Vous êtes sur la page 1sur 262

RHETORIC, POLITICS AND SOCIETY

GENERAL EDITORS: A. Finlayson; J. Martin; K. Phillips

DEBATES, RHETORIC AND


POLITICAL ACTION
Practices of Textual Interpretation
and Analysis

Claudia Wiesner,
Taru Haapala,
Kari Palonen
Rhetoric, Politics and Society

Series Editors
Alan Finlayson
University of East Anglia
Norfolk, United Kingdom

James Martin
Goldsmiths, University of London
London, United Kingdom

Kendall Phillips
University of Syracuse
Syracuse, New York, USA
Rhetoric lies at the intersection of a variety of disciplinary approaches
and methods, drawing upon the study of language, history, culture and
philosophy to understand the persuasive aspects of communication in all
its modes: spoken, written, argued, depicted and performed. This series
presents the best international research in rhetoric that develops and
exemplifies the multifaceted and cross-disciplinary exploration of practices
of persuasion and communication. It seeks to publish texts that openly
explore and expand rhetorical knowledge and enquiry, be it in the form
of historical scholarship, theoretical analysis or contemporary cultural and
political critique. The editors welcome proposals for monographs that
explore contemporary rhetorical forms, rhetorical theories and thinkers,
and rhetorical themes inside and across disciplinary boundaries. For infor-
mal enquiries, questions, as well as submitting proposals, please contact
the editors: Alan Finlayson: a.finlayson@uea.ac.uk James Martin: j.martin@
gold.ac.uk Kendall Phillips: kphillip@syr.edu

More information about this series at


http://www.springer.com/series/14497
Claudia Wiesner • Taru Haapala • Kari Palonen

Debates, Rhetoric
and Political Action
Practices of Textual Interpretation and Analysis
Claudia Wiesner Kari Palonen
University of Jyväskylä and Technical University of Jyväskylä
University Darmstadt Jyväskylä, Finland
Darmstadt, Germany

Taru Haapala
University of Jyväskylä
Jyväskylä, Finland

Rhetoric, Politics and Society


ISBN 978-1-137-57056-7    ISBN 978-1-137-57057-4 (eBook)
DOI 10.1057/978-1-137-57057-4

Library of Congress Control Number: 2016956403

© The Editor(s) (if applicable) and The Author(s) 2017


The author(s) has/have asserted their right(s) to be identified as the author(s) of this work
in accordance with the Copyright, Designs and Patents Act 1988.
This work is subject to copyright. All rights are solely and exclusively licensed by the
Publisher, whether the whole or part of the material is concerned, specifically the rights of
translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on
microfilms or in any other physical way, and transmission or information storage and retrieval,
electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information
in this book are believed to be true and accurate at the date of publication. Neither the pub-
lisher nor the authors or the editors give a warranty, express or implied, with respect to the
material contained herein or for any errors or omissions that may have been made.

Cover illustration: © EyeEm / Alamy Stock Photo

Printed on acid-free paper

This Palgrave Macmillan imprint is published by Springer Nature


The registered company is Macmillan Publishers Ltd.
The registered company address is: The Campus, 4 Crinan Street, London, N1 9XW,
United Kingdom
Preface

‘Nothing is more interesting than politics’. This does not seem to be a


popular claim today, as ‘politics’ rather has a bad reputation and ‘popu-
lism’ is on the rise. In the present volume we defend this claim, and we
hope that our readers will be intrigued to learn the reasons for taking such
an interest in politics, to analyse and study it and be consciously engaged
in it, at least as ‘occasional politicians’, to quote Max Weber (1919, 41).
To say that politics is interesting already presupposes that we do not
know in advance what is meant by politics. If there were a well-known and
commonly shared definition for ‘politics’, debates on what is meant by
politics or ‘the political’ could reach a definite conclusion simply by check-
ing the lexicographical authorities. This is not the case. During the past
150–200 years when parliamentary and democratic forms of politics have
been high on the agenda of a number of debates, among scholars as well
as the more or less professional political agents themselves, the concept of
politics has been continuously contested. Politics, therefore, is something
that cannot be understood simply by looking at what is said about it in a
dictionary. On the contrary, the writing of dictionaries and giving defini-
tions to concepts are themselves a political activity, a rhetorical move in a
debate in history and in the contemporary context of using concepts.
We also cannot distinguish certain views of politics on the basis of the
positions of political actors. There is not one typical governmental or oppo-
sitional view, or a right or left view of politics, that has historically strug-
gled for supremacy. How politics or the political should be understood
is a matter for debate. Politics is a controversial concept, and politicians,
scholars, journalists as well as citizens at large may make ­contributions to

v
vi  Preface

the controversy. They may contest certain uses of the concept or invent
new ones, even without necessarily intending to do something new. The
debate is ongoing in the sense that there is no final authority to decide
when the debate is over nor is there any a priori limit to the possibilities
for inventing new perspectives on the concept of politics.
The conceptual history of politics underlines its contestedness.
Expressions using ‘politics’ or ‘political’ can be traced back to the ancient
Greek polis, to the city-republic as a political form, as opposed to ‘despotic’
monarchies and imperia (on the opposition between politikós and despo-
tikós see e.g. Aristotle’s Politics [Ta politikà], on the commentaries see e.g.
Meier 1980; Finley 1983). But the history of the concept in the different
European languages has not been just a translation from the Greek: the
different languages have developed their own vocabularies with different
conceptual resources. Above all, the ancient and early modern thinkers
could not have understood politics as a contingent activity, as this concep-
tual horizon was new, introduced in the course of the nineteenth century
in particular (see esp. Palonen 2006). This horizon of seeing politics as a
contingent and controversial activity is the one within which we move in
this volume.
Seeing politics as an activity, what we do in this volume is to analyse
the practice of debating as a distinct form of politics. We understand the
concept of ‘debate’ broadly, referring to a wide range of activities that can
be explicit and implicit, living and frozen, regulated and irregular, and so
on. Debate can as well as be debate between many or few agents, or debate
as in a debate event in the sense of a single actor’s action to intervene in
an ongoing debate.
The book offers typical as well as untypical examples of debate viewed
as a political activity. The examples are related in part to our own previous
or ongoing research, the convergence of which also prompted us to write
this book together. We realised that nobody had written a book on how
to analyse debates politically, and our publisher Palgrave Macmillan also
understood that this was the case.
While we aim at indicating possible paths, approaches and practices
for studying politics and debates, this is no ordinary textbook. We share
Robin George Collingwood’s well-known view, in his posthumously
(1946) published The Idea of History, that textbooks often treat the read-
ers as children who are expected to learn what is taught to them, keeping
them, as he puts it, in a condition of statu pupillari. As opposed to such
a view, we hope to encourage the independent and critical thinking of
Preface   vii

our readers, which might lead far away from what we have presented or
anticipated in this book.
This book is based on our common engagement in such projects as: The
Finnish Centre of Excellence in Political Thought and Conceptual Change;
The Politics of Dissensus: Parliamentarism, Rhetoric and Conceptual
History (the Academy of Finland research project); the Marie Curie proj-
ect Conceptualizing representative democracy in the EU polity by re-thinking
classical key conceptual clusters for the EU multi-level polity (EUPOLCON);
the Standing Group Political Concepts of the European Consortium for
Political Research (ECPR); and most recently, the Finnish Distinguished
Professorship project Transformations of Concepts and Institutions in the
European Polity (TRACE), directed by Professor Niilo Kauppi. The book
is also extraordinary in so far as we have tried to turn our different career
lengths, academic tracks, generational experiences of politics and practices
in analysing debates to our advantage in our combined efforts to write
this volume.
We are scholars in the Continental academic tradition. Although the
opposition between Anglophone and Continental political thought and
philosophy has become very relative in recent debates, our teaching and
research are still shaped strongly by the German (CW) and German-­
indebted Finnish (TH, KP) academic cultures. We consider however that,
for the purposes of this book, it is more convenient to use examples that
are based on sources written or spoken in English. This is seen in most
of the examples presented in the book, especially the ‘exercises’ con-
ducted in Chapter 4. The Hansard documentation of the British House of
Commons debates from 1803 to 2003 has provided the main source for
two sections, the Oxford and Cambridge Unions the source for another
section (see also Haapala, forthcoming 2017), and the English version of
the European Parliament’s Rules of Procedure for another (cf. Palonen
and Wiesner 2016).
Our work is also inspired by a number of continental classics, Hannah
Arendt and Max Weber in particular, or contemporary political thinkers,
such as Reinhart Koselleck and Pierre Rosanvallon. Our principle has been
to quote from the sources in their original language, although we have
paraphrased the passages in English and referred also to the existing trans-
lations, in so far as they are available, both in the references and in the
bibliography.
This book is not a volume with three editors but a joint volume of three
scholars. We think, however, that the metaphor of ‘division of labour’ is
viii  Preface

misleading to describe the art of cooperation practised in this volume.


Even if only one of us wrote the first drafts of the first three chapters, the
others have commented extensively on each of them and revised them so
thoroughly that none of them can be attributed to one of us as the main
author. To some extent the situation is different in Chapter 4, in which we
have used our different scholarly background and profiles. However, the
first example in that chapter also illustrates how each of us, when looking
at one and the same debate in the British House of Commons, has found
and focused on different aspects of it. Altogether, we hope to have grasped
some of what makes politics so interesting, and to be able to transmit this
to the interested reader.

Darmstadt, Germany  Claudia Wiesner


Jyväskylä, Finland Taru Haapala
Jyväskylä, Finland Kari Palonen
Contents

1 Understanding Debate as Politics1


1.1 The Conceptual Aspect of Politics3
1.2 Politics as an Activity5
1.3 Four Aspects of Politics: Politicisation—Polity;
Politicking—Policy9
1.4 Types of Politics—Types of Debate12
1.4.1 Debate, Discourse, Dispute, Discussion
and Dissensus14
1.4.2 Debate as pro et contra and as a Peaceful
Form of Struggle16
1.4.3 ‘Live’ and ‘Virtual’ Debates18
1.4.4 Documents as Contributions to Debates19
1.4.5 Political and Academic Debates: Similarities
and Differences21
1.5 The Presence of Politics in Texts and Debates23

2 Reading Debates Politically25


2.1 Political Literacy26
2.2 Parliamentary Debate as an Ideal Type29
2.2.1 On the History of Parliamentary Debates31
2.2.2 Extensions of Parliamentary Debates34
2.2.3 Limits of Debate36
2.2.4 Restrictions of Debate Inside Parliaments37
2.3 Forms, Dimensions and Characteristics of Debates41

ix
x  Contents

2.3.1 Rules and Procedures of Debate41


2.3.2 Debates on the Agenda and Debates
of the Agenda44
2.3.3 The Actors in Debate47
2.3.4 Regulators of Debate49
2.3.5 Multistage and Multilayer Debates50
2.3.6 The Scarcity of Time for Debate54
2.3.7 Terminating Debates56

3 Research Practices and Operations in Studying Debates


and Documents59
3.1 The Researcher’s Perspective on the Sources60
3.1.1 Keeping Your Distance While Being Part
of the Debate61
3.1.2 On the Veto Power of the Sources64
3.2 Approaches to Linguistic Political Action:
Discourse Analysis, Rhetorical Analysis
and Conceptual History67
3.2.1 Discourse Analysis68
3.2.2 Rhetorical Analysis70
3.2.3 Concepts as Nodal Points of Debates77
3.3 Analysing Debates and Documents: The Essential Steps81
3.3.1 Finding a Research Topic and Fixing
the Research Question84
3.3.2 From Research Topic to Research Question
in the Exemplar Study85
3.3.3 Determining the Research Material87
3.3.4 Material Selection in the Exemplar Study90
3.3.5 Analysing the Material93
3.3.6 Ordering, Coding and Extracting94
3.4 Categories for Analysing Debates and Documents:
Explaining the ‘How’ and the ‘Why’97
3.4.1 Numbers and Quantities97
3.4.2 Finding Out the ‘How’ and ‘Why’
in the Exemplar Study99
3.4.3 Further Useful Dimensions for 
Analysing Debates103
Contents  xi

3.5 Building Typologies, Drawing Conclusions


and Answering Questions103
3.5.1 Key Themes, Topics and Rules of the Discourses
in the Exemplar Study104
3.5.2 The Discourses in a Comparative Perspective106
3.5.3 Computer-Supported Analysis106

4 Examples of Analysing Debates as Politics109


4.1 Three Interpretations of a Parliamentary Debate:
The Case of Constitutional Renewal
in Westminster, 9 June 2009111
4.1.1 Procedure, Concepts and Time as Topics
of the Debate—Kari Palonen111
4.1.2 Analysis of a House of Commons
Debate—Claudia Wiesner121
4.1.3 A House of Commons Debate
Analysis—Taru Haapala131
4.2 Conceptual Controversies in Parliament—‘Politics’
in the House of Commons142
4.2.1 The Four Aspects of Politics
in Parliamentary Debate142
4.2.2 ‘Political’ in British Constitutional Debates
from the Late 1940s144
4.2.3 ‘Politicking’ in the Westminster Debates152
4.2.4 ‘Politics’ in British Parliament154
4.3 Reading and Analysing Political Debate
in Debating Societies155
4.4 Parliamentary Rules of Procedure:
The Case of the European Parliament161
4.4.1 Rules of Parliamentary Procedure
as Documents in Debates161
4.4.2 Conceptions of Parliament162
4.4.3 A First Look at the EP as a Parliament163
4.4.4 EP’s Procedural History164
4.4.5 French and British Procedural Styles165
4.4.6 EP’s Francophone Committees167
4.4.7 Free Mandate versus Party Groups170
xii  Contents

4.4.8 The Intervention of the Commission


and the Councils 173
4.4.9 The Parliamentary Regulation of Debate 176
4.4.10 The EP President as a Negotiator 178
4.4.11 The Politics of Time in the EP 180
4.4.12 The MEP as a Politician 181
4.4.13 Conclusions183
4.5 Doing Politics via Policy Documents and Laws:
The Case of EU Citizenship Rights184
4.5.1 The Conceptual Context: Citizenship
in Western Nation-­States and the EU 184
4.5.2 The Institutional Context 186
4.5.3 Citizenship Rights in the Treaty of Rome 187
4.5.4 Creating and Interpreting Union Citizenship 190
4.5.5 Implementing Union Citizenship: The Example
of the Right to Free Movement 192
4.5.6 Shaping the Principle of Antidiscrimination 196
4.5.7 In Conclusion199
4.6 What Is Said and What Is Not in Press Debates:
The Silencing Strategy in Germany200
4.6.1 Case 1: Marginalising EU Critics
in the Mainstream Parties 202
4.6.2 Case 2: The PDS 206
4.6.3 Case 3: Peter Gauweiler 210
4.6.4 In Conclusion212
4.7 The Politics of a Subtext: Football as a Polity215
4.7.1 Text and Subtext 215
4.7.2 The Political Subtext of Football 216
4.7.3 The Politics of Football Tactics217
Notes222

