4

Public Policy and Discourse Analysis

Discourse is not easy to define. Many authors in the primary anti-positivist traditions
of discourse analysis—hermeneutics, post-structuralism, and post-Marxist—use the
term differently, making it difficult to offer a fixed or commonly accepted definition.
All rights reserved. May not be reproduced in any form without permission from the publisher, except fair uses permitted under U.S. or applicable copyright law.

Various characteristics of discourse, however, are relatively clear. Discourse, as

Howarth (2002: 9) writes, refers 'to historically specific systems of meaning which
form the identities of subjects and objects'. Discourse theory, as such, starts from the
assumption that all actions, objects, and practices are socially meaningful and that
these meanings are shaped by the social and political struggles in specific historical
periods. Through a range of linguistic and non-linguistic materials—verbal statements,
historical events, interviews, ideas, politics, among others—the goal of discourse
analysis is thus to show how these actions and objects come to be socially constructed
and what they mean for social organization and interaction. Included among the
methods for doing this are the tools of rhetorical analysis, hermeneutics, deconstruc-
tion, and genealogical approaches to discourse analysis. The discussion that follows
draws on these various approaches without adhering to a particular school of analysis.
Clearly, then, discourse is more than just synonymous with discussion or talking.
The meanings of the words used and the statements employed in a discourse depend
on the social context in which they are uttered, including the positions or arguments
against which they are advanced. At the level of everyday interaction, discourses rep-
resent specific systems of power and the social practices that produce and reproduce
them. As MacDonell (1986: 2) illustrates the point by showing how discourses func-
tion in different social settings: the kinds of speech appropriate on the factory floor
differ sharply from those in the corporate boardroom, just as doctors in a hospital
speaks differently from their patients. Different social groups, especially groups with
differing degrees of power and authority, may use the same words differently in their
interpretations of social and political situations.
For Hajer (1995a: 44) discourse is 'a specific ensemble of ideas, concepts, and cat-
egorizations that are produced, reproduced, and transformed to give meaning to phys-
ical and social relations'. As such, a discourse and discursive practices circumscribe
the range of subjects and objects through which people experience the world, specify
the views that can be legitimately accepted as knowledge, and constitute the actors
taken to be the agents of knowledge. Through such discursive delimitations, as
Copyright 2003. OUP Oxford.

Shapiro (1981: 130) puts it, a discourse establishes 'norms for developing conceptu-
alizations that are used to understand the phenomenon'. Discursive practices,
which Foucault took as his basic unit of analysis, are the widely held and oft-repeated
interpretations of social conduct that produce and affirm behaviours. Over time these
EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL
AN: 176953 ; Fischer, Frank.; Reframing Public Policy : Discursive Politics and Deliberative Practices
Account: s6670599.main.eds
 74 Public Policy and Discourse Analysis

interpretations become unreflectively taken for granted; they are scarcely noted by
the actors who employ them. As generally accepted presuppositions, they become
embedded in the institutional deliberations and practices that produce and govern
basic societal relations.
Focused on virtually all research concerned with language in its social and cog-
nitive context, including scientific disciplinary discourses, legal discourse, everyday
narrative discourse, and so on, discourse analysis has given rise to numerous
approaches. Basic to these orientations, though, is a concern with defining the units
of analysis in various modes of deliberation and inquiry (Meinhof 1993). A dis-
course, in this respect, is not just any collection of words or sentences. Rather it is
an integration of sentences—spoken or written—that produces a meaning larger than
that contained in the sentences examined independently. Each kind of discourse links
the sentences that compose it according to distinct patterns of reasoning. Physics, for
example, is one kind of discourse; social science another; and law still another. And
there are, of course, many others, including everyday discourses.
Fairclough (1992) offers a method that combines elements of hermeneutic, post-
structural, post-Marxist and rhetorical approaches to discourse analysis. Following
his approach, a 'discursive event' can be at once be seen as text, discursive practice,
or social practice (Fairclough 1992: 4). 'The "text" dimension attends to language
analysis of texts.' Focus on this aspect of discourse is the primary activity of lin-
guistics (Johnstone 2001). Here a basic task is to understand how a discourse links
together its utterances according to distinctive discursive forms—for example, ref-
erential, narrational, persuasive, expressive, or poetical (Kinneavy 1971). A discurs-
ive practice, of particular interest to literary scholars, focuses the processes of text
production and interpretation. The analysis of discursive practices involves, among
other things, an examination of which types of discourses are employed in particu-
lar contexts, especially which discourses are privileged in particular areas of policy-
making and which are excluded. And as social practice, discourse refers to issues
such as the institutional and organizational circumstance of the discursive event and
how that shapes the nature of the discursive practice (Fairclough 1992: 4). Here one
would focus on the kinds of assumptions that underlie practices of particular policy
institutions—for example, the way an office of education might be governed by a
particular educational philosophy, or the way in which a particular methodological
orientation might be embedded in the rules and procedures of a political science
association and its official journal. This concern with social practices is a primary
interest of the social scientist.
Whether directly expressed in discussion and debate among speakers, or embedded
in the practices of societal institutions, discourse communicates on two basic levels:
the broad cultural level and the everyday level of communicative interaction.
Although perhaps somewhat simplistically, these two levels can be usefully thought
of in terms of micro everyday (first-order) and macro socio-cultural (second-order)
discourse, which is the primary concern of this chapter. At this socio-cultural level,
macro discourse transmits basic values and gives cohesion to shared beliefs, similar
to what Schmidt (2002: 210) describes as 'communicative' discourse in Chapter 2.

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 Public Policy and Discourse Analysis 75

Among other things, it supplies society with basic stories that serve as models of
behaviour, both positive and negative.
Examples of discourse at the macro level are not difficult to find. Perhaps the best
example in Western countries is Christianity. In Christian cultures, the most profound
illustration is the story of Jesus of Nazareth, which provides insights for all members
of society, even including people who are not especially religious. Such discourses
convey to a society its basic socio-cultural identity, that is, where it comes from, how
it got there, and what its goals and values are. Or take the case of modern-day Israel.
It exists in significant part because of the Holocaust and the story is told and retold to
remind Jews of who they are and why they must defend themselves. In Germany, of
course, the same story conveys a very different message of shame and guilt.
At this cultural level, discourses function epistemically to regularize the thinking of
a particular period, including the basic organizing principles of social action (such as
the rules of feudalism or capitalism). Functioning as deep socio-linguistic structures,
discourses organize the actors' understandings of reality without them necessarily
being aware of it. In the Foucaultian sense of the terminology, such epistemic
discourses have formative or constitutive power that structures basic social definitions,
meanings, and interactions in a socio-cultural system. As large encompassing systems
of meaning embedded in and transmitted by culture, macro discourses constitute the
'residua' of a society's or group's collective memory. They do so primarily in the form
of stories that can be taken as the engrams basic to our modes of thinking and action.
Consider a couple of examples. Basic to French history and culture is idea of the
French nation as carrier of civilization to other parts of the world, an idea grounded
in the French Revolution and extended by the far-flung missions of Napoleon. This
contrasts with the English tradition that emphasizes the establishment of a parlia-
ment and the rights of Englishmen. Whereas French political discourse stresses the
universal rights of man, English political discourse is grounded in the particular
rights of Englishmen and a concern about parliamentary encroachments. Such larger
discursive themes serve as points of orientation and departure in more concrete polit-
ical and everyday discourses in these different political cultures, especially when it
comes to dealing with each other. Under these broadly structuring discourses, polit-
ical stories and narratives are told which not only reflect these general systems of
meaning but also work them out in the concrete practices of the everyday world of
social action. In the politics of the European Union, for example, these differences
offer underlying reasons why the British are reluctant to turn over their sovereign
powers to a continental governing body (Schmidt 2001). For the French, they con-
stitute grounds for advancing the European Union, so long as the French government
is one of its leading directors.
An even more deep-seated example of such cultural residua is found in the way
common everyday phases of a language can carry and sustain more fundamental
norms which speakers no longer recognize as such. For example, consider the
expression 'alles in Ordnung' in the German language, which is regularly repeated
in the course of the most mundane events. No one thinks about it, but the phrase no
doubt has roots in the emphasis on discipline and authority that runs through

