=0142-5692%28199906%2920%3A2%3C189%3ATDO%27AT%3E2.0.CO%3B2-Q

Knowledge, Learning and the Curriculum of the Future

Michael Young
British Educational Research Journal, Vol. 25, No. 4, Post-Compulsory Education. (Sep., 1999), pp.
463-477.
Stable URL:
http://links.jstor.org/sici?sici=0141-1926%28199909%2925%3A4%3C463%3AKLATCO%3E2.0.CO%3B2-7