Concluding Notes: Reappraising Politics and Debate223

References229

Index243
List of Figure

Fig. 3.1  Number of articles per day 99

xiii
List of Tables

Table 3.1  Steps in the coding 96


Table 3.2  Eight dimensions of a discourse 101
Table 3.3  Studying the what, how and why in the discourse 102
Table 3.4  The main themes and levels of reference
in the French discourse 104
Table 3.5  The main rules, topics, themes and levels of reference
in the German discourse 105
Table 3.6  A comparison of the French and the German discourse 106
Table 4.1  Six dimensions in the constitutional reform debate 122
Table 4.2  Overview on the codings in the constitutional reform debate 125
Table 4.3  Three ideal types of parliamentary assemblies 162

xv
CHAPTER 1

Understanding Debate as Politics

Our aim in this book is to introduce an approach to the interpretation of


‘politics’ that focuses on debates as political action. We regard politics as
an activity, and debate as both an arena and a means of politics. Political
debates can be distinguished from academic ones, for example, by con-
sidering that the latter often are conducted for the sake of argument, or
over matters of taste in which dispute does not matter much. Political
debate relates to changing policy, alternative directions, or the forms of
power and regime of a polity, which can have consequences for the lives
of millions.
Our core idea is to present and explain how different actual and virtual
debates can be understood and analysed as political actions. We want to
provide tools for grasping the complex phenomenon of politics and new
ways to conduct its analysis. It will be shown that, by concentrating on
debate, politics can be understood in a way that allows for a more nuanced
understanding. Such an approach does justice to the aims of political
actors, taking into account their actions, interests, moves and strategies,
and situating them in relation to the different contexts in which their con-
tributions make a difference.
We intentionally include debates carried out in, or expressed by, docu-
ments. In the following chapters, we will outline our methodological

© The Author(s) 2017 1


C. Wiesner et al., Debates, Rhetoric and Political Action,
DOI 10.1057/978-1-137-57057-4_1
2  C. WIESNER ET AL.

­ ackground, approaches and practices, illustrated with examples. In the


b
first chapter, we present our theoretical and methodological background
by answering two crucial questions: Why study debate as politics? What
exactly do we mean by ‘politics’? In Chapter 2 we discuss the concept
of debate, taking the concept of ‘political literacy’ as our point of depar-
ture. Chapter 3 contains research practices and operations that are useful in
studying debates and documents. In Chapter 4 we shall present exemplary
analyses that illustrate conceptual and rhetorical perspectives in studying
politics.
Debates and documents have become important sources for the analy-
sis of politics, especially in rhetoric and discourse studies and in the history
of political thought, but also in comparative politics and European studies.
Despite this growing interest, debates and documents are often treated as
mere reflections of actors’ objectives and motives, as expressions of power
relations, or as simple preludes to the things that ‘finally matter’ such as
voting outcomes or policy outputs.
Political analysis of debates can now be conducted much easier than
only one decade ago. The digitalisation of parliamentary debates, news-
papers and other primary sources for politically relevant debates has facili-
tated both the access to and the possibility for analysing them. This should,
however, not serve as an excuse for not reading printed sources and using
libraries and archives for analysing texts and debates.
Parliamentary debates are probably the most explicit and best regulated
form of debates as political action, based on speaking pro et contra. In the
media as well as in research it is a frequent practice to interpret them for
the purpose of understanding related political processes, with the aim of
analysing or measuring ‘deeds’ or the outcomes of parliamentary votes.
This logic aims at directly relating and ‘explaining’ the words and deeds
of political agents. Such approaches risk losing sight of what is political in
debates: the activity of politics itself. We argue that besides the purposes
that debates and documents may also serve, it is crucial to see debate and
the documentation as a form of political action that transcends both party
divisions and the majorities expressed in votes. Thus, debates constitute a
crucial research material for anyone interested in analysing or understand-
ing politics.
Debates are especially important for the study of politics as no phe-
nomenon has a necessary political aspect, but every phenomenon can
become political. Things are not by nature political or apolitical, but they
must be marked, read and interpreted as such, which is itself, of course,
UNDERSTANDING DEBATE AS POLITICS  3

also a matter of debate. The political aspect of an issue or an act lies in its
relationship to its surrounding circumstances. This means that politics has
to be interpreted as such before it can be analysed. Sometimes interpreters
can detect political aspects in actions not previously considered politically
significant or, conversely, suggest that a previously major controversy has
lost much of its former political significance.
Debate differs from ordinary dialogue in that it involves a presenta-
tion of the pros and cons of an issue in which both sides actively aim at
persuading to come to their side. Whenever there is debate on a topic, it
necessarily shows the controversies embedded in it. Therefore, debates
can provide fruitful material for the political analysis of those controver-
sies. Controversial interaction (i.e. debate) between two opposing sides
is what makes it possible to analyse their arguments in relation to each
other. Perhaps the most interesting debates are those in which at least
some debaters change the sides, ‘cross the floor’ in the Westminster ter-
minology. We thus take the speeches and writings of politicians as speech
acts in debates that are to be considered as political actions in themselves,
instead of treating them as symptoms of something else.

1.1   The Conceptual Aspect of Politics


In the language of politicians and other political agents, what constitutes
‘the political’ or ‘politics’ has never been something definable only by an
entry in a dictionary or academic textbook. Neither can we point to the
existence of a few clearly distinct conceptions of politics to which political
agents should commit themselves. On the contrary, studies on the actual
usage of vocabulary related to politics easily illustrate how the concept has
multiple aspects, dimensions and historical layers that cannot be strictly
separated from each other (see e.g. Crick 1962; Marchart 2007; Palonen
2006; Steinmetz et al. 2013).
Debates very seldom show the different conceptions of politics
explicitly, even though there can be situations in which, for example,
government representatives claim that something is ‘not a political
question’, whereas members of the opposition insist otherwise. The
character of a question or phenomenon can always be turned into a
matter of dispute. Therefore, it cannot be assumed that there exists
one neutral judge that could resolve the dispute. Instead, it should be
noted that the conceptual controversy itself is part of political action
(cf. Skinner 2002a, 7).
4  C. WIESNER ET AL.

The starting point of any analysis of politics-as-action should be that the


participants in a conceptual dispute do mean what they say. This implies
that political agents are performing actions in accordance with a logic of
their own, even if they seem to be lacking competence to identify some
of the nuances of the expressions. The role of the scholars is to clarify and
make explicit the conceptual horizons of debates. Scholars can identify
what is typical in them, what aspects of the concept have been thematised
and how this has been done.
To follow Austin, some speech acts are just conventional, locutionary
statements that ‘say something’, such as ‘Britain is a member of the EU’.
Others possess a specific illocutionary point: ‘Saying something is to do
something, or in saying something we do something, and even by saying
something we do something’. Then uttering the sentence ‘Britain is a
member of the EU’ is not just an affirmation of a current state of affairs.
It can be a statement of support (fortunately, this is the case), of blame
(unfortunately, it is), of excuse (therefore we cannot change many things),
of duty (therefore Britain must carry its responsibilities) and so on. After
the referendum in June 2016, the utterance got a temporal dimension (it
still is; it has not yet left the EU) (Austin 1962, 94).
Such illocutionary speech acts may, firstly, affirm the existing conventions,
but, secondly, they might also revise them and introduce conceptual innova-
tions. For example, the slogan ‘the personal is political’ was once a subver-
sive act, when it was militantly used by feminists beginning in the 1960s, and
it is still frequently disputed. In this sense both the history of the concept of
politics and the analytical distinctions used in the scholarly debates can play
a role, although for each case some specific criteria are needed, as will be
discussed in the section on political literacy below (Section 2.1).
This second perspective is based on the idea that the scholar who
focuses on the political aspects of a debate can detect concepts that the
debaters do not notice or mention at all. Carl Schmitt once formulated
the idea that an extremely intense way of doing politics is to accuse oth-
ers for politicking while denying that one is doing it oneself, affirming
that one’s own view is objective, scientific and so on (Schmitt 1932, 21).
Schmitt’s example illustrates that, although we must take into account
the language of political agents, it cannot be taken at face value. We have
to learn to read and interpret the political aspects of actions both when
the agents deny that they are acting politically as well as when they do
not even realise they are doing so, though their action is clearly politi-
cally significant. To give an example, we can analyse both the politics of
UNDERSTANDING DEBATE AS POLITICS  5

anarchists who reject the state, parties and frequently politics per se and
bureaucrats or experts who think themselves to be above the ‘dirty’ game
of politics.
It is important to take the language of political agents seriously as a mark
of their self-understanding of politics. Such understanding is informed not
only by what they say they do politically, but also by what they can in fact
do. Traditionally only state-based politics has been understood as ‘serious’
politics. In the twentieth century, party-based politics became the only
real form of politics for many politicians (see the example in Section 4.2).
They were unable to ‘do’ politics otherwise than via parties. They could
not recognise the politics of either informal ‘movements’ or intellectuals,
whose political action is characterised by its opposition to interest-based
politics and who intervene in questions and debates that ‘do not concern
themselves’, as Jean-Paul Sartre put it (Sartre 1966).
Noteworthy for political analysis are those situations in which the
political aspect is more obvious to outsiders or in hindsight judgement
than it is for the agents. This kind of politics, which is so self-evident as
to be unrecognised by those who are actually doing it, can be found for
example in the politics of street naming. Political decisions are built-in to
considerations of whether to use commemorative names of persons (rue
Bonaparte), historical events such as military battles (place de Stalingrad)
or concepts (avenue de la République), to use French examples. The same
holds for decisions on street names of a certain profile that combine a
person with what is being commemorated; for example, Rosa Luxemburg
and Karl Liebknecht retained ‘their streets’ in the eastern part of Berlin
after 1990. But also choices of names that appear to be ‘natural’ or from
the built environment are also thoroughly political, both in the sense that
they mean an exclusion of the commemorative alternatives and that street
naming per se is a matter of debate and decision-making. Interesting cases
are streets named after people’s first names, not only of monarchs and cler-
ics but also sometimes of high officials, as in the Helsinki of the nineteenth
century. From this perspective we can distinguish between cities of higher
and of lower political profiles (see Palonen 1993).

1.2   Politics as an Activity

Sometimes we hear MPs or ministers announcing that they are ‘leav-


ing’ politics. This kind of expression stems from a rather commonly held
assumption that politics is a separate sphere in which you enter and from
6  C. WIESNER ET AL.

which you can leave. As a field of study politics is often understood in


such a way, either deliberately or unwittingly. Politics is treated as a given
and well-known subject matter. More commonly than not, it is assumed
that ‘politics’ refers to some fixed and pre-defined space or area of exper-
tise, which is filled by professional politicians and is separate from every-
day life or the economy, culture, law or other spheres of life. Instead, we
argue that politics is an aspect of human action and that the politics-as-­
a-­separate-sphere type of thinking tends to hinder rather than promote
efforts to understand the multiple aspects of the phenomenon.
To think of politics in terms of spatial metaphors, such as a sphere,
field or sector, is historically relevant to the concept of politics. It still
seems to be the most common way of understanding politics within aca-
demic, everyday and media discourse. Sometimes it has become reified
into an acting entity, for example, in such expressions as ‘politics must take
responsibility’ over the refugee crisis or some other pressing phenomena.
This refers, in fact, already to politics as doing something, but makes it an
impersonal force above the action of individual human beings.
In the history of the concept, politics-as-activity is the second main
alternative for understanding politics (discussed from a formal perspective
in Palonen 2003, from a historical perspective in Palonen 2006). Unlike
the sphere concept, understanding politics as an activity does not demar-
cate politics from other phenomena, such as culture or economics, but
enquires into the political aspect of such phenomena. For example, many
artists today understand well that they act politically, and the same can be
said of persons speculating in the stock market, even if they more seldom
admit that they are thereby also engaged in politics.
For us, politics represents an aspect of human action that is present—at
least potentially—in every aspect of human life. Politics can be related to
schools, kindergartens and universities, to trees and gardens (and where to
place them or not), to group interests or to traffic regulations. The ques-
tion, then, is how and why such issues become questions of political action
and debate. A half-century ago it used to be a matter of ridicule when
someone said that the weather is a political matter, but in our age climate
change is hardly something anyone denies.
Studying politics as an activity contrasts our approach with that which
views ‘the political’ as an ‘ontological structure’ behind ‘mere politics’. A
major recent discourse, indebted directly or indirectly to Carl Schmitt’s
famous booklet Der Begriff des Politischen from 1932, claims that the
activity of politicians is only a surface phenomenon, behind which there
UNDERSTANDING DEBATE AS POLITICS  7

is a deeper layer or structure of ‘the political’, an ‘ontological’ level con-


sidered more important than the daily activity of politics. This has been
manifest especially in the Francophone philosophical discourse since the
early 1980s (see Marchart 2007), in which the influential work of Chantal
Mouffe (2005) can also be included. From our point of view, this stand-
point tends to devaluate the activity of politics, to ‘explain away’ its inher-
ent contingency.
We see politics as being about chances, choices and the inevitable plu-
rality of opinions that are part of human existence. In other words, as an
activity, politics is contingent and controversial, and we regard this as what
should make it intelligible and interesting to scholars. As James Martin has
eloquently expressed it:

Often the uncertainty or ambiguity of the world forces us to confront a


plurality of contrasting perceptions of our situation and opposed views of
how to act. […] If everything was certain and clear, if nothing were open
to chance, it would be a world without choices, a strangely unhuman world
devoid of the anxieties such choices generate. However attractive that sounds
to you, it would be, nonetheless, a world without politics. (Martin 2014, 1)

To characterise politics as a contingent and controversial activity means


that, in politics, acting otherwise is always possible. To put it another way,
there are always alternatives, chances to act differently, the outcome of
which is, more often than not, unexpected. The contingency of politics
involves potential and real changes of opinion and direction. However, this
should not be conflated with the view that anything can be done. The
range of realistic alternatives is restricted not only by others who inten-
tionally oppose a policy, but also by the complexities of the context and
of the historically formed situation. It should be noted, however, that
even though you can always act differently, this does not mean that you
can always change the result. In Humanisme et terreur (1947) Maurice
Merleau-Ponty discusses the situation of Nikolai Bukharin in the Moscow
show trial of 1938. Bukharin knows that he will be sentenced to death, but
still defends his past policies in the face of the accusation of conspiracy and
of having been an ‘objective traitor’, as the Stalinist jargon of the prosecu-
tor Vyshinsky would have it. The idea is that the way one chooses to face an
unavoidable fate still matters.
To understand politics as an activity also means to regard it as an activity
of operating not only in time, but also with time. This is related in part to a
8  C. WIESNER ET AL.