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 76 Public Policy and Discourse Analysis

Prussian history. (A famous story is told of the Prussian king who fired his generals
because they failed to follow his orders to the letter, although their actions enabled
them to win the Franco-Prussian war.) Contemporary Germans today do not attach
such deep meaning to the regular repetition of 'alles in Ordnung', which can be
likened to Americans casually asking someone 'how it's going?' A discourse analysis
would nonetheless take note that the phrase is still there and resonates, even unin-
tentionally or unconsciously, with part of German history. In this way, it carries for-
ward important aspects of a culture. One would not want to draw immediate
consequences from such an utterance, but would nonetheless want to ask about the
way such an expression still tends to influence contemporary cultural orientations,
including important social constructs such as efficiency or cleanliness.1 Albeit
always in new ways, a culture constantly reproduces itself. The process of reproduc-
tion is one of the ways a people show or reveal themselves.

DISCOURSE ANALYSIS
Discourse analysis in politics begins with the recognition that discourses are dis-
tributed across institutions. In addition to the dominant discourses, competing dis-
courses struggle to gain recognition and power. A key task for the analyst is to
account for the viewpoints and positions from which actors speak and the institu-
tions and processes that distribute and preserve what they say. As MacDonell (1986:
2-3) writes, a discourse 'may be identified by the institutions to which it relates and
by the position from which it comes and which it marks out for the speaker'.
The speaker's position, however, does not stand alone. It can only be understood 'as
a standpoint taken up by the discourse through its relation to another discourse', as
when the owners speak to the workers.
The discursive constitution of society, as Fairclough (1992) makes clear in his very
useful explication of the theory and methods of discourse analysis, does not stem
from a free play of ideas in people's heads. Rather, it emerges from practices that are
rooted in and oriented to basic social structures and ideological practices, particularly
the tensions and conflicts to which they give rise. This connects discourse to practices
that constitute—and naturalize, sustain, and change—the basic social and political sig-
nifications of the world. Politically, such practices not only construct but also
support and alter specific power relations between collective entities (classes, groups,
communities, and so on) through which they are expressed. The ideological
significations of these entities, as Fairclough makes clear, are generated within power
relations as part of the exercise of and struggle over power. Discourse in politics is
thus not only an activity in a power struggle, but also a stake in it as well.
The question of the relations among discourse, power, and ideology is the subject
of complicated debates. Indeed, the term 'discourse' was in part introduced to deal

1
No contemporary German would deny that an emphasis on order and efficiency is embedded in
Germany's institutional culture, but many complain that it is not always that easy to find in practice.

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 Public Policy and Discourse Analysis 77

with problematic issues related to particular theories of ideology.2 Here, though,

there is no widely accepted statement in the literature. Although discourses are not
ideologies, they do intersect with ideologies, which supply the words of a discourse
with different meanings. Built into the various meanings of discursive practices,
ideologies—as interpretive constructions of reality—contribute to the production
or transformation of power relations and political domination (Fairclough 1992:
86-91). Pecheux (1982) argues that discourses, which have their basis in language,
nonetheless are always uttered in the social sphere, which is constructed ideolog-
ically. They always find their social meanings by reference to an ideological position
(MacDonell 1986: 43-55). Even though they are not ideologies per se, discourses
will generally have an intentional or unintentional relationship or position to one.
Drawing on Althusser(1971) and Pecheux (1982), as MacDonell (1986: 102) writes,
a discourse 'is pinned down where it serves as a weapon in an ideological struggle'.
It is in these struggles that a discourse finds its social meanings.
A first-rate example of the relationship of discourse to ideology is found in the
academic discourse of economics. Western industrial societies are organized as cap-
italist systems; thus it is difficult to render any statement completely outside the
realm of capitalist ideology. Nonetheless academic economics presents itself as
a value-free scientific discourse independent of capitalist or socialist ideologies.
It purports to identify and describe casual economic relations, often portrayed as
'natural', and offers an apolitical way to discourse about them. Despite the fact that
modern-day 'neo-liberaP capitalist discourse can arguably be defined as the hege-
monic discourse of our time, many contemporary introductory economics texts
never—or almost never—mention the world 'capitalism', thus extracting the dis-
course from its historical and ideological context. A socialist, however, is correct to
identify the capitalist understandings embedded in these discourses. The key words
of such the discourse—'management', 'workers', 'profit', and 'corporations', and so
on—are interpreted to have very different meanings by socialist ideology. Where a
capitalist reads the economic text and unreflectively takes the word 'profit' for
granted, the socialist reads the same word against a concern for exploitation.
Because ideologies reflect basic material and social relations in a society, they also
supply people with different social identities. The various groups in society thus
orient themselves ideologically in different ways to particular discourses. To stay
with economics, the British Labour Party, or at least the left wing of the party, keeps
its distance from the free-market discourses of contemporary economics. In this
respect, as Pecheux (1982) has argued, people construct their consciousness in terms
of three basic orientations to the dominant discourses and their ideological orienta-
tions. Some freely consent to and identify with the social image held out to them by
2
Initially, the concept of political discourse was introduced to understand the historical evolution of
social systems by analysing the nature, origins, and evolution of its discursive formations. In particular, it
has been developed to escape the problematic aspects of the concept of ideology, especially its relation-
ship to the Marxist concept of 'false consciousness' and the limits of more materialist approaches that
assigned cultural and other ideational dimensions to a secondary status. In the process, though, this
relationship of discourse to ideology has become a complicated theoretical question.