broader type of conceptual change, which Reinhart Koselleck (e.g. 2000)


has called the ‘temporalisation of concepts’, including a transformation of
originally spatial concepts, such as progress, to gain a primarily temporal
meaning, or to increase the gap between the space of experience and the
horizons of expectation. With this change, time is no longer experienced
merely as something consumed by action, but as an independent dimen-
sion of the action itself. Or, we could say that the present has become
something more than a line between the past and the future. In this sense
we can speak of a politics against time, of time and with time. The first
highlights the experience that the time available for political action is
always scarce, the second the struggle over the distribution of time and
the third the utilisation of time as a resource in the political struggle (see
Palonen 2006, esp. 17–19 and 294–295).
In parliament the scarcity of time is recognised by the old Westminster
principle that a member in a plenary debate can speak only once—which
leads to the question of when is it most opportune to speak (see e.g.
Hamilton 1927, 40). Since the nineteenth century additional devices
to restrict speaking time, such as the clôture (a motion to end debate)
and the guillotine (a time limit or scheduling of the entire debate), have
been accepted in most parliaments (see the discussion in Palonen 2014c,
212–220). The debates over how to fairly distribute the available time
between government and opposition and between frontbenchers and back-
benchers as well as between parties are another matter (ibid., 221–244).
Parliamentarians who master the procedure have a great advantage, and
part of this mastery consists of the competence to use different aspects of
parliamentary time to one’s own purposes. This can include moving for
‘order’, for example making the Speaker of the House of Commons decide
whether the debate is still dealing with the ‘matter’ on the agenda. This
motion compels the Speaker to interpret whether the current speech is or is
not in fact ‘in order’. Also motions for adjournment not only contain a post-
ponement of the debate to get more time, but also interrupt the ongoing
debate and open a new one on the justifications for the adjournment itself.
The scarcity of time in politics can be used to one’s advantage as well,
which relates to the struggle with time. Politicians become skilful at using
time to their benefit. This can be a fascinating part of political life, for it
is impossible to predict the twists and turns of situations and events, and
how unexpected circumstances may result in benefit or detriment to the
various sides. And that is, undoubtedly, one thing an MP learns and lives
with once she gets elected. Even though she may stand down in elections
UNDERSTANDING DEBATE AS POLITICS  9

and ‘leave’ the public stage, the mode of acting politically is not some-
thing she will part with quite as easily. The ability to identify and make
use of the chances possible when acting in the moment is a skill that can
develop over time. It can be argued that this skill to use time as a political
resource is an integral part of being a politician.
It is the position of a student of politics to analyse the use of those skills.
Moreover, it is not enough to study this phenomenon among established
politicians. The skill, or craftsmanship, can be obtained in a variety of
circumstances and simultaneously reveal new ways of conducting politics.
While identifying these may be difficult at first, there are ways to improve
one’s own skill at it (see Section 4.3 on the Oxford and Cambridge Unions
as political training grounds).
Politics is a phenomenon that can be read out of or into a myriad of
situations and events. As a contingent action, politics can at any time
become controversial between the participants. Any proposal or rule can
be contested. This is why debating between alternatives lies at the crux of
politics.
The disagreement and controversies integral to politics are best cap-
tured by the concept of debate. In his The Elements of Rhetoric, James De
Mille, a nineteenth-century Canadian professor of rhetoric, made a clear
distinction between ‘debate’ and ‘oratory’: ‘Oratory is the discussion of
a subject by one; debate is the discussion of a subject by more than one.
Oratory considers the subject from one point of view; debate considers
the subject from two or more opposed points of view’ (De Mille 1878,
471). He noted that debate involves two or more sides to a question,
which makes it a problematising, adversarial activity.
The adversarial character of debates, however, is not always self-evident.
Just because an opposing side exists does not necessarily make a debate
‘political’. Controversial conceptualisations and underpinnings may give
clues to finding out the kind of politics that is being exercised. A reading
of politics has to be conducted with the context and audience(s) of the
debate taken into account.

1.3   Four Aspects of Politics: Politicisation—


Polity; Politicking—Policy
In political science it is common to distinguish between three dimensions
of ‘political’. While polity refers here to the form of the political entity and
its formal framework, policy names the political output with the legislation
10  C. WIESNER ET AL.

and actions taken in different fields. In the triad, politics stands for the
political process not for the activity (see e.g. Rohe 1978).
Our view takes a broader look: we focus not only on the political
process, but also on the actions and activities that drive and constitute
politics. In relation to our approach of regarding politics as an activity con-
cept, we differentiate four dimensions or aspects of politics (cf. Palonen
2003). While these include the two commonly used expressions of ‘pol-
ity’ and ‘policy’, we interpret them somewhat differently. Our approach
also ­contains two new expressions: ‘politicisation’ and ‘politicking’. These
four dimensions are very helpful for distinguishing the different aspects of
political action. Their heuristic value for us is their use in demonstrating
how politics can be understood as an activity in a multifaceted way.
Polity in current usage refers to an established political system and its
institutional framework. However, we consider the construction of a ‘pol-
ity’ to be the result of (contingent) political action, not something that
is uncontested and essentialistically developed. Polities involve the setting
up of rules and regulations that guide their practices and conventions, yet
they are not unchangeable or permanent. Parliaments are obvious exam-
ples of ‘polity’. They are ‘metaphorical spaces’ (Palonen 2003, 179) that
are designed to contain politics within a certain range of action. To give
an example, parliamentary politics is conducted within a certain building
or among elected members of the parliament.
Policy in standard political science refers to the contents or outputs of
the political process which are mostly interpreted in terms of laws, regula-
tions and measures taken by the executive. We regard policy as a normative
and teleological dimension of politics; it is an acquired or intended line of
conduct or a project (Palonen 2003, 175). It points to future action by
giving direction. While representing a ‘programme’ or norms to be fol-
lowed, it guides political action. Policies can be altered or revised by con-
frontations and unexpected inconsistencies. They can also be abandoned
when a considerable number of the elements that constituted them in the
first place are no longer considered valid.
Politicking refers to the performance dimension of politics, that is, to the
act of doing politics, and hence to politics as an activity. It highlights that
politics cannot be reduced to results and outputs alone, but must also be
understood as practical performance (Palonen 2003, 177). Performance
of this kind can take many forms. For example, politicking can take place
in the European Parliament, at university board meetings or in student
groups. Styles of politicking vary according to duration and rhythm as do
UNDERSTANDING DEBATE AS POLITICS  11

styles of performance in the performing arts (ibid., 178). The moments


of politicking can also be very different in a variety of settings. In the case
of the European Parliament, politicking takes place on a public and global
stage, making its work dependent upon a schedule completely different
to that of university board meetings, which can, for the most part, decide
their own pace and calendar.
Politicisation, in our view, is to be understood as an act or a process
that turns a matter into a political issue (Palonen 2003, 182). Again, this
view is broader than current interpretations of politicisation that refer to
a growing interest of political parties in certain matters, such as European
integration, or to new political ruptures, or increased citizen participation
or interest. From our point of view, these may be consequences or signs of
politicisation, but they do not in themselves describe or define politicisa-
tion. The politicisation of an issue is mostly done by naming and through
interpretation. Politicisation occurs, for example, when governments try
to legitimise surveillance of the Internet in the name of ‘security’. Further,
this shows that politicisation opens up new horizons for politicking. In
the case of surveillance, securitisation gives the possibility to re-arrange
the political agenda and to accommodate for speeches on threats and fears
related to hacking and terrorism. Another form of politicisation relates
to the interpretation of ‘necessities’ and questioning an established pri-
oritisation of meanings. By identifying the political potential of ongoing
changes, shifts or processes, previous interpretations of what is ‘necessary’
or ‘impossible’ can be rendered obsolete (ibid.).
All four dimensions of politics can be found linked together in certain
cases. Politicisation can put into question the legitimacy or even lead to
the demise of a polity. To take an example, the dire national debt situation
in Greece together with the immigration crisis within the European Union
(EU) made some of the member states’ governments and citizens ques-
tion the integration process and the validity of common European values.
These developments have brought about what is often called the ‘crisis’
of European integration, and hence the EU polity is being questioned.
While the European polity has thus seemingly weakened, new horizons for
politicking have appeared. For example, the immigration crisis has caused
national governments to raise the need to protect their borders despite
the Schengen Agreement. At the same time there are signs of new forms
of integration during the crises, for example, the strengthening and accel-
eration of such policy areas as the planning for a pan-European border.
The unexpected state of emergency has made it possible to introduce acts
12  C. WIESNER ET AL.

and ideas that would not have been approved of or even been likely at all
under ‘normal’ circumstances. As member states have started acting more
or less independently, new impetus has at the same time been given for a
common EU policy.
In this volume the fourfold scheme of politics serves as a heuristic
instrument for the examination and interpretation of the different aspects
of acting politically and their interrelationships (the use of the scheme in
political analysis will be further illustrated with case examples in Chapter 4).
The analytic scheme is especially applicable to the interpretation of politi-
cal moves in debate, in relation to the language of the actors in their
self-understanding as well as interpretations that attempt to go beyond
them. It allows us to direct our ‘searchlight’ in certain areas and situations
to target certain types of politics, actions and linguistic expressions, and
thereby allows more specific interpretations to be made by analysing the
political constellations in the actual contexts as well as the more detailed
formulations and speech acts of the agents.

1.4   Types of Politics—Types of Debate

Our core thesis in this book is that debate should be understood as an


intensive form of political action. A debate only makes sense if there is
something to debate, if things are contingent and, therefore, able to be
altered with alternative directions for change. In a debate everything can
be questioned, and every debate can also be terminated in multiple ways.
Nevertheless, there may be differences of degree in the de facto ‘question-
ability’ or openness of the debate.
In twenty-first-century Europe, the everyday language of ‘debate’ refers
above all to two types of situations, which can be described as reflecting
differing ideal types. In the first place, debate can mean a confrontation
between representatives of two (or more) parties when presenting and
making public their views to a wider audience, without reconsidering their
own standpoints in the course of the debate. The pre-election debates of
party leaders are the most obvious example of this, while TV talk shows
could also be included. Secondly, debate can mean a speech competition
between two (or more) participants on any matter with the aim of being
a better debater than the adversary. Standpoints may be changed in so
far as a participant regards it as advantageous for winning the debate.
The debate competitions between school and university teams were first
established in the USA in the nineteenth century and have been adopted
UNDERSTANDING DEBATE AS POLITICS  13

elsewhere since as an important part of rhetorical training for a variety of


purposes.
Both ideal types are marked by certain core characteristics. In the first
case, ‘debate’ appears as a show between already known and unchanging
positions. In the second, debate is an occasion to display one’s debating
skills, for the purposes of which the content is irrelevant. In the former
situation, sides have already been taken beforehand, that is, inside the
parties rather than between them. Politics precedes the formal debate in
the sense that the choices and standpoints of the parties have already been
decided in advance. The debaters are merely expected to formulate their
standpoints in a manner that attracts votes, not to judge the strengths and
weaknesses of the proposals as such unless to make the adversaries change
their stands. In the speech competitions, the participants do not have any
commitment to their standpoints beyond the actual debate itself. Politics
is inherent in the practice of presenting ideas and arguments pro et contra,
but this is unconnected to the personal views of the partisans.
In practice, debate often takes place between these ideal types, although
it might also indicate aspects beyond them. A debate in parliament, for
instance, refers to a public exchange of standpoints where the emphasis is
on the notions of ‘public’ and ‘exchange’, and yet it is an integral part of
the process to take into account the standpoint of the other party. In par-
liamentary debates, it is perfectly possible that a standpoint may be altered
in the course of the debate, for example, when government supporters
consent to accept an amendment by an opposition member. Especially in
committee debates, the questions frequently concern the formulations,
though these are often politically significant and the majority might yield
to formulations that are more acceptable to the minority. Parliamentary
debate also includes a competitive aspect in the sense of who can come up
with the most persuasive argument, and thereby appear to be ‘winning’ in
the eyes of the audience. The person with the most persuasive arguments
will probably rise in status, even if her group or opinion loses the vote
afterwards.
Debate in the presence of adversaries forms the ideal type to which
other debates must be compared. As political action, a debate between
opposed points of view need not necessarily take place between agents
who address to each other in the same audience. This kind of debate can
also be conducted at a distance, between agents separated in space, time,
language and other contextual criteria. Therefore, we suggest a distinction
between actual (‘live’) and virtual (‘frozen’) debates. The latter generally
14  C. WIESNER ET AL.

includes academic debates, for example. Such debates are frequently car-
ried out under the direction of, or frozen in, documents, which seldom
refer to the target of their polemics. In a wider sense, not only speeches
but also writings and documents are thus contributions to this kind of
debate. This obviously holds for scholarly works, pamphlets, newspaper
articles, novels and so on that have been written in order to take a stand in
an ongoing debate or eventually to launch a new one.

1.4.1  Debate, Discourse, Dispute, Discussion and Dissensus


‘Debate’, as we understand it, is the practice of speaking pro et contra on
an item on the agenda between a plurality of agents. This understanding
differs from other related concepts, such as ‘discourse’. It implies a slightly,
but crucially different direction of analysis than ‘debate’, as it mainly
focuses on how meaning is constructed and reproduced. ‘Discourse’ thus
refers to a set of practices which constitute meaning (see also Section
3.2)—whether or not these are related to debate. The focus on ‘debate’,
as we see it, puts more emphasis on the adversarial aspects of the activity
of politics.
Debate includes putting questions on the agenda, taking a stand on
those already there and persuading the participants to alter their stands
or to rethink their justifications. In a parliamentary context, for exam-
ple, ‘debate’ does not refer to a single event but to the entire journey
of a motion on the agenda through the plenum and committee sittings.
Debate focuses on controversies over subject matter, the normative colour
or style of the argumentation, whereas discourse is concerned with the
tacit assumptions behind the controversies. Debate depends on the prob-
lematic, which is more interesting.
In his Rhetoric, Aristotle divided the art of speech into three genres:
deliberative speech for and against a policy in public assemblies; forensic
speech for accusing or defending a person in the popular courts (dikaste-
rion) for the person’s past deeds; and the epideictic genre for praising and
blaming in performative or ceremonial situations, which can also be found
in judging works of art (Rhetoric, 1358b–1360b). Later other genres
have also arisen, such as the rhetoric of negotiation in diplomacy between
nation-states or in labour bargaining between employers and employees.
Deliberation is the only genre decided among the participants them-
selves without relying on an external authority, such as a judge in a court-
room, an expert in the arts, or an arbitrator in negotiations. ‘Debate’ can
UNDERSTANDING DEBATE AS POLITICS  15

be seen as closely related to the rhetorical genre of deliberation as no exter-


nal authority determines the final outcome. In deliberative a­ ssemblies, the
ultimate authority is the chairperson, whose main responsibility is merely
to keep order and whose decisions may in certain cases be disputed by the
majority of the assembly.
As a rhetorical genre, deliberation refers to any debate in an assembly or
public meeting in which the participants themselves decide upon a topic
and perform the debate by speaking for or against a specific proposal.
There are, however, various terms other than ‘debate’ that have diffuse
conceptual borders and that can be used to describe deliberation, includ-
ing ‘dispute’, ‘discussion’ and ‘dissensus’.
‘Dispute’ refers to a specific controversy on a distinct topic between
persons defending opposite standpoints. It is less regulated and has no
time limitations. ‘Discussion’ is sometimes limited to a debate in which the
participants agree upon some basic ground rules and which overlaps with
polite conversation and generally aims for consensus (see De Mille 1878).
In contrast, ‘dissensus’ refers to the presentation of opposite perspectives
in a situation in which judgments may be altered as to the strengths and
weaknesses of the motion on the agenda. The procedure for expressing
and regulating dissensus is a defining criterion for parliamentary debate.
The differences between these concepts refer to different aspects of
politics. In a conversation-like ‘discussion’ the viewpoints are in proximity
to each other, whereas ‘dissensus’ provides for a new dimension of politi-
cisation in which the entire setting of deliberations may be transformed
(see Palonen 2014a). In ‘discussion’ the participants are more likely to
find closure. ‘Dissensus’, on the other hand, questions the allegedly shared
presuppositions of a debate. If ‘discussion’ is directed at closing a debate,
‘dissensus’ initiates new ones. Both refer to the non-regulated borders of
deliberation. ‘Debate’ and ‘dispute’, however, are likely to mean the very
opposite. Both can be found in very regulated forms of deliberation where
agendas are set and rules are agreed on beforehand.
The conceptual boundaries just explained have been drawn to make
sense of the various usages of these closely associated terms. However, it
should be noted that ‘debate’, ‘dispute’, ‘discussion’ and ‘dissensus’, as
well as ‘discourse’, are all used in different ways by different agents. For
example, sometimes ‘debate’ is meant when the term ‘discussion’ is used.
Walter Bagehot in his Physics and Politics (1872a) talked about ‘govern-
ment by discussion’. But the expression can be interpreted in terms of
‘debate’. Therefore, it is not possible to give strict definitions to any of
16  C. WIESNER ET AL.

these terms. They can only be used as markers that suggest which concep-
tualisation is behind them.

1.4.2  Debate as pro et contra and as a Peaceful


Form of Struggle
In certain circumstances debate can be regarded as a way of fighting fairly,
of waging a peaceful struggle. In our view, debate also means speaking
pro et contra on a proposal (motion) or on an analysis of the situation at
hand. In ‘live’ debates performed in deliberative assemblies, the opponent
is confronted in the moment and has to be respected as the opposition
by the same audience. In most assemblies there are established rules of
procedure that ensure all sides will be treated fairly. Compared to more
open settings, the debating is well regulated, and regarding verbal and
physical attacks, certain restrictions are observed. While parliamentarians
have the free mandate as well as parliamentary privilege, they still have to
be careful to avoid being seen as using ‘unparliamentary language’. One of
the official duties of the Speaker of the House is to keep order, and a part
of that is to sanction unwanted behaviour. In the parliamentary setting,
insults can be used to evoke emotions in adversaries in a way that makes
them more likely to show ‘unflattering aspects’ of their personalities (Ilie
2004, 52). It can also divert the attention from the actual debate, which is
one of the reasons why insults are not permitted.
A debate contains both adversarial and dissensual aspects. It can take
place, for example, between political theorists, between blog writers or at
party conferences. However, the parliamentary setting can be regarded
as the paradigm to which other types can be compared. The paradig-
matic format of debate ensures that all possible opinions can be uttered,
although within the bounds of the rules and procedures. Parliamentary
debate presupposes a respect for the rules of debate and for one’s adver-
saries. The same is less achievable in blogs or other public fora where a
certain anonymity is always part of the setting. In parliaments all partici-
pants are publicly known, and when they speak, they all have equal status
as representatives. The parliamentary procedure ensures that speakers as
well as agenda issues will be fairly treated (see Section 2.1).
The attention given in the procedure to the conditions and forms of
debate provides for a radical diversity of perspectives and opposing in
debate. Sometimes parliaments are referred to as ‘political theatre’ due to
the public performance aspects of debate. But it would not be justified to
UNDERSTANDING DEBATE AS POLITICS  17

assume that the performance is only for show. It should be remembered


that parliamentary procedure does not allow speakers to deviate from the
issue at hand, and therefore, debates are not merely entertainment or a
pastime. Whenever an item is put on the parliamentary agenda, it cannot
be removed without the consent of the House. It is also worth noting that
time limits are set for each session. Debaters cannot afford to waste their
allotted time slot.
Parliamentary procedure can be modified, and so it does change over
time. For example, the British Parliament has revised its rules of debate
several times, even though the changes have not always been major ones
(see Redlich 1905; Palonen 2014c; Vieira 2015). Still, those changes can
indirectly affect the dynamics of the debates. Another important factor to
consider is that the rules of debate can be bent to a certain extent.
No rule or law can cover all thinkable cases or predict future events.
Rules can also contain expressions that might seem ambiguous or impre-
cise in certain situations. In these circumstances, the letter of the law
must be related to such broader principles as justice and equity. Laws
require judges and courts for the interpretation of legal cases, similar
to the rules in, for example, football or other sport needing referees to
interpret ambiguous situations. The role of the chairperson in a parlia-
ment is comparable to a referee, and sometimes even to a judge (on the
use of juridical terms in parliamentary speeches and debates, see Bentham
1843). This does not mean that the rules are arbitrary, but that there is
some degree of latitude in interpreting them. There are, however, inter-
esting differences regarding the position of the Speaker or the chairper-
son. Since the mid-­nineteenth century the Speaker in Westminster has
been excluded from voting, even when a tiebreaker is needed. Other
parliaments, such as the Finnish Eduskunta, have followed this practice,
whereas for example in the European Parliament the president can even
participate in the debate simply by leaving the chair to one of the vice-
presidents (see Section 4.4).
Sometimes debates consist of contributions by various interlocutors
of different times and places. Intellectuals and theorists, for example,
take part in long-standing debates and controversies through their writ-
ings and speech acts all the time. Quentin Skinner’s The Foundations of
Modern Political Thought (1978), which deals with theorist-politicians
and politician-­ theorists of the Renaissance and Reformation periods,
offers a scholarly model for this kind of study. Intellectuals may contrib-
ute to debates that were launched by other scholars in other countries
18  C. WIESNER ET AL.

long before they themselves were even born. However, it should be noted
that the debates are not always identical to their earlier forms. Changing
­circumstances or approaches can alter or even invert the original ques-
tions. Political debates can also take place in other types of ‘texts’, such as
party programmes, strategy and policy documents.

1.4.3  ‘Live’ and ‘Virtual’ Debates


Debates, in a broader sense, can be either ‘live’ or ‘virtual’. In live debates,
the proponents and opponents of a motion face each other in the same
physical space, as is the case with parliaments and other assemblies that
follow their model. In public meetings, too, the adversaries may confront
each other, and the public decides who ‘wins’. In contrast, virtual debates
are extended to adversaries who do not share the same space or time. In
electoral campaigns, the candidates usually debate with each other not in
order to persuade their adversaries, but to attract followers. In these situ-
ations, the voters themselves do not have to debate, only vote. In party
conferences, debaters speak in front of an audience sharing the same basic
ideology or cause. The speeches are targeted against outside adversar-
ies, even though the participants may also be competing with each other
within the party framework.
Like discourse, virtual debates can extend over long periods, such as
months or years. Even daily newspapers polemicise against their competi-
tors’ editorials of the previous days or weeks expecting replies from them.
However, the ideal type of debate in a strict sense is limited to exchanges
that last a shorter time, usually hours or days.
Both live and virtual debates require interpretation, and neither form
should be considered more ‘real’, ‘authentic’ or ontologically primary
than the other. The very first delivery or act of verbal speech represents
for some the most ‘accurate’ form, and therefore, it seems necessary to
experience it first-hand in order for it to be analysed properly. But it is
debatable whether this immediate experience could provide enough ana-
lytical distance for the scholar (cf. Palonen 2013). Distancing oneself from
the subject matter is crucial, especially for political scientists who can easily
get trapped by their own attitudes or prejudices (Vaarakallio and Haapala,
eds. 2013b).
As distance is a precondition of political analysis, the difference between
live and virtual debates seems relative in the context of textual interpre-
tation. Another important point is that, even in live situations, it is very
UNDERSTANDING DEBATE AS POLITICS  19

difficult to distinguish the complete range of intended or pertinent audi-


ences. For example, during delivery a speaker may allude to other debates
by a g­ esture or even a facial expression that targets a certain audience.
However, these may still be lost to some of the audiences, and even to
some of the intended receivers for that matter. Textual formulations and
the style of performance can certainly provide clues for identifying other
debates alluded to by the speaker, implicitly suggested standpoints of the
speaker and contradictions as well as compromises between them. But
such information can often still be obtained even without having any
access to the actual debates.

1.4.4  Documents as Contributions to Debates


A key point in our argument is that political analysis should consider not
only spoken debate, but also written debate in the form of minutes, pam-
phlets, treatises or other documents. But in what way are documents to be
considered a part of debate?
We will first turn to impersonal documents that are not published by
named authors. We shall return to documents that are more personal in
the next section when discussing the relationship between academic and
parliamentary debate.
Depending on the research interest, documents of any kind can be
relevant for political analysis: parliamentary minutes, reports of commit-
tee hearings, press articles, interview protocols, laws, policy documents,
resolutions, official records, administrative statements, white papers,
party manifestos, constitutions or peace treaties. They can be clear parts
of debates or merely indications of debates. Alternatively, they may be
approximations to a live debate.
All legitimate and formal decisions are preceded by debates which
follow distinct procedures and are usually recorded in protocols, nowa-
days widely available online. Debating on both sides of the items on the
agenda, including the agenda-setting itself, is a methodical principle of
parliamentary politics and the basis for its rules of procedure for enabling
and regulating the debates. Parliaments use a distinct type of procedure to
decide on the majority of important law projects not to mention the fate
of governments.
Parliamentary rules and procedures are also applied in the debates of
meetings, associations, organisations or parties. The difference of these to
parliaments is that in the majority of cases, the adversaries remain o
­ utside
20  C. WIESNER ET AL.

the decision-making process. Michael Oakeshott (1975) calls such parties


or interest groups ‘enterprise associations’, which are united by a common
purpose, whereas parliaments and debating societies are ‘civil associations’,
united by common rules. This difference in purpose does not prevent
them from applying, to a greater or lesser degree, parliamentary rules in
their debates over internal controversies. For the analysis of debate, the
point of the distinction is that with civil associations the adversaries are
expected to sit in the same audience, whereas in enterprise associations
they are assumed to be outside while the enterprise members are ‘united’
against them.
Virtual or frozen debates can also extend over longer periods and take
place by or be reflected in policy papers, law texts or governmental pro-
ceedings (see Section 4.5). Debates in their non-public forms (e.g. par-
liamentary committees) are also related to controversies and exchanges
of opinion, so even in these cases the specific direction of policies and
programmes remains a matter of debate. In media articles, public and/
or mediated debates are reflected (see Section 4.6). Usually they relate to
specific topics in or outside the decision-making process, and they can also
wield considerable influence in the political system.
These various types of texts are important objects of research for under-
standing the way politics is carried out. Documents store a wealth of infor-
mation about what kind of political action has been practised in live debates.
They also can represent frozen debates, or they can be seen as the result
of past debates, or as marking turning points in ongoing ones, sometimes
even as the starting points or targets of future debates. Documents can
be considered as mere end products or ‘outputs’ of political discussion.
They may represent the ‘afterlife’ or continuation of debates, or indicate
the limits of a preceding debate, causing us to question things that were
not seen as controversial or were tacitly accepted by all the participants in
the debate at the time. It can be ­worthwhile to discuss why something was
absent in a document that, for outsiders or in posterior judgment, would
have been expected. The reasons may lie in explicit rules or regulations, in
‘hegemonic’ political constellations that in practice exclude certain stands,
or in implicit conventions of the situation that nobody in the context even
could have thought to question.
Using examples that represent these different types of debate, we will
illustrate both the depth and breadth of debates as political action as well
as the range of politics that can be studied from the debate perspective
(in Chapter 4). It will be shown that political activity can be read out of
UNDERSTANDING DEBATE AS POLITICS  21

documents of various kinds by concentrating on the rhetorical moves and


conceptual changes in the texts. In that way documents can be seen as
more than the mere end products of live debates. They can be considered
as pieces of broader debates that are not tied to just one space or time.

1.4.5  Political and Academic Debates: Similarities


and Differences
Quentin Skinner recommends his students to treat Thomas Hobbes’
Leviathan like a ‘speech in parliament’ (Skinner 2008). The quote reminds
us that the line between political and academic debates is not as clear as
is commonly assumed, and sometimes it can even be questioned whether
the line exists at all. When considering how some scholars do take part in
political debates and how some intellectuals are noted for their involve-
ment in political matters, it could be argued that one of the most impor-
tant tasks of a scholar is to raise matters in the hope of them reaching the
official political agenda. Some scholars are more active in this respect than
others, but the main point is that the ethos of a scholar makes her a poten-
tial politician in the field of public debate.
Skinner’s point on Leviathan indicates that academic debates on con-
cepts, theories, facts and approaches have much more in common with
political debates than what the textbooks assume. In scholarly debates
everything is controversial and alterable in principle. In this respect they
do not differ from (other) political debates. Contributions to such debates
can consist of speeches but also of scholarly works, pamphlets, newspaper
articles, novels and so on that have been written for the purpose of taking
a stand in an existing debate or launching a new one.
An argument for the debate-character of all scholarship can also be
found in Max Weber’s concept of ideal types, as presented in his essay
on ‘objectivity’ from 1904. With the work of Nietzsche and the rhetori-
cal tradition on the background, Weber proposes a perspectivist vision
of knowledge. ‘Reality’ is so rich, multifaceted and constantly chang-
ing that it cannot be exhausted by any conceptual apparatus. For the
Weberian ‘science of reality’ (Wirklichkeitswissenschaft), it is important
to present different and intentionally one-sided perspectives in order to
offer alternative interpretations of selected aspects of reality. One impor-
tant consequence is that, from a perspectivist view, there are no ‘facts’
independent of the perspectives of interpretation. A topic that appears
to be an ‘indubitable fact’ from one perspective may, from a different
22  C. WIESNER ET AL.

one, be seen to lose its factual character (Weber 1904, esp. 170–171,
180–181).
For the study of politics this kind of tendency should be rather evi-
dent. Skinner refers to the scheme of paradiastole used in ancient and
Renaissance rhetoric that emphasises the normative dimension in debates
of pros and cons, and to the possibility of constructing perspectives that
either devaluate the virtues or extenuate the vices of the concepts and
arguments (Skinner 1996, esp. Chapter 4; also Skinner 2007). Nietzsche
used this scheme in his famous principle of the transvaluation of values
(Umwertung der Werte) (cf. in particular his essay Zur Genealogie der
Moral, 1887).
Even if ‘arguing by numbers’ might seem to be an effective rhetori-
cal means in many contexts, we always have to question the perspective
behind them. It is a common experience that after elections almost every
party claims to have won, and they frequently do this with good grounds.
One party may speak of wins in percentage, another in seats, a third in the
absolute number of votes. Or they may choose some previous elections
(national, local, European, etc.) as the point of comparison. Even opinion
polls can be chosen as a reference point, and the victory (or loss) can then
be regarded as greater or smaller than expected, according to the party’s
own purposes. If we exclude some landslide victories or losses in which
the different numbers all point in the same direction, all such comparisons
may be justified, and there cannot be any definite overarching perspective
that would be absolutely superior. Even landslide victories may be seen by
adversaries and eventually by some partisans as Pyrrhic victories.
The Weberian point is that any fact can be disputed if one succeeds in
constructing a perspective from which doubt might be casted upon it. It
is always possible to find such perspectives, although they might not be
persuasive to the target audience. The ideal type, Weber’s special tool for
perspectivist interpretation, is a perspective that one-sidedly accentuates
the formal and pure possibilities of how to think about a question. For any
given topic we can think of a limited number of ‘pure’ possibilities that
offer alternative interpretations of the phenomenon or alternative courses
of action for the political actor. They do not correspond to ‘reality’ or offer
any hypotheses to be tested. In their pure and one-dimensional character
they can offer interpretative tools for analysing and evaluating cases in the
research material (Weber 1904, esp. 190–214). For example, the history
of voting systems—majoritarian vs. proportional, and so on—could be
analysed by devising and combining different ideal-typical alternatives in
UNDERSTANDING DEBATE AS POLITICS  23

relation to the shifting political constellations. However, scholarly debates


that apply ideal types to the complex and impure realities of political life
sometimes seem to be lacking. Focusing on pure possibilities is certainly
one of the reasons why academics frequently make mediocre politicians.
Academic debates consisting of opposing perspectives are analogous to
parliamentary debates on the pros and cons of motions and arguments.
But there are, of course, also important differences. Academic debates are
mainly ‘virtual’ in the sense of operating with written words, as opposed
to the oral character of parliamentary debates, and the distance in space
and time between adversaries creating advantages for them. In academic
debates there is neither a shared agenda nor common ‘questions and
fields’ of debate (Weber 1904, 184–185), but rather the polemicising is
more indirect, whether it concerns the academic agenda itself or the items
currently on the agenda. In this sense scholarly contributions to debates
have more latitude in terms of the substance and timing of the critical
comments.
Furthermore, in academic debates there is no strict end point similar
to the parliamentary ones of voting and taking sides by division of the
assembly. Indirect analogies of voting, however, can be seen, for example,
in citation indexes, which give an indication of the reputation of a scholar.
There is also an analogy to parliamentary majority when talking about
the so-called mainstream view, which is much harder to measure. At the
same time, the appeal to the mainstream seems to be a form of reputation
building as disputable as government proposals to parliamentary sessions.