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 78 Public Policy and Discourse Analysis

the dominant ideology, for example, working class or upper class (they are the 'good
subjects'). Others refuse the social categories that are offered, adopting a 'counter-
identification'. These 'trouble-makers' include criminals, but also often liberal
reformers. The third group, Pecheux argues, takes the orientation of 'disidentifica-
tion'. Whereas the counter-identification position defines itself in oppositional terms
('what you call the political crisis'), this third group sets out a new discourse.
Although the identifications established by the dominant ideology can never be
entirely evaded, this orientation seeks to transform and displace them. A traditional
example is the Marxist theory of capitalism; a contemporary illustration is radical
ecology. It here that one can understand the revolutionary Sartre's argument that the
trouble with being radical reformer is that the opposition always sets your agenda. The
alternative, he argued, is to be a revolutionary and seek to set the agenda yourself.
Gramsci's concept of hegemony provides a fruitful way of conceptualizing and
exploring these interrelated ideological and political aspects of discourse, in particu-
lar the ways they both contribute to and are shaped by wider processes of social
and political change (Laclau und Mouffe 1985; Fairclough 1992). Hegemony, in
Gramsci's (1971) work, emphasizes domination across the economic, ideological,
cultural, and political domains of society. It refers to the power over society by one
of the economically defined classes in political alliance with other societal forces.
For Gramsci, though, this does not mean a fixed or firm social equation. Hegemony,
as he defines it, is an unstable equilibrium that always remains partial and temporary.
For this reason, the focus is on the strategies of political leadership required to sus-
tain the hegemonic balance of forces. Constructing and maintaining the alliances
necessary to sustain the equilibrium, as discursive power, involves integrating rather
than simply dominating subordinate classes to win their consent. Through a com-
bination of ideological means and material concessions, hegemonic politics emphas-
izes the constant struggle around the points of greatest instability between classes or
groups, in an effort to build, sustain, or block alliances. Hegemonic struggle, as
Fairclough (1992: 92) explains it, 'takes place on a broad front, which includes the
institutions of civil society (education, trade unions, family), with possible uneven-
ness between different levels and domains'.
Gramsci (1971: 324) offers a conception of political groups as structured by the
diverse ideologies implicit in their discursive practices. Given the sometimes conflict-
ing or contradictory components that make up such understandings of groups, Gramsci
speaks of their 'strangely composite' character. Basic to these composites is the way in
which common-sense knowledge serves as both 'a repository of diverse effects of past
ideological struggles' and a regular target for restructuring in social and political
struggles, hi common-sense knowledge, 'ideologies become naturalized, or automat-
ized' (Fairclough 1992: 92). On this view, ideologies are understood to be fields of 'con-
flicting, overlapping, or intersecting currents or formations', or what Gramsci referred
to as an 'ideological complex'. This necessitates an examination of the processes that
articulate and restructure ideological complexes. In more recent years, a newer group of
'neo-Gramscian' scholars have sought to revitalize Gramsci's insights in the area
of international relations and foreign policy (Gill 1993; Cox 1994; Sinclair 2000).

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 Public Policy and Discourse Analysis 79

Such an approach to political struggle offers a dialectical perspective on the

relationship between discursive structures and events. Viewing discourses as the
relatively unstable and often contradictory configurations that hold together the equi-
libria of political hegemony, 'the articulation and rearticulating of orders of dis-
course is correspondingly one stake in hegemonic struggle' (Fairclough 1992: 93).
Furthermore, 'discursive practice, the production, distribution, and consumption
(including interpretation) of texts, is a facet of hegemonic struggle which contributes
in varying degrees to the reproduction or transformation not only of the existing order
of discourse, but also to the reproduction of existing social and power relations.'
To illustrate the relationship between discourse and social change Fairclough offers
a number of examples. His concern is with broad tendencies in discursive change
affecting the societal order of discourse, which are related to more general directions
of social and cultural change. One of his examples to explain the ways that language
and discourse are used to facilitate social change is that of 'commodification'. This
involves analysing a process whereby social domains and institutions, whose concern
is not producing economic commodities in the narrower economic sense of goods for
sale, come nevertheless to be organized and conceptualized in terms of commodity
production, distribution, and consumption. 'In terms of orders of discourse', he writes
(1992: 207), 'we can conceive of commodification as the colonization of institutional
orders of discourse, and more broadly of the societal order of discourse, by discourse
types associated with commodity production.' Specifically, the process of commod-
ification is used to show the way in which a language text can be employed to
mediate a social transformation to a more entrepreneurial market-oriented culture, a
phenomenon of no minor importance. Since the Thatcher years in Britain and the
Reagan period in the USA, such a process of commodification has been increasingly
extended to one institutional domain after another.
Education in both countries is an important example. Towards this end, Fairclough
concretely illustrates the role of the commodification discourse in mediating the
process of social change through the example of the transformation of the traditional
liberal educational curriculum to a more practical training orientation for the labour
market. In the process, he shows how education is increasingly reconceptualized as
an 'industry' or 'enterprise' concerned with producing and selling cultural or educa-
tional commodities to their student 'clients' or 'consumers'. A widespread feature of
contemporary educational discourse is the wording of courses or programmes of
study as 'commodities' to be 'packaged' by curriculum designers and 'marketed' in
a competitive bid for 'customers'. (See Appendix.)
In order for a specific discourse to become hegemonic, according to Laclau
and Mouffe (1985), a particular set of 'nodal points' must be established in the dis-
cursive field. Given the 'surplus of meanings' that characterize the discursive fields
of most policy issues, as illustrated in Chapter 2, a candidate for hegemony must
compete with a wide range of potentially relevant but different discourses. Within
the discursive field, as Battistelli and Ricotta (2001: 4) write, 'to become hegemonic
the specific discourse needs to dominate the field in which it was formed,
blocking the flow of differences in meaning and setting itself up as the center of

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 80 Public Policy and Discourse Analysis

interpretive processes'. This happens as the discourse succeeds in the 'structuring of

privileged discursive points, or nodal points of discourse' that facilitate discussion
about some things and not others: that is, when its concepts and definitions have suc-
ceeded in fixing specific meanings to be employed in interpreting a wide spectrum
of social issues, problems, objects, actors, and situations.
For Fraser (1989: 164) such discourse politics is understood in terms of struggles
over the interpretation of issues rather than over substantive questions. This 'politics
of conflict interpretation, includes not only the interpretations of the problems, but
also the types of discourses they should involve' (Braun and Herrmann 2001: 2).
This involves examining questions such as whose interpretation becomes authorit-
ative, as well the reason why. Are the dominant forms of political discourse available
for interpretive argumentation shown to be acceptable and fair to the parties
involved? Do the arguments favour the dominant political groups or do they assist
challengers in making their oppositional claims? To get at these questions the
interpretive analyst has to examine the social and institutional logics that underlie the
interpretive processes. In which societal institutions are the interpretations
developed, and what forms of social relations bind interlocutors together?
Towards this end, discourse analysis must also include a focus on the available dis-
cursive resources in a given social situation and how they are stratified in accordance
with societal patterns of dominance. Discursive resources involve accepted idioms
in which claims can be advanced, such as talk about rights, or economic talk; avail-
able vocabularies, such as medical, religious, or feminist vocabularies, frameworks
of argumentation (especially as they pertain to debates among citizens, experts,
politicians, and others), and narrative approaches or conventions that are available
for the construction of stories and their telling. Especially important in Fraser's
scheme is an emphasis on modes of subjugation. How do different discourses posi-
tion or subjugate participants in a communicative exchange (for example, are they
experts or citizens, politicians or businessmen, and so on?).
Such an orientation would also require us to examine the distribution of discursive
capabilities across a society (Beck 1986; Schiller and Symthe 1992). Some people
can speak regularly, while others can hardly ever be heard, unless they threaten viol-
ence. A key aspect in modern society is the role of the media, particularly how it is
controlled. As is well documented, the media in the United States and Europe is
increasingly controlled by smaller numbers of people and the content is more
and more influenced by the advertisers who pay for the programmes. This strongly
supports the ideological complex with a distinctive consumer-oriented business
bias. Moreover, much of the news is limited in the perspectives it offers.
Consequently, such research shows that modern citizens are socialized more into
roles as passive members of the audience than active participants. From this
perspective, education and access to information and information technologies, such
as computers and the Internet, as resources becomes an important topic. Some
writers such as Beck (1995) have, in this respect, described the defining tensions of
the twenty-first century in terms of a conflict between those who have expertise and
those who do not.

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 Public Policy and Discourse Analysis 81

Fraser distinguishes between the dominant means communication and interpreta-

tion that are officially approved or sanctioned, and the subordinate discourses that are
normally denied easy access to the primary discursive arenas. Discursive politics,
from this perspective, is an essential strategy of political resistance (Fraser 1989:
165). The arguments of the subordinate groups, argues Fraser, cannot be 'translated
into the dominant discourses without being redefined in ways that shed important
meanings'. Towards this end, Fraser calls for the development of 'social spaces'
where groups with unequal discursive (and non-discursive) resources can deliberate
about different ways of talking about social issues and problems.
Discursive understanding of power and political struggles, as Fraser's
Habermasian focus also makes clear, holds open the possibility of moving beyond
the traditional negative connotations associated with ideology and power. In much of
social science power and ideology are understood negatively; they refer to domina-
tion, manipulation, and control. Missing from this view, as we noted in Chapter 2, is
the more positive understanding of discourse. Communicative power, as Habermas
and others have made clear, can create new consenses that open the way to altern-
ative identities and courses of action. Moving beyond domination or the mobiliza-
tion of resources, discursive power is productive power. It is, as Litfin (1994) puts,
it, 'capacity-giving', a topic to which we return in later chapters.

POLICY DISCOURSEAND ARGUMENTATIVE STRUGGLE

Can we apply the discourse model more directly to the policy research? Such work
is increasingly emerging. In Governing Molecules, for example, Gottweis (1998;
2002), has examined the discursive construction of genetic engineering in Europe
and the United States to show that the genetic controversy is a process inseparable
from the social construction of political, economic, and scientific worlds. Calling for
a 'poststructural policy analysis', Gottweis employs discourse-analytic methods
based on insights from Derrida and Foucault to demonstrate the need to examine
how discourses and narrative stories that create orientation and meaning constitute
the policy field of genetic engineering. Critical of the assumption that political sci-
ence can simply assume the 'existence' of actors and structures in politics, as well as
the narrow understanding of politics that has informed conventional political stud-
ies, he argues for 'conceptualizing the process of policymaking as situated at the
intersection between forces and institutions deemed "political" and those appar-
atuses that shape and manage individual conduct in relations to norms and objectives
but are deemed "nonpolitical," such as science or education.' Rather than 'assuming
stable boundaries between sectors such as politics, the economy, and science', he
calls for the study of 'the micropolitics of boundary drawing' (Gottweis 1998: 27).
Through this perspective Gottweis (1998) shows that the politics of genetic engin-
eering cannot be reduced to citizens demanding the regulation of recombinant DNA
technology. The politics of genetic engineering also takes place when scientists carry
out specific experiments in genetic engineering and subsequently make claims
about the importance of their research for the understanding of both nature and

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 82 Public Policy and Discourse Analysis

social behaviour. This approach allows Gottweis to identify a much more sophisti-
cated regime of 'governability', a system of fields and sites ranging from research
laboratories to legislatures that negotiate and deploy approaches designed to control
and manipulate genetic materials, or as he puts it, 'to make them governable'.
In Ozone Discourses, Litfin (1994: 3) offers a discourse or 'reflective' approach to
the 1987 negotiation of the Montreal Protocol (on substances that deplete the ozone
layer). Essentially, it is a story about how a dominant anti-regulatory discourse suc-
ceeded in supplanting a new environmental regulatory discourse. Through the nar-
rative discourse she illustrates how mainstream functionalist, agency-centred, and
epistemic community approaches downplay, even ignore, the ways in which the informa-
tion of scientific experts mainly reinforces or reinforces existing political conflicts.
On the surface of the matter, this landmark ecological treaty would seem to be the
outcome of a rigorous risk analysis process and astute diplomatic negotiations. With
complex atmospheric models as the scientific basis for negotiation and decision-
making, this account has led writers such as Haas (1989; 1992) to explain the
Montreal process in terms of an epistemic community composed of atmospheric sci-
entists. In this view, knowledge is taken to be a body of concrete and objective facts
that were not only relevant to the actors' perceived interests, but were fundamentally
implicated in questions of interpretation and policy framing. Knowledge provided
the means that reoriented actors' understandings of their own interests, thus making
the agreement possible.
Litfin (1994: 46-51), in sharp contrast, offers a very different picture of what hap-
pened. The 'knowledge' at issue was itself 'framed in light of specific interests and
preexisting discourses so that questions of value were rendered as questions of fact,
with exogenous factors shaping the political salience of various modes of interpret-
ing that knowledge', including the discovery of an ozone hole of the Antarctic that
helped to empower the subordinate regulatory discourse. By neglecting issues of
interpretation and legitimacy, mainstream theories offer explanations that fail to cap-
ture the multi-dimensional interpretive and interactive processes that link science to
politics. In so far as these reflective elements are essential to knowledge-based
power, she argues that standard accounts have to be extended to include a discursive,
productive, or capacity-giving conception of power.
In The Politics of Environmental Discourse, Hajer (1995a) has analysed the pol-
itics of the acid rain controversy to show how different applications and understand-
ings of acid rain have generated very different types of politics in two countries.
Because Hajer's concept of discourse coalitions most directly engages the analytical
and methodological debates in policy analysis, particularly in terms of his critique of
the influential work of Sabatier and Jenkins-Smith on advocacy coalitions, we turn
to a more detailed examination of his contribution in the section below.
Devoted to a comparison of acid rain policies in Britain and the Netherlands, Hajer
offers a full-scale study of discursive policy analysis. Building on Foucault's theoret-
ical work, Hajer labours to translate this lofty theoretical contribution into a frame-
work for the analysis of concrete policy problems. The task, as he explains, is to
understand how particular discourses in a distinct political domain come to influence