1.5   The Presence of Politics in Texts and Debates


Political agents use language to define what they consider to be ‘political’
or a part of ‘politics’. No single definition is correct, and therefore, no
single definition is shared between dictionaries or between textbooks. Nor
can we speak of the existence of a few distinct conceptions of politics to
which all agents should commit themselves. On the contrary, politics has
to be read out of the texts and the debates. The formulations and styles of
performance have to be contextualised and analysed, and studies need to
be conducted on the actual usages of the vocabulary. This approach can
also easily illustrate how the concepts used have multiple aspects, dimen-
sions and historical layers that cannot be strictly separated from each other.
To understand politics as an activity concept is to focus on activity that
does not operate only in time, but also with time. A politician is not ­simply
24  C. WIESNER ET AL.

restrained by time; she can use it to her own advantage in a number of


ways. That includes using the rules of debate in a way that is the most
­beneficial for her cause. This kind of politics is found not only in live
debates, but also in written text and documents.
Debates very seldom explicitly introduce different views on what is
political and what is not—even in such situations in which, for example,
government representatives claim that ‘this is not a political question’,
whereas the opposition members insist on its political character. In other
words, the political character of some questions or phenomena is itself a
matter of dispute, and there does not exist any neutral judge to resolve
the question.
The conceptual controversy over definitions is itself part of political
action. Therefore, it is crucial to pay attention to, rather than discard,
other suggested definitions and meanings, even if all parties cannot iden-
tify the exact nuances of what is meant by a given expression. What the
scholar can do is clarify and make explicit the conceptual horizons of
the  debate. Scholars can identify what is typical in them, what aspects
of the concept have been thematised and how this has been done.
The second perspective relies on the idea that the scholar who focuses
on the political aspects of a debate can detect them, even if the debat-
ers themselves do not speak of politics at all. More important might be
those forms of politics in which the avoidance of political commitment
is expected from all—and still in such situations political aspects may be
found by readers who look at it from a distance and who possess the ana-
lytical tools for studying politics. Perhaps these kinds of tools can be best
applied to documents, for example, to the politics of street naming (cf.
Palonen 1993), or even to an implicit political subtext that regulates the
entire activity (see Section 4.7 on football).
CHAPTER 2

Reading Debates Politically

As was said in the previous chapter, we understand politics as a contingent


and controversial activity, and debate as one of its major forms. Our lead-
ing thesis in this chapter is that parliamentary debate, especially the type
of parliamentary debate formed in the British parliament at Westminster,
forms a historically paradigmatic form of debate.
The ideal type allows us to classify structure and ultimately analyse
debates by different criteria, such as topics, parliamentary speech acts or
participants. The types of parliamentary debate and the political constel-
lations sketched below thus provide a necessary background for debate
analysis. They present a number of issues relating to the range and forms
of debate from a variety of perspectives.
But everything cannot be analysed simultaneously: it is up to the
scholar to choose a specific debate or a problematic as the specific topic of
analysis. To make such choices, political literacy is a necessary prerequisite.
As discussed in Chapter 1, political activity is a multifaceted and complex
phenomenon that requires careful contextualisation of the actual debates.
The reading of debates has to take into account the agenda, its origins, ini-
tiators and purposes. What eventually becomes political during the debate
has a lot to do with the circumstances of the moment. Therefore, sensitiv-
ity to the participants’ rhetoric and use of language is needed for gaining
literacy in politics.

© The Author(s) 2017 25


C. Wiesner et al., Debates, Rhetoric and Political Action,
DOI 10.1057/978-1-137-57057-4_2
26  C. WIESNER ET AL.

Opening a debate always means to mark an issue as political, as a theme


of politics, in many cases one that had previously been considered as given,
‘natural’ or at least commonly accepted. The presence of Christian sym-
bols in schools, for example, has been taken for granted in most European
countries for centuries, but more recently it has become politicised as a
sign of partiality against non-Christians in a secular state. But the debate
continues, because some non-Christians have interpreted these symbols as
a part of their cultural heritage, worth knowing for all citizens indepen-
dently of religious belief (or the absence thereof). Whether decisions on
matters like this can be made by the schools themselves and by whom, or by
local, regional, national or EU-wide bodies at an administrative, govern-
mental or parliamentary level is a controversial matter itself and indicates
the thoroughly political character of the debate and the decision-making.
In historical studies the problem also sometimes arises that we do not
know much of the context and it is difficult to identify to which debate a
document contributes. Therefore, it is frequently difficult to understand
what in a text is just a repetition of the conventions of the time and what
might constitute an innovation or break with convention. It is necessary
to use independent contextual evidence together with the formulations
in the text to understand to what context of debates a text belongs and
what kind of speech act it was in its context (see Skinner 1970). A major
contribution of Quentin Skinner is to have situated classical works, includ-
ing pamphlets (Skinner 1978), rhetorical textbooks (1996) and the foren-
sic genre of rhetoric (2014), back in their original setting, among their
contemporaries.

2.1   Political Literacy


The concept of political literacy does not refer to the ‘locutionary’ mean-
ing of words and sentences, to fluent command and a knowledge of their
familiar meanings, but to the ‘illocutionary’ meaning, that is, understand-
ing the speech act involved in the sentence (in the sense of Austin 1962).
It is crucial for political literacy to be willing and competent to judge
actions, situations, practices and institutions in terms of political struggle.
What kinds of aims can we identify in various utterances, arguments or
topoi, and how may we assess their consequences?
For politicians this means acquiring a certain practical knowledge of
the history of the issues debated and of the concepts involved as a precon-
dition for understanding ongoing debates. For example, the concept of
READING DEBATES POLITICALLY  27

‘question’ in the parliamentary sense has several historical layers of mean-


ing (see Palonen 2012b). Parliamentary newcomers must learn the rules
of parliamentary debate—not just the written ones, but also how they are
applied—before they are adequately equipped to take part in plenary and
committee debates. Such political literacy increases a member’s chances to
politicise an issue into a subject of debate (see Flynn 2012).
Scholars of politics have an equal need for political literacy in order to
analyse the point and significance of the moves and judgements of debat-
ers. We have to avoid, however, hindsight in applying knowledge that
could not have existed for the politicians at the time of their actions. A
practical knowledge of the history of concepts is necessary in order to be
able to understand the ‘rules of the game’ and to participate in debates on
an equal footing with other competent players.
It is also important to be cautious to avoid judging the moves of politi-
cians according to the criteria of our favourite theories, understanding that
all theories are just theories and the political world is always much more
complex and nuanced than the best, necessarily one-sided theories can be
(see Weber 1904). This might also be one of the reasons why renowned
scholars tend to make miserable politicians: they analyse situations too
dogmatically, in the light of their favourite theories. Perhaps particularly
susceptible to this danger are economists, who are frequently used as
authorities in the media, but who tend to underestimate the translation
problems involved in moving from the ideal-typical level of their scholar-
ship to the ‘down and dirty’ complex world of politics.
A bit of practical advice for political literacy is to imagine oneself in the
politician’s shoes. Weber’s ‘occasional politician’ (1919) refers to the fact
that citizens of Western polities that include parliamentary and democratic
features tend to already have at their disposal an everyday knowledge of
political practices, as voters, debaters, participants in public meetings,
members of organisations, readers of newspapers and so on. To claim that
the world of politicians is entirely alien to the ‘ordinary person’ is another
myth that has the effect of maintaining a self-imposed political illiteracy.
The ancient Athenians used theatre as a medium for teaching political
literacy, and even started to pay a fee (misthos) to spectators. As criti-
cal spectators, the citizens of today also have some degree of political
literacy, though they may have lost the need to participate themselves in
debates or to take public stands. Weber wanted to stress to his student
audience (in his speech Politik als Beruf, originally presented in the revo-
lutionary Munich of January 1919) that even contempt for p ­ oliticians is
28  C. WIESNER ET AL.

a form of being an occasional politician (see Palonen 2002). One of the


main aims of the speech and booklet (published in summer 1919, after
the ­suppression of the Bavarian revolution) lies in persuading the target
audience to imagine themselves in the situation of an ­acting ­professional
politician.
Academics of today tend to be as contemptuous of politicians and as
lacking the willingness to imagine themselves in their situation as they
were in Weber’s time. This voluntary political illiteracy consists perhaps
first of all in reducing the value of politics to its results or to the hours
dedicated to it. Such illiteracy is of detriment to the scholars themselves in
terms of their intra-university politics. To recognise the existence of a dis-
tinct political literacy also allows one to overcome the prejudice that schol-
ars are wise and politicians dull and simple-minded. In Hannah Arendt’s
terms, presented in her The Human Condition (1958), politics is neither
fabrication nor labour, but action.
Part of political literacy is also the ability to recognise how the actors
themselves present in the situation understand politics, analyse the situ-
ation, assess the political constellation and so on. Here it is important
not only to avoid anachronism, that is, of being wise after the event, but
also to judge the political competencies of the actors in situ, taking into
account situations that have changed such that the professional politicians’
old practices are no longer viable, for in such cases outsiders may have a
more lucid view as occasional politicians. This might be the case in revo-
lutionary or crisis situations.
We can think, for example, of the role of intellectual politicians in the
East European regime changes of 1989 (see Konrad 1985). Vaclav Havel
in particular presented himself as an anti-politician, one who assumed his
responsibilities at the collapse of the communists’ ‘politics’. As a playwright
and internationally known dissident who had been imprisoned under the
communist regime, he also had competencies that were appropriate for
politicians in dramatic situations, but when such a situation was over, his
competencies lost much of their political value (see Zantovsky 2014).
Every politician does not have an explicit conception of politics at all,
although some references to their modes of conceptualising politics are
worth looking for in their speeches and writings. Besides levels of com-
petency, we could also speak of personal styles of political literacy, among
both professional and occasional politicians. Moreover, doing politics also
differs according to the situation. For example, election campaigns, par-
liamentary speeches, ministerial statements and diplomatic ­negotiations
READING DEBATES POLITICALLY  29

as well as politicians themselves can be judged according to the breadth


of their political register. The ministers of EU member-states cannot sim-
ply defend their ‘national interest’, but when ministers act as members
of the Council of the European Union and the prime ministers act as
members of the European Council (EC), they need both parliamentary
and diplomatic competencies (see Palonen and Wiesner 2016).
The concept of political literacy helps scholars to understand the rise,
the course and the results of controversies. This should not be interpreted,
however, as simply aiming to legitimise victory. On the contrary, political
literacy is a concept that helps us understand why politics, as a contingent
and controversial activity that depends also on the activity of others, might
be of high quality but ultimately not successful. Politicians rather frequently
choose to defend unpopular causes which they consider to be worth support-
ing. Sometimes they even support actions which they know to be doomed
to fail but regard despite that as worth defending. In Weberian terms, poli-
tics requires a combination of the ethos of conviction and ethos responsibility,
which frequently oppose each other (see Weber 1919, 79–88).

2.2   Parliamentary Debate as an Ideal Type


We cannot understand a debate politically simply by analysing the debates
that appear in the media, in speech competitions, or in parliaments or
assemblies. Our point is explicitly to avoid treating debates as if they were
everyday conversations and thus to trivialise politics as though it were
equivalent to small talk. Raising a topic to a subject of debate is an explicit
sign of its controversial character.
Political aspects can be found in every pro et contra debate, but their
importance and intensity can vary greatly. In parliamentary-type assem-
blies, including, for example, municipal councils or party conferences,
politics is the main reason why debates are held at all. Parliamentary-type
debates do not only concern the debaters themselves, but are also related
to the forms of the polity and the direction of the policy in question,
possibly with vital consequences for a great number of citizens and non-­
citizens. Such debates mark the horizon of the politicisation of issues in
the context in question and also give indications to the practices of poli-
ticking (for explanations of the various ‘polit’-terms, see Chapter 1, and in
detail Palonen 2003).
The approach in this book is to regard parliamentary debate as the
ideal type of political debate, a benchmark against which other debates can
30  C. WIESNER ET AL.

be compared. There are a number of reasons for this choice. Parliaments


are assemblies about which both politicians and citizens (in democratic
polities) hold the expectation that agenda items will be debated fairly with
regard to the pros and cons of the topics themselves. Indeed, in parliamen-
tary procedure opposed points of views are a presupposition: in no other
way is a thorough understanding of the issue possible (see De Mille 1878),
and members are therefore expected to invent alternative perspectives on
the question in order to be able to properly judge the strengths and weak-
nesses of the given motion.
Parliamentarians understand well that what looks like an indisputable
‘fact’ from the point of view presented by the government may be dis-
puted by members offering another perspective, making the ‘facts’ sud-
denly look very different. The sophist Protagoras once saw the task of a
debater to make ‘the weaker logos appear stronger’ (see the fragments in
Gagarin and Woodruff 1995, esp. 187–188). Nietzsche and Weber insist
that every ‘fact’ is dependent on the interpretations of it, and that new
perspectives can always make the world appear in a different light (see
in particular Nietzsche 1887, 860–861; Weber 1904, 180–181). In this
sense we do not mean by ‘parliamentary’ merely a type of government,
but also a mode of thinking in terms of debating alternatives and judging
their strengths and weaknesses.
Parliamentary-type actors are required to present and review their
opinions on any issue that might become politically relevant. In a debate
it becomes possible to see an issue from different perspectives, while the
perspectives themselves are also subjected to a fair debate. Although par-
liamentary debates concern taking a stand on a motion on the agenda,
the justifications for and against may differ and are themselves parts of
the debate.
Concerning the question of what is under debate, we insist on the
importance of agenda-setting. In parliamentary debate, putting a question
onto the agenda is a necessary condition for the realisation of the debate:
the members have an occasion to speak only if a motion is on the agenda.
Regulations concerning how, when and by whom a motion can be put on
the agenda characterise parliamentary-type debates. The analysis of debate
should always consider the rules of the assembly’s agenda-setting in order
to understand the moves that take place in the debate.
Parliamentary debates end with a vote as their last step, although
question hours and more recently occasions to debate day-to-day mat-
ters without a vote have been introduced. The vote expresses the view
READING DEBATES POLITICALLY  31

of the majority of the assembly on the item under debate. In matters of


­legislation, a vote is binding also on the minority, but minorities can main-
tain their views after the vote and try to alter the majority in the future.
Both the items on the agenda and the persons speaking are, in the
parliamentary order, clearly separated from each other in time. No pack-
age solutions can be accepted en bloc, but rather legislation is debated
paragraph by paragraph and budgets accepted subsection by subsection,
after the introductory presentations and general discussions. Unlike in a
referendum, a parliament has the power to debate the details and alter, in
committee or in the plenum, some points of the proposal. Few laws and
hardly any budgets in established parliamentary regimes go through with-
out any alteration. The budget deliberations tend to last several months
every year, offering members time to examine a government’s policies,
although majority governments might allow only marginal changes.