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 Public Policy and Discourse Analysis 83

discourses in other domains. Towards this end, he draws on discursive concepts

developed by the social psychologist Billig designed for a less abstract 'level of
interpretive interaction'. The result is a 'social-interactive discourse theory' geared to
the analytical tasks of policy analysis.
In Hajer's scheme, as in Foucault's, the dominant discourses supply social actors
with 'subject-positions' that define their social and power relationships in terms of a
principal narrative or narratives (1995a: 65). Discourse analysis shows the way in
which actors attach meanings to other actors in terms of the positions assigned or
attributed to them in the dominant or counter-discourses. Actors are 'positioned' with
regard to specific social attributes such as blame, and responsibility. In this view,
social actors mainly make sense of the world by borrowing terms and concepts from
the discourses made available to them in their social groups and the society generally.
That is, to be understood and considered relevant, a speaker has to situate his remarks
in—or relate them to—the recognized discourses in use at the time. A speaker's state-
ments must say something in terms of the ongoing flow of discursive exchanges.
Ideas that do not draw on or interact with the available discourses will be dis-
missed as strange or irrelevant. Put more concretely, before Adam Smith came along
and offered a fully developed discourse on the virtues of the free market, it was dif-
ficult, as economic historians show, for the small businessman to express or convey
such interests and concerns. Now capitalist discourse has long been one of the avail-
able discursive themes to which one can appeal—approvingly or disapprovingly—in
an effort to be understood. Today, one can say the same of environmental discourse.
Forty years ago it was difficult to express concern about the environment without
confronting furrowed brows. Now it is more than clear what is at stake when one
refers his or her argument to environmental concerns.
Actors in this discourse-analytic approach are constituted by discursive practices,
meaning they have to be understood through the concepts and languages employed
to describe their activities. In the case of the public administrator, for example, con-
ventional social science typically defines a public administrator in terms of his or her
functional role in the agency. But a discursive analysis would examine the other
stories co-workers tell that interpret the administrator's behaviour in different ways,
for example as party functionary, dedicated man of the people, or small-minded
bureaucrat, or perhaps all three, depending on the occasion.
Rather than a set of ritualized roles, social interaction is thus conceptualized as an
exchange of arguments, of competing, sometimes contradictory, suggestions of how
one is to make sense of reality. For political and social research this means exploring
discourse practices through which social actors seek to persuade others to see a par-
ticular situation or event. In this dialectical or perspectival view of knowledge, it is
important to know the counter arguments and positions of the others involved
(MacDonell 1986: 43-59). As discursive arguments always take their meaning in
opposition to other positions, the analyst always has to pay attention to the dialectical
tensions among rival positions, a point to which we return in Chapter 6. The discurs-
ive fields in which communicative exchanges take place thus need to be a central
interest of analysis. 'Language', as Shapiro (1981: 14) puts it, 'becomes part of data

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 84 Public Policy and Discourse Analysis

analysis for inquiry, rather than simply a tool for speaking about an extra-linguistic
reality.'
Underlying this approach are two basic analytical tasks, one concerned with the
substantive content of discursive statements and the other related to their implica-
tions for institutional practices. With regard to discursive content, one can easily
grasp the point by comparing the discourses of physics and environmentalism.
Physics is a discourse, produced, reproduced, and transformed through practices
like academic teaching, laboratory experiments, and peer-reviewed journals.
Environmental discourse, in contrast, is produced through activities such as prac-
tising of alternative life-styles, engaging in political demonstrations and protests,
making references to pristine forests, reciting folklore about nature, and the reject-
ing consumer culture. Both discourses have their own context or domain. No one
can successfully use the language or modes of argument of physics in everyday
discourse. And the same holds true of the language of environmentalism; it does not
get one very far in the physics laboratory.
Hajer's specific concern is with the discursive content of environmental policy dis-
course, yet another kind of discourse with its own characteristics. While we can iden-
tify a rational or systematic basis for policy discourse, it lacks the kind of coherence
of content found in formal discourses such as physics or law. Legal discourse, for
instance, is a highly structured, internally consistent mode of deliberation with its
own inherent criteria of credibility. Its systematized institutionalized practices are
reflected in the procedures of lawsuits, the preparation of legislation, appeals
processes to higher courts, and the like. Policy discourse is also dependent on its
institutional environment, but it cannot achieve the kind of internal coherence of a
discourse such as law, or even political theory for that matter. This has to do with its
position at the intersection of various discourses, combining a range of discursive
components—empirical, institutional, pragmatic, and normative factors—that give it
its unique character and requirements. In the case of regulatory policy, for instance,
Jasanoff (1990; 1995) has captured this in her concept of 'regulatory science', which
she juxtaposes to 'science' per se. Where the later follows the logic of empirical
causality, regulatory science involves an interdisciplinary interaction between legal,
economic, scientific, and administrative discourses. Each contributes to what can
become consensually established as a set of rules designed to judiciously rule over a
particular policy domain.
In Hajer's (1995a: 53) social-interactive approach, the interplay of such discourses
is not to be conceptualized as 'semi-static plays in which social actors have fixed and
well memorized roles of such as environmentalist, policy-maker, scientist, or indus-
trialist', or the firm belief systems stereotypically attached to these roles. To the
contrary, politics becomes an argumentative struggle in which actors not only try to
make their opponents see the problem according to their view, but also seek to posi-
tion or portray other actors in specific ways (for example, by showing the relation-
ship of a particular argument to the interests of an industrialist) that open up new
possibilities for action. Grounded in qualitative methods, the discourse-analytic task
is seared to the real world of social action. The evidence for this is found in the

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 Public Policy and Discourse Analysis 85

actors themselves. Discourse is not a category merely imposed on them by social

scientists; rather, it is something social actors constantly practise.
A significant part of the task is to identify the elements in argumentative interac-
tions that are essential to explaining the emergence and persistence of particular
discursive constructions. How, for example, do concepts such as 'sustainable devel-
opment', 'acid rain', or 'globalization' get constructed and shaped? What kinds
of forces—both political and rhetorical—are brought to bear to establish such
discursive formations, and how do they affect the behaviour of both individuals and
social groups? Once established, how can such constructions establish discursive
hegemony over a particular policy domain?
A social-interactive discourse approach takes social actors to be actively engaged
in choosing and adapting thoughts, shaping and fashioning them, in an ongoing
struggle for argumentative triumph over rival positions. Taking here a position that
would seem to draw as much or more from Habermas than Foucault, Hajer sees the
task of the discursive policy analysis as uncovering the defining claims of a particu-
lar position. That is, we need to examine the structure, style, and historical context
of an argument to determine why some modes of argumentation serve to effectively
justify specific actions in particular situations and others fail. The discourse analyst
needs to examine a particular conception of reality as something that is upheld by
key political actors through discursive interaction, as well as how the oppositional
forces seek to challenge these constructs.
Most fundamental to these discursive interactions is the question of how a particular
framing of an issue can bestow the appearance of problematic on some features of a dis-
cussion while others seem proper and fixed. The task of the analyst is to show whether
particular definitions 'homogenize' a problem, that is, render the problem understand-
able by situating it is a wider social frame, or whether definitions lead to a 'heterogen-
ization' that opens up established discursive categories and hence the possibility of new
courses of action. In the process, the opening up, or 'deconstructing', of a policy dis-
course can show how it emerges as the unintended consequence of a confluence of
events and ideas. It can also show how seemingly technical issues can conceal norm-
ative commitments, as well as what sorts of institutional arrangements make this pos-
sible. Discourse analysis, in this way, helps us see which institutional dimensions are
firmly entrenched and which structural elements are more open to change.
Important to understand is how social actors can contribute to the production and
potential transformation of a discourse. In this respect, Hajer (1995a: 55) adopts an
'immanentist' view of language. Like Marx, Hajer argues that discursive actors can
shape their own history, but not under conditions entirely of their own choosing.
Given that they are operating in pre-existing social structures, the political context of
interaction has to be situated in a discursive construction that offers both possibilities
and constraints. Oppositional groups must recognize that various modes of expression
have meaning or effect only to the extent that they are appropriate to the context in
which they are uttered. Because the rules and conventions that constitute the social
order have to be constantly reproduced and reconfirmed in actual speech situations—
whether in documents, media coverage, or public debates—the challenge for such

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 86 Public Policy and Discourse Analysis

groups is to find ways to break open these routines. In general, this happens through
the discovery of contradictions or paradoxes.
In view of the complex multi-dimensionality of a given social configuration, the
particular conceptions of reality that ruling groups advance to legitimize existing
institutions must necessarily cover up contradictions and paradoxes. The task for the
opposition is to open and exploit these ideological tensions and contradictions by
showing how they function to hide or conceal other realities (Mannheim 1936). At
times, such openings are facilitated by the emergence of exogenous crises or shocks
to the social system that make these tensions either visible or difficult to conceal.
However, unlike Sabatier's (1988) emphasis on external shocks, which tend to play
a quasi-deterministic role, the crucial issue from a discourse perspective is the kind
of strategic communications to which the such a crisis gives rise.
Discursive responses to a crisis are in no way automatic. The ways in which they
are portrayed can vary widely, with very different implications, ranging from 'there is
little that can be done; we must accept it' to 'dramatic action must be taken—those
responsible for allowing it to happen must be punished'. Compare here the rhetoric of
President Herbert Hoover during the Great Depression with that of President Franklin
Delano Roosevelt. Where Hoover could only ring his hands in despair ('There is noth-
ing more we can do'), Roosevelt energized the nation to rally around a new set of
institutional arrangements ('There is nothing to fear but fear itself). Without the
strategic discursive responses of a dynamic new leader, as Houck (2001) shows, the
course of American history would have surely been quite different. Nothing is given
in such political developments. How discursive responses will unfold are case-
specific, depending on a wide range of factors from how those who have power choose
to exercise it, to the personal characteristics of the particular actors on the stage at the
time, what sort of orientation the media is disposed towards, the mood of the public,
the resources of different groups, other related contextual factors, and more. The point
is that a systems crisis itself does not determine what follows; rather, it is only the
trigger for discursive responses that open or close off possible courses of action.

POLICY STORYLINES
The basic linguistic mechanism for creating and maintaining discursive order, or
responding to a destabilizing jolt to the discursive order, is the concept of the story-
line. A storyline, Hajer (1995a: 56) writes, 'is a generative sort of narrative that
allows actors to draw upon various discursive categories to give meaning to specific
or social phenomena.'(For example, 'there is nothing we can do' or 'we must take
immediate action'.) The primary 'function of story-lines is that they suggest unity in
the bewildering variety of separate discursive components parts of a problem' that
otherwise have no clear or meaningful pattern of connections.
To understand the world around them, most people do not draw on comprehensive
discursive systems for their cognitions. Instead, they rely on storylines. That is,
they do not appeal to well-developed theories of ecology or political philosophies for
cognitive assistance, but rather turn to discursive storylines as short-hand constructions.

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 Public Policy and Discourse Analysis 87

Storylines, in this way, function to condense large amounts of factual information inter-
mixed with the normative assumptions and value orientations that assign meaning to
them. As social constructions of particular events, storylines serve to position social
actors and institutional practices in ongoing, competing narratives. In the process, they

Globalization as Storyline
For the purpose of clarifying the concept of a storyline, the current politics of global-
ization provides an excellent illustration of the struggle to define and interpret reality.
Consider the anti-globalization protests that have taken place in Seattle, Gothenburg,
Genoa, New York, and so forth. Political elites portray or 'discursively position' pro-
testers as irresponsible anarchists hostile to an inevitable good that cannot—and should
not—be impeded. The actions of such people, according to President George W. Bush's
political narrative, will only deny the poor of the underdeveloped world the economic
benefits of globalization. Against this dominant storyline, which is widely supported by
the media, opponents of globalization and others sympathetic to their views argue that
the demonstrators are not simply anti-global, if anti-global at all. Rather they protest
against a particular capitalistic version of globalization that currently is beyond control.
In this view, the dominant political leaders are seen to only speak for transnational cap-
italists, whom they themselves cannot control, as the globalization process transcends
the nation state. Since globalization is taking place in the total absence of democratic
institutions, where the discussion of the development of globalization could take place,
the demonstrators turn to the only forum available to them, namely, the streets.
The point here is that there is no objective or fixed narrative construction of the
demonstrators and what their actions mean. Indeed, both interpretations of the actions of
the protestors may have truth to them. Much of the politics surrounding the issue,
however, involves a discursive struggle to establish a working definition of these people
and what they are up to. Regardless of what the final outcome will be, these meanings are
at present up for grabs and the outcome will depend on more than who has the most
power. In this respect, the ability to cleverly shape the discursive space can have an import-
ant and even decisive impact on how this struggle with evolve. The media will provide the
principal form for these debates, as that is where most of us will follow these events. Each
side will vie to place its own linguistic tropes on the events seen on television.
In the case of the protests in Genoa, the first media reports on German television
simply showed people throwing stones and referred to them as 'chaotic' people. This was
a clear expression of the hegemonic position, which a few days later waned as evidence
of police brutality began to emerge and movement leaders began getting the alternative
story across in the media. Prominent politicians, especially those supportive of the
demonstrators, began to call for an investigation of the Italian police force in Genoa.
The winner of the discursive struggle will define what will be taken to be reality with
categories that will at the same time suppress alternative conceptions. Through the delin-
eation of categories and practices, for example, the defenders of globalization will seek
to extend and expand a discursive field that will enable the continuing expansion of the
transnational-capitalist approach to globalization. Those who speak an oppositional
discourse will continue to be portrayed as either outmoded or anarchist. In the process,
their questions and arguments will be suppressed.