2.2.1  On the History of Parliamentary Debates


It is the debate character of the speeches that still gives Westminster its
paradigmatic position compared to parliaments where pre-written party
declarations dominate (for this perspective, see the numbers and discussion
in Proksch and Slapin 2014). Parliament has transformed the deliberative
rhetoric of ancient and medieval assemblies into a rhetoric of debating pro
et contra (for pre-revolutionary England, see Peltonen 2013). This his-
torically relates to the empowerment of parliaments over monarchs, courts
and bureaucracies, which is a change that also created the conditions for
the much later step-by-step democratisation of parliaments.
The expressions ‘parliamentary oratory’ or ‘parliamentary eloquence’
have been used since the mid-eighteenth century. At Westminster, the
second half of the century is frequently seen as ‘the golden age of parlia-
mentary eloquence’. In the nineteenth century, however, we can speak of
a transition from oratory to debate in parliamentary deliberations. The
ancient criteria became seen as obsolete and parliamentary speeches were
now judged as contributions to debate (see e.g. the young Gladstone’s
view of 1838). The ‘school rhetoricians’ of the time lamented this as a
decline in parliamentary speaking (see Palonen 2016).
The parliamentary reforms of 1832 and 1867 contributed to the
change in parliamentary speaking. The agenda was constantly growing,
and while some items were delegated to committees, the pressures on
parliamentary time continued to increase. Hour-long speeches became
32  C. WIESNER ET AL.

outmoded. In the unreformed parliament, speaking was a privilege of a


few leading parliamentary speakers, whereas after the reforms the vot-
ers in the constituencies and the local press began to expect that ‘their’
Member of Parliament (MP) would speak regularly in the plenary sessions.
To say something became more important than having something to say,
as the contemporary critique went (e.g. William Borlase in the House of
Commons, 20 March 1882, quoted in Palonen 2014a, 342).
What did all this ‘democratisation’ of parliamentary speaking mean for
the debates? Several commentators lamented the declining quality of the
speeches (e.g. Johnston 1927); others insisted that speaking became more
necessary than ever as a power share for the parliamentarians, especially
for political leaders (see Craig 1913; Curzon 1913; Weber 1918; 1919).
Instead of long speeches following the classical rhetoric’s canon of stages
of speech, debate speeches commenting on the previous speakers, includ-
ing replies and interjections from the floor, became ever more important.
What is a matter of legislation is also controversial, and it has changed
over the decades and centuries. For instance, many things previously for-
bidden by law are no longer so, especially in the lifestyle questions. In nine-
teenth-century Britain, restrictive oaths that prevented elected members of
parliament from taking their seats were gradually removed (these applied
to Catholics, Jews and, lastly, atheists, see May 1883; Redlich 1905).
The amendment is the distinctive instrument of the Westminster prac-
tice of debate, and it was known already in the seventeenth-century tracts
on parliamentary rules. Amendment does not mean merely ‘addition’,
but ‘Alteration, Addition or Omission’, as Henry Scobell wrote in 1656
(pp. 22–23). At Westminster, amendments to more or less radically alter
the political content of motions on the agenda are the main way of pre-
senting political alternatives, whereas in continental parliaments frequently
a full counter-motion must be presented.
Putting forth alternatives through amendments requires the special skill
to present the contrast in simple, concise terms. A classic example is John
Stuart Mill’s amendment motion to Benjamin Disraeli’s parliamentary
reform bill in the House of Commons on 20 May 1867: ‘Amendment
proposed, in page 2, line 16, to leave out the word “man”, in order
to insert the word “person,”—(Mr. Mill,)—instead thereof’ (http://
hansard.millbanksystems.com/commons/1867/may/20/clauses-3-4-­
progress-may-17). This move was directed at introducing female suffrage.
Amendments have three temporal aspects. We can illustrate them
with Mill’s motion. First, they suspend an ongoing debate. Secondly,
READING DEBATES POLITICALLY  33

they oblige the members to consider the strengths and weaknesses of the
­original motion (Is it really necessary to exclude women from voting in
parliamentary elections?). Thirdly, they open up a new debate (What could
be the political significance of introducing female suffrage?).
The principle that only one member may speak at a time has, how-
ever, important and legitimate exceptions. Cries of ‘Order! Order!’ in
Westminster are a regular part of the debate: by such cries a member can
direct the Speaker’s attention to a speaking member who, in the interjec-
tor’s view, is not sticking to the ‘matter’, that is, to the item currently on
the agenda, or to a member using ‘unparliamentary’ language towards the
parliament or its members. Members can raise other questions of order as
well, for example, by moving for a termination of the debate (clôture). By
raising a second-order debate on the parliamentary procedure, the mem-
bers of parliament are able to advise the Speaker and thereby serve as
guardians of parliamentary order themselves.
For studies on the application of parliamentary procedure, the ‘Order!
Order!’ shouts provide occasions to analyse how strictly the rule on speak-
ing to the matter is upheld and to what degree the Speaker takes notice of
such calls. Attempts to raise a question from the floor by individual mem-
bers may provide important occasions for creating precedents or changing
old ones.
Spontaneous interjections from the floor are not mentioned in the
rules of procedure, but they have been included in parliamentary records
(for a historical study on the German Zwischenrufe, see Burkhardt 2004).
Thomas Erskine May, the author of the most famous Westminster tract
on procedure, writes: ‘There are words of interruption which, if used in
moderation, are not unparliamentary’ (May 1844, 298). It is up to the
Speaker to decide whether such calls are unparliamentary or whether they
risk accumulating to a degree that would cast the ongoing debate into
a state of complete disarray. The study of such situations is important
for understanding parliamentary procedures as well as rhetorical practices.
For some famous parliamentarians, such as the German Social Democrat
Herbert Wehner, receiving an Ordnungsruf, that is, getting ‘named’ by
the Speaker, was used as a badge of honour in their political profile (see
Floehr and Schmidt 1984).
Studies of the art and degree of deviating from the principle that
only one person may speak at a time could be realised in historical stud-
ies of a parliament or assembly or in comparative studies of parliaments.
Characteristic of the powerlessness of this parliamentary order is the
34  C. WIESNER ET AL.

Reichstag of the last years of Weimar Germany, in which both Nazi and
Communist members showed open contempt for parliament and its mem-
bers, but the President was unable to keep them in order or force mem-
bers to retract unparliamentary expressions (see Mergel 2002).
A distinctive feature of Westminsterian parliaments lies also in their
combination of debate with the resolution to be voted. In the parlia-
mentary sense, the vote can be regarded as the last ‘speech act’ in the
debate. Gilbert Campion, a Clerk of the Parliament, formulated the point:
‘Motion, Question and Decision are all parts of a process that may be
called the elementary form of debate’ (Campion 1929, 143).
The matter was different in pre-revolutionary French assemblies, as
Jeremy Bentham analysed in his Essay on Political Tactics. One of the main
differences was that the debate and vote were not strictly separated in
time, but opinion and avis contained something of both. The members in
a sense ‘voted’ in the course of the debate before they had heard all the
arguments from the other participants. Bentham illustrates the difference
by legal analogy: ‘To vote for or against a motion, is to judge—to exercise
the office of a judge: to speak for or against it, is to exercise the function of
an advocate’ (Bentham 1843, Section VI.5.4).

2.2.2  Extensions of Parliamentary Debates


There are several possible extensions of the parliamentary paradigm. For
instance, the parliamentary paradigm is clearly visible in the debate order
of the assemblies. The US scholar of rhetoric John Marks Brewer makes
the point: ‘Only one topic will be under consideration at any one time,
and only one person at a time will be speaking’ (Brewer 1916, 291). This
parliamentary model has shaped the procedures and practices of all kinds
of meetings, associations and organisations (see e.g. Haapala, forthcoming
2017). For example, the famous procedural guide, Robert’s Rules of Order
(1876), was sketched by US Brigadier General Henry M. Robert for his
Baptist church congregation.
In debate competitions the stands taken do not have consequences
beyond the debate. The sides are given and a pair or team of debaters com-
petes. Victory in debate is the only aim and referees discuss and vote upon
who, in their opinion, won the debate. In parliamentary and academic
contexts, the question of who ‘won’ is always a matter of interpretation
and dispute—it being not always a matter of who won the vote. By partici-
pating in competitions, debaters can learn some features of p ­ arliamentary
READING DEBATES POLITICALLY  35

tactics and strategy, but not on all aspects of parliamentary politics, for
example, the politics of agenda-setting.
As we discussed after quoting Quentin Skinner in Chapter 1, academic
debates on concepts, theories, approaches and facts may also have much
in common with parliamentary debates. There are, of course, important
differences as well. In academic matters no strict votes between theories
or approaches are conducted, but indirect analogies of voting such as
the number of citations are a part of the current practice. The academic
analogy to the parliamentary majority is the appeal to the so-called main-
stream. But in scholarly debates the changes in mainstream views rather
resemble changes in fashion: what was ‘in’ yesterday will be forgotten
tomorrow. Like minorities in parliaments, unfashionable standpoints can
be tolerated and even respected among colleagues in a profession. A com-
mon element in both scholarly and parliamentary debates are thought
experiments (What about thinking in this or that manner?), as well as
inventing devil’s advocate objections or constructing novel perspectives in
order to make commonly assumed theories and facts disputable.
Besides these examples, we can also cite parliamentary-type debates on
lifestyle, which have been largely removed from political assemblies to the
private life of citizens. At the same time as gender and religious lifestyle
issues become a matter of debate among citizens, we can also detect a
parliamentary dimension to the debates on other everyday choices around
practices involving food, clothing and travel. Considering how the range
of parliamentary-style debates (as well as the open or implicit debates pre-
ceding them) have been extended to include the everyday decisions of
the individual, it is evident that today every citizen indeed has become an
occasional politician (Weber 1919, 41).
Last but not least in value is that seemingly non-controversial or neutral-­
looking statements in written texts may in fact contain important political
commitments (which is a point motivating our reading texts and docu-
ments as contributions to debate). Such texts can be better understood
as political speech acts if they are compared with previous documents of
the same genre or with documents written by rivals or adversaries. Studied
from this perspective, such documents or even single texts may suddenly
become ‘alive’.
Documents that are subject to regular revision, such as party manifestos
or programmatic statements on a policy, clearly require such a ­comparative
reading in order to understand their political point. In such cases the omis-
sions are frequently as important as terminological innovation or explicit
36  C. WIESNER ET AL.

revised stands on issues. The case might be more complicated with aca-
demic or literary works by individual authors, as they frequently tend to
use allusions without naming the targets of their polemics. These must
then be identified by the scholar and discussed from the point of view of
whether, for example, the critic understood the point or gave fair treat-
ment to their adversary’s position.

2.2.3  Limits of Debate
Parliamentary debate can be compared to settings that minimise debate or
attempt to exclude it entirely. Before entering into the analysis of debate,
we must thus discuss limit-situations on the borderline between debate
and non-debate. Such examples include: debates without an agenda;
acclamation (epideictic rhetoric); negotiations (diplomatic, labour market
or coalition-government negotiations); debates without a vote; and votes
without a debate.
In a parliamentary-style debate the practice of speaking separately and
in a consecutive order is linked to the question on the agenda, which
debaters must stick to in their speeches. We can think of limit-situations
of debate that have a minimal agenda. In the epideictic genre of classical
rhetoric, speeches were given at different kinds of ceremonies, and the
orations succeeded each other, but there was hardly debate with the pre-
ceding speeches. Still, even with a minimal agenda in an epideictic setting,
the actual occasion of the ceremony may introduce a moment of ‘debate’
in the form of the speaker’s ‘agenda’ to elicit the audience’s response
about whether to applaud or not. Even in speech competitions, the ideal
of speaking better than others marks a minimal agenda.
Teach-ins were a minimal form of debate that arose in the US student
movement in the 1960s. The teach-ins could last for hours, and everyone
was expected to speak from the heart on whatever matter one wanted. The
idea was to increase both participation and freedom of speech by remov-
ing the agenda: a succession of short individual orations without thematic
limits could produce a marathon of speeches. Those who had something
to say were expected to contribute, and the debate would therefore end
when no new ideas to offer the audience remained. Whether the inten-
tion was to proliferate matters independently of each other or whether
the contributions constituted a form of indirect debate is difficult to say.
Still, a teach-in that is organised around a question, that sets a minimal
agenda and that encourages participants to take a stand for or against the
READING DEBATES POLITICALLY  37

given issue, instead of being just a disjointed series of orations, could be


considered a debate.
In elections the agenda consists of electing candidates to parliament,
but there is no proper agenda for the campaign speeches. They may con-
cern the qualities of the candidates, the party ideologies, the interests of
certain regions or professions, a choice between the incumbents and the
non-elected or topics expected to be introduced in the next parliament’s
agenda. The procedure of the elections itself forms a minimal agenda that
regulates the situation, and the prospective agenda of the next parliament
enjoys a conceptual priority over other matters of debate, although there
is no need to debate with other candidates.

2.2.4  Restrictions of Debate Inside Parliaments


There are also tendencies within parliaments to replace debate by simply
reading pre-written speeches. In classical terms these may correspond to
the epideictic rhetoric of acclamation (typical of ceremonial occasions as
well as of referenda and presidential elections) or to the rhetoric of nego-
tiations (typical of diplomatic practice between given parties).
In the analysis of such limited debates we must consider the minimal
agenda as well as find out whether, when, where and how a debate will
arise. This can be the case also in parliaments with a high degree of party
discipline and the time for speaking allocated according to party quotas,
for in these cases only pre-written declarations without any reference to
other speeches would be heard. Does an assembly no longer constitute
a parliament once such declaratory speeches in the epideictic style have
become the rule?
If the government regards the slightest alterations to its motions as ques-
tions of confidence with party whips trying to impose ‘discipline’ upon the
backbenchers, then the ‘debate’ will have an epideictic character, typical
of referenda or of presidential regimes. In so far as parliamentary politics
in a country approaches a one-dimensional government versus opposition
divide, we can speak of presidential tendencies existing also within the par-
liamentary system, for in these cases parliament plays only a ratifying role.
The Westminster parliament, however, has found counter-measures to this
tendency by institutionalising in its rules of debate a second dividing line
between frontbenchers and backbenchers. By creating special occasions for
individual members to initiate debates (see e.g. Campion 1958; Griffith
and Ryle 2003), the House of Commons has avoided becoming a merely
38  C. WIESNER ET AL.

acclamatory assembly. Recent procedural reforms during the Brown and


Cameron-Clegg governments have reactivated the powers of the House
and the backbenchers to speak from perspectives beyond the government-­
opposition divide (see Crewe 2015, Chapter 5; Wright 2012).
Another limit-situation for debate lies in the cacophony of simultane-
ous speeches by members, due to the decline of keeping ‘parliamentary
order’. Speakers compete for who will be heard in such a ‘state of nature’,
as in a contest for whose voice can override the others. The ‘speeches’ will
then be cries, added on top of the noise of acclamation or disapproba-
tion of varying duration and volume. In such situations the parliamentary
principle of voice as an interruption of business-as-usual does become a
noise without specific intention or object. While in extreme cases, causing
a tumult in parliament can be an effective protest against a government, it
can also furnish the government with an excuse to dissolve or even abolish
the parliament.
In French revolutionary assemblies following the Rousseauvian tradi-
tion, debating and voting were separated to different assemblies. This
practice was institutionalised by the Napoleonic constitution with debat-
ing performed in the Tribunat and voting performed silently in the Corps
législatif (see Garrigues 2007, 114–122). The former expressed the ‘opin-
ion’ and the latter the ‘will’ as the decisive moment (see Urbinati 2006,
2014). The suspicion against debate is based on a certain form of egali-
tarianism, understanding parliamentary politics as rule by the eloquent or
by the ‘bavards’, a suspicion subscribed to by later anti-parliamentarians
such as Thomas Carlyle (1850) and criticised by John Stuart Mill (1861).
Voting without debate is mere acclamation, and has been explicitly sup-
ported by such anti-parliamentarians as Carl Schmitt (1928).
The Rousseauvian brand of egalitarianism assumes that citizens always
have political opinions, and that these should be taken as they are, however
prejudiced they might be. Frank Ankersmit has provided a well-founded
critique of this view with his thesis that the act of representation concep-
tually precedes the formation of the represented and the representatives
(Ankersmit 2002, esp. 115). Citizens have definite political opinions only
when they are asked to take a stand on a question on the agenda, institu-
tionalised in elections or in parliamentary-type assemblies. In such situa-
tions the vote is taken after a debate, although the electoral campaign is
only a partial surrogate for the parliamentary form of debate.
A debate without a vote tends to reduce the powers of the debat-
ing assembly. However, as mentioned, the Westminster tradition has
READING DEBATES POLITICALLY  39

developed special situations of debating without voting in order to avoid


the opposite danger, that is, of becoming a mere ratifier of government
proposals, in which case voting is more an acclamation than the last
step in the debate. All this allows us to analyse the politics in different
types of assemblies and meetings, with their different profiles relating
to debating and voting, by considering the special characteristics of the
assembly and its rules of procedure for debating and voting.
A different limit-situation of debate is negotiation, in which there is
neither a debate nor a vote. This can take place in the form of replac-
ing the parliamentary style of deliberating pro et contra with the diplo-
matic style of negotiations. For the danger of reducing parliaments to
this level, see Edmund Burke’s famous formula from his 1774 Bristol
speech on the threat posed to parliaments if representatives become
more like delegates and the parliament a constituency-based ‘congress
of ambassadors’. Another possibility was supported by early socialists
who wanted to see parliamentarians as delegates of their own party
(Kautsky 1911).
In terms of classical rhetoric, we can consider negotiations towards a
common aim between a given set of participants as a borderline case of the
deliberative genre. The parties to the negotiations strive for an agreement
as advantageous as possible for them. The paradigm for such negotiation
is diplomacy, in which the ambassadors of different states negotiate in
secret to reach an agreement, which may involve adjustments to their own
separate aims and interests in order to reach a compromise. But when the
diplomats, parties in labour negotiations, the old estates of the realm or
even modern coalition governments negotiate, they do not debate and
vote on what is valuable or commendable (see Weber 1917b, 169).
If no agreement can be reached, the parties do not persuade each other,
but often resort to various kinds of threats, such as wars, strikes, boycotts
or refusal to cooperate. Debating and voting are, however, needed within
each party in order to ratify the agreement the negotiators have proposed,
but the ratification then resembles an acclamation in a referendum rather
than a parliamentary vote on a resolution. The more parties there are
involved, the more difficult it is to reach a compromise.
In some cases the boundary between negotiation and deliberation
is diffuse: for example, the Swedish estates, according to both the 1809
Instrument of Government and the Finnish Diet Act of 1869, were for-
bidden from using the imperative mandate in their separate internal
deliberations, although negotiation regulated the relationship between
­
40  C. WIESNER ET AL.

estates (see Pekonen 2014). The European Council (EC) consisting of the
heads of governments and state as an assembly is a mixture between an
inter-governmental ‘congress of ambassadors’, a presidency based on the
epideictic rhetoric of acclamation and the second parliamentary chamber of
the EU (Palonen and Wiesner 2016).
In coalition governments a kind of intra-parliamentary diplomacy is a
regular and necessary phenomenon. Such coalitions are based on a coali-
tion agreement, which the parties might invoke when the opposition
between them arises. In a broader sense all ministers in a parliamentary
government must strike a balance between their status as parliamentar-
ians based on a free mandate with government loyalty while at the same
time advancing the interests of their ministry within the government (see
Bagehot 1867). In such situations the question of relative weight of par-
liamentary debate versus the intra-governmental type of negotiation must
be analysed in detail.
There is a Platonic and Hegelian tradition, today represented by Jürgen
Habermas and a number of theorists of ‘deliberative democracy’ that
follow him (see e.g. Steiner et al. 2005), that assumes that debate ends
‘naturally’ by finding a ‘consensus’, without a vote. Debate—or rather
discussion—is presented as a necessary but passing, cathartic stage, which
concludes when the participants have found an optimal solution, by force
of ‘the best argument’. This argument presupposes that the criteria for
‘optimal’ or ‘best’ are given in advance and well-known, that no disputes
of the criteria are legitimate, and that when a ‘solution’ has been found,
everyone spontaneously agrees—or else is excluded from membership in
the unit.
For a political analysis of debates such claims cannot be taken at
their face value. However, historical examples of the rise of the cathar-
tic moment, a moment when consensus suddenly dawns upon everyone,
as an attempt to reach the end of politics, would be worth analysing,
although a textual documentation of it might be difficult. The analogy
to conversion—whether to Communism, Christianity or Islam, for exam-
ple—is striking, and such conversions have been described in several auto-
biographies. Equally interesting is the opposite of consensus: exclusion or
excommunication and the forms of sanctions they involve, for which there
are also autobiographical and literary descriptions among former members
of Communist or Christian sects (for a scholarly interpretation of such
excommunications, see the ‘oath group’ [groupe assermenté] in Jean-Paul
Sartre’s Critique de la raison dialectique, 1960).
READING DEBATES POLITICALLY  41

This discussion of the forms—whether hypothetical or historical—that


would limit or exclude debate is pertinent to the analysis of actual debates.
With respect to the limit, we should keep in mind that the freedom, fair-
ness and even the existence of debate at all rest on certain political condi-
tions (for example, that give legal protection to or that institutionalise it).
Parliaments in the Westminster tradition made a historical breakthrough
when they institutionalised the procedure of looking at items from oppo-
site points of view. In the actual analysis of debates we should also remem-
ber the opposite ideal type—the possibilities for eliminating or minimising
debates—in order to better understand the specific conditions of actual
debates.

2.3   Forms, Dimensions and Characteristics


of Debates

As we have seen, also the concept of debate contains multiple levels of


meaning and has a history of its own. To analyse debates politically, it
is important to consider the actual dimensions of debate from different
perspectives. Taking the parliamentary debate as a paradigm, we can take
up here a closer discussion of the following aspects that characterise and
structure it: its rules and procedures, the agenda-setting and the items on
the agenda, the participants in the debates, the regulating of debates, its
stages and layers and, more generally, the political times of debates.

2.3.1  Rules and Procedures of Debate


The ideal-typical and paradigmatic character of parliamentary debate is
connected with the explicitly procedural character of parliamentary poli-
tics as well as with the controversies over procedure as a core feature of
the rules themselves. Such controversies involve the role of interpretative
struggle and remind us of the possibility to change the rules when new sit-
uations arise for which the old rules appear insufficient or have unwanted
consequences. For example, the increasing willingness of parliamentary
members to intervene in plenary debates has necessitated restrictions of
speaking time and opened up the question of how to do this in a fair way.
The specific rules of debate vary to a considerable extent according to
the type of parliament in question. Roughly speaking the British parliament
is the best historical approximation of a deliberative assembly, while the
US Congress may be regarded as a parliament of more n ­ arrow scope, that
42  C. WIESNER ET AL.

is, as a legislature, and the French Assemblée Nationale has been regarded
as the model of a representative assembly. In continental European coun-
tries, elements of all three are available in different mixtures, whereas the
European Parliament was formed according to the French model, for
example, in its committee system and in the powers of the President. The
three types of parliament have much in common, but also differ from
each other in remarkable ways. For the study of parliamentary debate,
the deliberative character (as a rhetorical genre, as opposed to forensic,
epideictic and negotiating genre) of the British House of Commons offers
the best historical model, and will be used below as our main point of
reference.
There are few written rules in the Westminster parliament. The
Westminster tradition of parliamentary rules relies on precedents, as mani-
fested in the title of John Hatsell’s four-volume Precedents and Proceedings
in the House of Commons, with Observations (1779–1796, revised edition
1818). In contrast, French parliamentary procedure relies on written
rules, règlements (see esp. Pierre 1887). Also in Westminster since the
early nineteenth century the role of Standing Orders, that is, procedural
rules that have been adopted until further decision, instead of being valid
only for the current parliamentary year, have become more important.
The prolongation of the parliamentary agenda has made such a tendency
unavoidable.
Regulations based on precedent could be the more flexible. Gilbert
Campion defends the Westminster practice of ‘the unwritten rules, or
practice’ of the House which exist principally for the sake of ensuring fair-
ness and fullness of debate, and are on the whole in the interests of Private
Members; and the Standing Orders which aim at ‘getting work done, and
are on the whole in the interest of Government’ (Campion 1929, vii). His
concern is that the individual members are losing their control of the rules
of parliament to the government, which threatens the ‘fairness and full-
ness of debate’. While nowadays the regulations have grown more com-
plicated, the parliament has also taken precautions to give more occasions
for debate initiated by the backbenchers of all parties (see Griffith and
Ryle 2003). Every parliamentary session can also alter its Standing Orders.
There is a distinct type of ‘parliamentary order’, but its exact content
remains a matter of debate. For instance, the distinction between ‘parlia-
mentary’ versus ‘unparliamentary’ language and conduct has been used
in Westminster since at least the 1620s. It presupposes a divide between
what is legitimate to say and do inside the parliament and what is not. The
READING DEBATES POLITICALLY  43

tracts on parliamentary procedure are major resources for discussing such


questions (see Palonen 2014c).
The tracts by Hatsell, Bentham, May and Campion on parliamentary
procedure are largely inventories of the history of the procedural con-
troversies and their current status. There are frequent possibilities for
applying different rules when making decisions on how to proceed, and
new realities—the Internet, for example—prompts situations for which
no rules exist, but in such cases members can apply analogies to existing
rules of debate. Do the blogs of parliamentarians, for example, fall under
the domain of ‘unparliamentary language’, or must this concept be strictly
restricted to speeches held in parliament (see Vaarakallio 2015)?
In extreme cases, as when the obstructive speeches of Irish members
were paralysing the House of Commons in 1881, Speaker Brand, sup-
ported by Prime Minister Gladstone, invented a rule of urgency to stop
the debate. It was recognised that the continuous functioning of the par-
liament is something presupposed by its rules, even if it was not and could
not have been regulated by a distinct rule (see Redlich 1905; Vieira 2015).
In other words, behind the existing rules and actual controversies on their
interpretation, there exists a second-order debate concerning parliamen-
tary order. It is conducted by the members themselves, by the Speaker, by
the committees on procedure and by the authoritative interpretations of
the tracts of procedure, written by the parliamentary staff or by members
themselves. Contributing to the parliamentary character of the rules of
debate is the fact that they themselves are subject to debate as to how they
should be interpreted.
Recently intra-parliamentary debates on procedure have been con-
trasted with attempts to apply universalistic rules of ‘good governance’ or
‘codes of conduct’ to parliaments and their members (see Beetham 2006).
Such quasi-legal instruments might help, for example, in regulating the
financial dependencies of the members, but in principle they are as contro-
versial as intra-parliamentary rules, and raise the question of parliamentary
autonomy against the powers of outside legal or other experts, who are
subject to no parliamentary-style debate.
This contrast is between the parliamentary principle of fair play and the
forensic principle of good order. Parliamentary self-regulation as practised
in Britain and in the Scandinavian countries is in contrast to the powerful
constitutional courts of the USA and post-war West Germany (for a defence
of the former systems, see Tomkins 2005, Bellamy 2007; for the latter,
see e.g. Rawls 1993). For the analysis of debate, the intra-­parliamentary
44  C. WIESNER ET AL.

debates on procedure are most interesting, but struggles between parlia-


mentary bodies and Supreme Court or Bundesverfassungsgericht decisions
are also a highly relevant topic for a second-order debate.
Comparative analysis of intra-parliamentary debates on procedure and
the changing interpretations of procedure form another valuable topic
for debate analyses. The analyses may concern, for example, judgements
about speaking to ‘the matter’ or ‘the question’, the powers of the Speaker
(President), or the changing interpretations of unparliamentary language
and conduct (which in some parliaments are based on a list of prohibited
expressions, in others on more contextual criteria). Accusing a member of
‘lying’ has been a classic example of unparliamentary speech, which from
early on has been circumvented by a variety of expressions such as being
‘economical with the truth’.

2.3.2  Debates on the Agenda and Debates of the Agenda


The Westminster procedural tracts and documents mainly focus on
debates on parliamentary agenda items. They present a refined set of rules
for debate and discuss controversies over their interpretation. The debates
on amendments and adjournment as well as on the committing of bills—
sending them to the committees—as well as on the raising of procedural
questions (discussed in the previous section) must begin with a motion
already on the agenda. The struggles over the agenda itself have gained
much less attention, although they are increasingly important politically.
The early twentieth-century political debates were largely based on differ-
ent answers to the same questions; today the questions themselves and the
controversies over setting them on the agenda are at least as important as
the political dividing lines.
The parliamentary rule is that a motion on the agenda can only move
‘forward’ to the final stage of vote. At Westminster a motion already con-
tains a resolution, a question to be put to the vote. ‘Motion, Question and
Decision are all parts of a process that may be called the elementary form
of debate,’ as Gilbert Campion put it (1929, 143). Therefore, no separate
motion to oppose the original one is needed. A motion is taken off the
agenda, rejected without a vote, if it lacks a ‘seconding’ by another mem-
ber. Another way of silent rejection is the adjournment of a motion sine
die, to an indeterminate future date. A third way is ‘burial’ of the motion
in the committees, which prevents it from ever coming to a third reading
READING DEBATES POLITICALLY  45

or vote before the next elections. Such filibusters are practised especially in
the US Congress. All of these means are legitimate tools in political tactics,
but they can also be used to threaten to cause a destructive paralysis of the
entire parliament.
A motion on the agenda can always be dealt with by amendments, which
are, as mentioned, the Westminster way of presenting an alternative to a
motion, whereas in continental parliaments full counter-motions opposed
to the original are required. In many assemblies, including the Oxford and
Cambridge Union Societies, when an amendment is voted for, the original
motion is lost without ever arriving to the vote (see Haapala 2012).
Reginald Palgrave, who succeeded May as the Clerk of Parliament,
insisted, however, that in Westminster this is not the case. The amendment
interrupts debate on the original motion, but when the debate on the
amendment has terminated, it will be confronted again with the original
motion: ‘The systematic discussion, both of a motion, and of its amend-
ment, or of an amendment to an amendment is impossible, without an
adoption of that method for deliberation which is expressed in that for-
mula used by Parliament’ (Palgrave 1878, vii). The main point is to have
‘an alternative choice between amendment or motion’ (ibid., 7). In other
words, the procedure of Westminster excludes the possibility of eliminat-
ing the original motion by a popular amendment (which removes the
original motion from the agenda) that has no real chance of passing all the
stages of parliamentary debate. In the continental parliaments, competing
proposals of no confidence in the government by competing opposition
parties are a part of political tactics.
Besides the legal initiatives—including today EU legislation—there are
other regular items on the agenda of parliamentary plenary sessions which
are debated according to specific rules. The annual budget debates are the
oldest layer of parliamentary control of government. Votes of no confi-
dence in the government were initiated by Sandys’ motion to the Walpole
government in 1741 (Turkka 2007). Additional forms to control the gov-
ernment include written and oral questions to ministers. Furthermore,
government declarations on foreign policy, on treaties to be ratified and
so on are today important occasions for debate, and the opposition parties
as well as the backbenchers across party lines also have their own occasions
to initiate debate.
The entry of topics onto the parliamentary agenda deserves to be ana-
lysed in terms of the heading of the topic and by whom it was initiated.
46  C. WIESNER ET AL.