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 88 Public Policy and Discourse Analysis

stress some aspects of an event and conceal or downplay others (covering over potential
problems—tensions or paradoxes—that might be embedded in the story).
Maintaining the social order depends on the successful reproduction of particular
understandings of the relevant actors and institutions—that is, the capitalist and the
free market economy require legitimization and active participation. At the same
time, critical tensions in the free-market storyline ('the capitalist is only interested in
his profit margin') coupled with the emergence of counter storylines based on new
problematic events or revelations (such as a sudden and dramatic increase in oil
prices) open the way for efforts to introduce stronger market regulations. Finding or
reconstructing the appropriate storyline becomes a central form of agency for the
political actor. Effective communicative skills and argumentative strategies are thus
important resources that play a significant role in determining the shape of events.
Like an art form, such communicative skills hold out the possibility of making peo-
ple see things differently, and, in the process, shifting the course of political struggle.
The relevance of discursive strategies is particularly evident in environmental pol-
itics. The environmental movement has long been haunted by the dilemma of
whether to argue on the terms set by the government or to insist on their own mode
of expression. In the latter case, of course, the movement runs the risk of losing its
direct influence and therefore often trades its expressive freedom for influence on the
policymaking process.3 Given a hegemonic discourse, people who try to challenge
the dominant storyline are often expected to position their contribution in terms of
established categories. Indeed, this is the primary way a hegemonic discourse
exercises it power. The concept of 'sustainable development' is a good example of
the process.4 Because the earlier environmental storyline, 'limits to growth', proved
to be a non-starter for the industrial community (as capitalist industry exists to
grow), there was a need to innovate a new storyline capable of working for both
environmentalists and industrialists. The response was the concept of 'sustainable
development'. Shaped by leading industrials, it refers to environmental protection
but without sacrificing industrial development, which environmentalists have long
defined as the very source of pollution.
Environmentalists can—and often do—reject the terms set by the dominant
discourse, insisting on the importance of their alternative discourse. The disciplinary

3
The point is well illustrated by the Green Party in Germany. The party gained political influence by
joining with the Social Democratic Party to form the governing coalition in 1998. But the Greens' posi-
tion as junior member of the coalition caused the party to soften its more critical political positions, such
as opposition to military engagements, at considerable cost. In response, many party loyalists resigned
their memberships, charging the leadership with having politically sold out in the interest of gaining
power. In the process, the party created political identity problems for itself that have led many voters to
see the party as a superfluous feature of the German political landscape.
4
The concept of sustainable development was advanced by the Bruntland Commission and was the
focus of discussions at the second international Earth Summit in Rio de Janeiro. The Commission denned
sustainable development as a 'framework for the integration of environmental policies and development
strategies—the term "development" being used in the broadest sense of the term'. Sustainable develop-
ment is designed to meet 'the needs of the present without compromising the ability of future generations
to meet their own needs' (World Commission on Environment and Development 1987: 42).

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 Public Policy and Discourse Analysis 89

force of discursive practices, as Hajer (1995a) explains, often depends on the

implicit assumption that subsequent speakers will respond within the same discur-
sive frame. Such discursive challenges may consist of rejecting an understanding of
events and actions in terms of established categories or, often even more powerfully,
in establishing new combinations with the operative, routinized discursive struc-
tures. Indeed, this is what much of contemporary environmental politics is about.
Although this social-interactive theory of discourse focuses on how people take
up or deny particular positions, it also recognizes the ways in which power is
embedded in the reifications of a discourse and, as such, remains hidden. People use
particular terms without giving much thought to the way they sustain a specific
social order. Merely consider the difference between saying 'Negro' and
'Blackman'. Both words refer to the same physical person, but each rests on and
reproduces a very different conception of power relations between the races. Many
people simply assume that such a word is a formal name for such a person with a
particular skin colour, or that it is 'just the way one talks' on particular occasions,
without being explicitly aware of (or taking into account) the social implications
embedded in the language. In such cases, people can support a given power structure
while explicitly denying that they are necessarily doing so. They do not use the spe-
cific words, but their listeners read racist meanings into them. Because routine forms
of accepted discourse generally express continuous but opaque power relationships,
they can also be especially useful to particular groups seeking to convey particular
social messages while avoiding direct confrontations.5
Through an analysis of the social meanings embedded in the discursive practices
of the societal and political institutions, the analyst can thus investigate the nature and
reproduction of the power structure. At the 'macro' level of analysis, the investigator
is able to focus on how political and economic elites construct and maintain a
societal-wide hegemonic discourse that makes clear what is on the agenda and what
is not, as well as how oppositional groups seek to make their voices heard. Today we
can easily understand this in terms of the construction of a market-oriented neo-lib-
eral ideology that now dominates not only Western industrial countries, but much of
global discourse as well. The success of such a hegemonic discourse is in significant
part measured in terms of the difficulties it poses for those who argue for government
interventions in both the economy and the provision of social redistributive pro-
grammes. At the 'micro' level of analysis, the focus turns to the relationships between
discourse and specific institutional practices. The goal, following Foucault, is to
undercover the ways that discourses embedded in institutional practices function to
reproduce the existing power-structure relationships from the bottom up. Here,
through the workings of professional expertise, ideas and knowledge can serve
as an independent force stabilizing or cementing society together. In terms of neo-
liberalism this can take the form of cost-benefit oriented policy analysis in agency

5
A good example here is the political rhetoric of Joerg Haider, the right-wing Austrian politician, who
uses words employed by the Nazis. The words say nothing about Nazis but they serve to recall specific
memories of a particular political past.

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 90 Public Policy and Discourse Analysis

bureaucracies, which implicitly insures the primacy of market criteria, over in

preference to social agency practices that emphasize 'workfare' counselling. One the
surface, cost-benefit analysis looks like—and is presented as—a scientific technique
for rational data collection and decision-making. But most people do not see that
the data collected by the methodology channels discussion in directions biased toward
a market orientation of society (Fischer 1987: 117).
Emphasis on institutional practices also helps to clear up the widespread mis-
understanding that such argumentative research just deals with languages and, in the
process, loses connection with the 'hard' institutional realities of policymaking. To
the contrary, for the kind of discourse analysis advanced by Fairclough and Hajer,
understanding what the relevant utterances mean requires approaching the institu-
tional setting as an 'argumentative field' in which statements are made. The focus is
thus not simply on the language or discursive constructs but on the ways utterances
relate to the specific institutional contexts and practices in which they can be mean-
ingfully stated and understood. Indeed, Hajer (19956) takes the point a step further
and argues that we need engage in 'institutional constructivism'. This involves
analysing which sorts of practices actually produce or facilitate particular policy
changes. In the acid rain case, for example, one of his most interesting discoveries is
that, after the environmentalists had succeeded in publicly winning the policy
debate—at which point everybody began speaking the environmental language, or a
version of it—no policy change could be detected. Why? Because the technically
oriented practices of the environmental agencies remained unchanged. Although the
public political discourse had accepted the environmental understanding of acid rain,
the environmental ministries continued to monitor the problem with their traditional
methodologies (Hajer 1993). The practices themselves determined the policy,
despite the new environmental discourse.

CONCLUSION
Discourse, as we have seen in this chapter, is an ensemble of ideas and concepts that
give social meaning to social and physical relationships. Operating at both the macro
and micro levels of society, discourse both transmits the cultural traditions of soci-
ety and mediates everyday social and political interactions. Focusing on the rela-
tionship between discourse and social practices, the task of the discursive analyst is
to explain how specific discourses become hegemonic, explicate the characteristics
of discursive fields (including the nodal points that privilege some arguments over
others), identify the defining claims of the particular positions, clarify how indi-
vidual discourses come to influence others, determine the structure of the arguments,
identify which styles of discourse make them effective in given contexts, uncover the
ways that the discursive resources are distributed across social systems, and show
how particular socio-historical constellations serve to justify specific courses of
action. Emphasizing the ways discursive production socially constructs reality,
such analysis examines the way hegemonic conceptions of reality are upheld and
reproduced by key political groups, while oppositional groups seek argumentative

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 Public Policy and Discourse Analysis 91

strategies to challenge these dominant social constructs. Although discourses take

place in pre-stractured contexts that limit or impede the range of possible actions,
persuasive discourses can even in the face entrenched social and material forces
open new paths to action.
More specifically, we saw the ways discourses establish the terrain on which polit-
ical struggle takes place. Assigning subject-positions to social actors in both cultural
narratives and ongoing storylines, they define the actors' social and political rela-
tionships, attributing to them various social attributes such as virtue or blame. In
argumentation at the micro or practical level, participants exchange competing and
sometimes contradictory arguments to make sense of social reality, including the
interpretation of political conflicts themselves. Operating in a field of existing
narrative discourses, social and political actors are seen to choose and adapt thoughts
and ideas in an ongoing effort to triumph over rival arguments.
The nature of these discursive interactions is further explored in the next chapter,
which presents a comparison of Sabatier's theory of advocacy coalitions withHajer's
theory of discourse course coalitions. Of particular importance, the comparison
offers the opportunity to take a closer look at the essential role of discourse and
interpretation in understanding the processes of social change. Where the former
seeks to advance an empiricist theory of policy change, the latter details the postem-
piricist discursive alternative. The two approaches are compared in terms of a con-
crete policy issue, environmental protection.
Finally, it should be noted that most of the points in the discussion were made or
illustrated through policies pertaining to environmental politics. In significant part,
this is because the fields of environmental and science studies, given their close rela-
tionship to social constructionist perspectives, have generated a great deal of the
research about policy discourse. But this should not be taken to limit the value of the
perspective. In the expert-oriented information society in which we live, the role of
discursive politics in policymaking is expected to become increasingly apparent in
other policy areas as well. If, as some have argued, the struggle between those with
and those without knowledge is one of the key socio-political dynamics of this new
century, the discourses of knowledge in their various forms will be a central aspect
of political conflict across the policy spectrum.

APPENDIX
DISCOURSE AND SOCIAL CHANGE:
COMMODIFYING EDUCATIONAL POLICY
Fairclough's (1992: 207-15) analysis of educational commodification begins through a textual
analysis illustrating the way particular words and images are mixed together to create a social
meaning. Towards this end, Fairclough focuses on the how the word 'skill' is employed in a
university catalogue designed to attract and inform students. The text is structured to impli-
citly—if not explicitly—inform the students that they need skills to get ahead in their lives
and that the educational institution in question is well prepared to supply them with the
appropriate skills.

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 92 Public Policy and Discourse Analysis
After having analysed the linguistic presentation of the text, Fairclough then turns to the
analysis of discursive practices to show the way the word 'skill' is given a double meaning.
By strategically overlaying a term of business discourse on an educational discourse, educa-
tional planners and curriculum designers create a rhetoric of skills that can be interpreted in
more than one way. As a commodified educational discourse becomes 'dominated by a voca-
bulary of skills, including not only the word "skill", but a whole wording of the processes of
learning and teaching based upon concepts of skill, skill training, use of skills, transfer of
skills, and so forth', the concept permits 'two often contradictory constructions of the learner
to coexist without manifest inconsistency, because it seems to fit into either an individual view
of learning, or an objectivist view of training'. In addition to the semantic history of the word,
as Fairclough writes, 'this ambivalence is reflected in the history of the concept within liberal
humanist and conservative economic discourse'. The concept of skill, on the one hand, 'has
active and individualist implications: skills are prized attributes of individuals, individuals
differ in types and degree of skill, and it is open to each individual to refine skills or add new
ones'. The concept of skill, on the other hand, 'has normative, passive and objectifying
implications: all individuals acquire elements from a common social repertoire of skills, via
normalized training procedures, and skills are assumed to be transferable across context,
occasions, and users, in a way which leaves little space for individuality'.
Drawing on laissez-faire, market-oriented economic discourse, 'the message to teachers and
course planners is a subtle version of the marketing axiom of consumer sovereignty: "Give the
customers what they want"'. This juxtaposing of 'wordings facilitates the metaphorical transfer
of the vocabulary of commodities and markets into the educational order of discourse'. But, at
the same time, 'this commodified educational discourse is commonly more self-contradictory
that this might suggest'. A clue to such contradictions in the blurring of clients and consumers
is an ambiguity about to whom educational packages are being sold. Who is doing the learn-
ing? Is it the students or the companies that are likely to hire the students? The company may
in fact need clients 'in the direct sense of paying the learner to take a course'. In this way,
'learners are contradictorily constructed'. They are constructed 'in the active role of discerning
customers or consumers aware of their "needs" and able to select courses which meet their
needs'. But they are constructed, at the same time, 'in the passive role of elements or instru-
ments in the production processes, targeted to train in required "skills" or "competences",
courses designed around precise "attainment targets" and culminating in "profiles" of learners,
both of which are specified in terms of quite precise skills'. This subtle intermingling of active
and passive constructions of student learners 'facilitates the manipulation of the learners
through education, by overlying it with what one might call an individualist and consumeristic
rhetoric'.
From the perspective of social practices the university has conflated these separate dis-
courses in such a way that, while they benefit different participants differently, they gradually
come to be taken as the same thing. That is, obtaining skill comes to be seen as the essence of
education, although the curriculum for these skills is no longer shaped around the liberal arts
components, but rather the needs of industry. Instead of the traditional understanding of edu-
cation as a way for the individual to broaden his or her relationship to the world in which he
or she lives, thus opening up new possibilities, it is in fact narrowed down to a package that
is shaped to meet the needs of business and industry rather than the individual per se. This is
not to say that the individual cannot gain from this restructuring, but it is to say that the
choices are dramatically limited to those that also fit the interests of the employers. Subtly,
though, in the process the two come to be conflated. The university tries to say at the same
time, 'come to us and get a well-rounded education', but at the same time it is saying 'come

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use
 Public Policy and Discourse Analysis 93
to us and get a job'. The two are not the same and the tension has to be mediated. This
happens through the concept of skill, which blurs the tension in subtle ways. In the process,
the university gets restructured. It becomes more of a training centre for industry, with the
business school more and more emerging as the primary faculty, rather than a traditional
centre for knowledge and learning for itself. Whereas the traditional university was adminis-
tered by the faculty, as they possessed the knowledge of the curriculum and could best decide
the issues of learning, today the deans become managers and the faculty members are increas-
ingly their subordinates. The facade of faculty governance remains, but it increasingly serves
to hide the real nature of institutional power and control.

EBSCOhost - printed on 6/9/2020 4:41 PM via PONTIFICIA UNIVERSIDAD JAVERIANA - BIBLIOTECA CENTRAL. All use subject to
https://www.ebsco.com/terms-of-use