When studying debates on a specific item on the agenda, we have to


interpret the character and political significance of the item in relation to
the parliamentary manner in which it is put on the agenda and its place in
the political constellation. This is not always clear in advance; for example,
a government may interpret the passing of a resolution as a question of
confidence, even if the opposition does not have the intention to cause
the fall of the government on that occasion.
The questions of how, when, by whom and concerning what mat-
ters a motion can be put on the agenda are also regulated by the proce-
dure. Since the early nineteenth century a distinction has been made at
Westminster between regular motions (orders of the day) and facultative
motions (notices), that is, between issues the parliament is expected to
deal with and items that members want to raise onto the agenda. In the
course of the nineteenth century this distinction began roughly to cor-
respond with that between the government’s bills and motions versus an
individual member’s motions.
Walter Bagehot (1867, esp. 23) understood the cabinet (the govern-
mental core) as the executive committee of the House of Commons.
Subsequently this principle was interpreted as giving priority to govern-
ment initiatives on the parliamentary agenda. Justifications for this pri-
oritisation referred to the fact that major legislative reforms needed an
administrative apparatus for their preparation, which the individual mem-
bers and even the opposition parties did not have at their disposal. This
government priority is even more pronounced in legislative types of parlia-
ment, including the European Parliament.
The danger of reducing parliament into a mere ratifying assembly or a rub-
ber stamp was a frequent topic around 1900, and in connection with which
the Whittaker Committee (1913–1914) discussed proposals to strengthen
the status of individual MPs (discussed in Palonen 2016). With respect
to agenda-setting, we can analyse the debates on how to strike a balance
between efficient decision-making and ensuring fairness of debate (see the
debate on ‘fair play’ in the procedural legislation of 1882, in Palonen 2014a).
Today governmental initiatives have a much greater chance of being
passed by the parliament (although their details are frequently modified
by  the debate) than do individual member’s initiatives, which seldom
reach the final vote at all. However, the latter initiatives are important,
because the topics are often picked up later in the government’s motions
or in party manifestos. Above all, the initiatives of backbenchers illus-
trate the political imagination of parliamentarians, the limited intellectual
READING DEBATES POLITICALLY  47

horizons of governments and majorities and the resourcefulness of indi-


vidual members in bringing to the agenda new and unknown topics (for
recent Westminster debates, see Wright 2012; Crewe 2015). We can also
study new interpretations of concepts or the launching of new slogans and
catchwords through the initiatives of backbenchers.

2.3.3  The Actors in Debate


The rhetorical profile of parliamentary speakers will depend on the debaters.
Ministers versus opposition leaders is the basic setting, but in Westminster
the front versus the back benches is another important dividing line. For
backbenchers, skill in debating is an important skill for it increases their
visibility and competence in scrutinising government and compensates as
well for their lack of numerical support. In continental parliaments the
contrast between party and parliamentary leader is frequently pertinent.
In the more specialised committee systems of the continental variety, the
committee chairs also hold a privileged position as parliamentary speakers,
as do also the rapporteurs in the European Parliament (see Section 4.4).
When elections became regularly contested, the divide between parlia-
mentary and platform speeches, especially those in the electoral campaign,
became more relative. For the debate character the parliament remains the
model, because in elections the debate with competing candidates is only
indirect—nobody expects any candidate to persuade the other or to vote
for their own rival. The strength of the parliamentary culture largely deter-
mines how much the campaign debates are shaped by the parliamentary
agenda, instead of by the personalities of the candidates (in line with the
ethos in epideictic rhetoric) or by pandering to the prejudices of the local
electorate. In public meetings and party conferences, unlike parliaments,
the adversaries are beyond the pale, and the debate among partisans con-
cerns rather the tactics towards the adversaries.
The tendency to speak in parliament ‘to the peanut gallery’, to the
electorate, the party or the press, has obviously let ‘platform’ tendencies
invade the parliaments. Many speeches tend to be declarative rather than
contributions to debate, although being formally addressed to the parlia-
ment (on the origins of this phenomenon, see e.g. Vieira 2015).
Instead of parliamentarians adapting themselves to extra-parliamentary
audiences, we can, conversely, think of the parliamentarisation of these
audiences. The spread of parliamentary pro et contra debate to meetings,
organisations, assemblies and institutions outside parliament can be an
48  C. WIESNER ET AL.

object of study. An eloquent defence of this practice was presented in


Walter Bagehot’s The English Constitution, in which he gave the following
characterisation of parliament:

The great scene of debate, the great engine of popular instruction and
political controversy, is the legislative assembly. A speech there by an emi-
nent statesman, a party movement by a great political combination, are the
best means yet known for arousing, enlivening, and teaching a people. The
cabinet system insures such debates, for it makes them the means by which
statesmen advertise themselves for future and confirm themselves in present
governments. It brings forward men eager to speak, and gives them occa-
sions to speak. (Bagehot 1867, 14)

For Bagehot the parliamentary ‘cabinet government’ is a regime that,


unlike US presidentialism, promotes the issues and practice of debate and
political controversy to the citizenry outside the parliament. ‘The nation
feels that its judgment is important, and it strives to judge. It succeeds in
deciding because the debates and the discussions give it the facts and the
arguments’ (Bagehot 1867, 15). In the 1872 edition of his book Bagehot
still insists on the omnipresence of the parliamentary model of ‘govern-
ment by discussion’:

The distinguishing quality of Parliamentary Government is, that in each


stage of a public transaction there is a discussion; that the public assist at
this discussion; that it can, through Parliament, turn out an administration
which is not doing as it likes, and can put in an administration, which will
do as it likes. (Bagehot 1872a, 99, see also 1872b)

The debating parliament is the main counterforce to arbitrary rule by


the administration, also in the sense that ministries are led by politicians
who are responsible to the parliament. The wider significance is that the
parliamentary model of a debate-based institution has been extended to
the rules of ordinary meetings (cf. Robert’s Rules of Order, 1876), and
in  ­nineteenth-century Britain also to the Union Societies (see below in
Section 4.3) as well as Parliamentary Debating Societies at the local level.
Through the conduct of meetings, the formation of voluntary associations
and self-­governing institutions, including universities, something of the
parliamentary model has become an inherent part of the European polities.
For the analysis of debate it is important that the electoral model with
its plebiscitarian elements of epideictic rhetoric, and the parliamentary
READING DEBATES POLITICALLY  49

model based on deliberations pro et contra rely on opposed principles.


Demands for surrogates of an imperative mandate that is party or con-
stituency based, advocated by many nationalists and socialists (see e.g.
Pekonen 2014), are wary of debate or of questioning every view, because
they claim to ‘know’ what should be done and therefore do not want to
leave matters to parliamentarians to argue for and against and possibly
decide otherwise. The democratisation of parliamentary culture, in con-
trast, both increases the number of debaters and parliamentarises the rules
of debate both in public assemblies and in private associations. The parlia-
mentary model also insists that the vote is no ‘voice of God’ (vox populi,
vox Dei), but just the last move in a debate.
In relation to the vote, we can, however, distinguish between two types
of parliamentary debate—symmetric and asymmetric. In the symmetric
struggle the parliamentarians are equal and debate with each other on the
strengths and weaknesses of a motion or resolution. In the asymmetric
struggle the parliamentarians oppose members of government and admin-
istration, examine their policies and procedures and aim at changing an
existing policy or replacing the existing personnel (see esp. Weber 1918,
235–237). The symmetric debate is even then conducted among parlia-
mentarians, some defending the government and others opposing it, or all
of them critical of the government but in different degrees. The antago-
nism between frontbenchers and backbenchers mixes both aspects, with
the focus more on individual members’ parliamentary rights than on the
strengths and weaknesses of a government’s measures or of a backbench-
er’s amendments.
The rhetorical profile and competencies of the individual actors must
also be considered, including the possibility that in a new debate a poli-
tician may always show some abilities unseen in her previous contribu-
tions to debates. Every new debate can also be understood not merely as
an exercise in the existing rhetorical repertoire, but also its expansion or
renewal.

2.3.4  Regulators of Debate
In the early histories of parliament, the juridical element, the High Court
of Parliament, was prominent, and the British House of Lords, for exam-
ple, still has legal powers. Inside parliaments the Speaker or President, as
the case may be, and committee chairs have quasi-legal powers, analogous
to that of a referee in football, and the same holds for the chair of any
50  C. WIESNER ET AL.

meeting. Like a referee, the Speaker is the interpreter of the rules during
debate, although, unlike the players in football, the parliamentarians can
under certain conditions act as the guardians of the rules and take back the
powers delegated to the Speaker.
The Speaker in Westminster has always represented parliament before
the government and the Crown, and in cases of obstruction, the Speaker
also defends the parliament against its own members. The Speaker,
although elected by the parliament, does not represent  the  major-
ity, but is in a position above and beyond the contest between the
government and the opposition, or the majority and the minority. In
Westminster this position was finally confirmed in the mid-nineteenth
century, when the Speaker was denied the possibility to participate in
either plenary or committee debates. In the European Parliament, how-
ever, the President can still participate by leaving the chair (as discussed
in Section 4.4).
The regulation of debates by the Speaker in Westminster follows the
principle of recognising speakers in alternating turns, for and against,
among those rising to intervene in the debates. The Speaker has consider-
able powers to anticipate what each member will likely say in the debate,
as the partisan divisions facilitate this task. In multiparty parliaments the
position of parties has been more firmly institutionalised and members are
expected to speak in the order in which they are registered on a list.
The possibility of replies has elevated the role of the Speaker in most
parliaments, and the Speaker’s control in judging whether a speaker on
the floor is speaking to the matter, and in making spontaneous interpre-
tations and so on, has also contributed to the same. A special case is the
Speaker’s powers to ‘name’ a member for irregular conduct in a session.
The member is asked to take back an unparliamentary expression, and if
she refuses to do so, the Speaker names the member, or calls her ‘to order’.
Other cases include the power to interrupt debate through the clôture and
guillotine (see Section 2.3.7).

2.3.5  Multistage and Multilayer Debates


In everyday language, debate is a single event. In the parliamentary sense,
however, debate refers to a whole series of events or types of events, to all
phases of parliamentary process, including the vote as its last move. The
examples from Hansard discussed in Sections 4.1 and 4.2 illustrate how
past debates can also become part of current debates, including through
READING DEBATES POLITICALLY  51

­ recedents, allusions to previous debates and an MP’s own past speeches


p
and positions.
In this section the focus is on the paradigmatic form of how parlia-
ment debates motions or bills in several stages. Combining the classical
three readings of a bill with the regular inclusion of the committee stage,
Gilbert Campion presents ‘principal stages in the discussion of Bills’ at
Westminster:

Introduction
Second Reading
Committee Stage
Report Stage
Third Reading, and
Consideration of Lords’ Amendments
(Campion 1929: 176)

The formation of distinct parliamentary rules and practices for agenda-­


setting and for debate transforms the concept of debate from an isolated
event into a political institution. One of the major features of parliamen-
tary politics is the extensive time it provides for debate for dealing with
items on the agenda thoroughly and from opposite points of view. Each of
the stages that Campion mentions refers to a different perspective on the
item on the agenda: the introduction presents the motion to the parlia-
ment with a preliminary debate; the second reading is the classical stage
for thorough and substantive plenary debate; the committee stage—which
may contain several different committees—is dedicated to discussion of
the details of the motion, each committee from a different viewpoint; the
report stage focuses on amendments proposed by the committee, accord-
ing to which the plenum might revise the motion; the third reading is the
stage to debate whether the motion on the table shall be passed or not; the
last stage is a bicameral moment, which has dwindled in significance after
the Lords lost their powers, but can still be important in more technical
matters or in terms of the entire timetable of the motion.
The multiple stages of debate mark the singularity of parliamentary
politics. Each stage has different aims, different types of motions and dif-
ferent rules of debate (including, for example, between the House plenum
and the committees—or the Committee of the whole House—or depend-
ing on the different persons participating in the debate, for example, sit-
ting on committees with selected membership). One of the strengths of
parliament is its ingenious use of time as a political medium for inventing
52  C. WIESNER ET AL.

new perspectives, for developing new arguments and for reconsidering


arguments and positions taken at the previous stages. The political use
of time as an intrinsic part of parliamentary debate is a key feature of
Westminsterian parliaments, making them the model par excellence of a
deliberative assembly.
The political significance of parliamentary moves can be illustrated by
three key aspects of parliamentary politics: the separation of items in time,
the amendments and the adjournments. These aspects have been devel-
oped most fully in the Westminster parliament, but can be, mutatis mutan-
dis, applied to other parliaments and the debates of other assemblies.
Parliamentary time is discontinuous: it consists of separate, successive,
non-simultaneous and irreversible items on the debating agenda. In a par-
liament there is always some minimal interval put between the items on
the agenda. The moves between items must be proposed by someone,
either by the Speaker or by a member of parliament. Until the second half
of the nineteenth century also technical transitions, such as that from the
House plenum to the Committee of the whole House and vice versa, were
(mis-)used as occasions to debate, but the practice was changed to enable
these shifts to take place without debate (see May 1883). To the disconti-
nuity in parliamentary time we can relate the rule that parliament cannot,
within the same session, replace a resolution that it has already passed (see
Hatsell 1818/2, 132–133).
Campion formulates the aim of an amendment thus: ‘An amendment
is a subsidiary motion which interposes a new process of Question and
decision between the main question and decision […] for the purpose
of altering its terms’ (Campion 1929, 48). As we have seen with John
Stuart Mill’s amendment, moving an amendment is also a play with time,
an interruption of the present debate, to reassess the strengths and weak-
nesses of the original motion and to begin a new debate in which those
that spoke on the original motion gain the right to speak again.
The motion for adjournment in the middle of a debate has a roughly
similar temporal structure. The adjournment does not just propose a
­postponement; Adjournment of debate is done in order to create more
time to rethink and reformulate a stand, for example, to rally the majority
backbenchers to support the government. The motion for adjournment
also creates a sudden interruption in a debate, creating time for reconsid-
eration of the strengths and weaknesses of the adjourned motion. Hatsell
notes that a motion for adjournment has become a possible strategy in
cases where ‘questions are moved, upon which the House do not wish to
READING DEBATES POLITICALLY  53

give any opinion’ (Hatsell 1818/2, 112). This is most definitely the case
with adjournments sine die. In terms of time, we can distinguish between
postponing, suspending or interrupting adjournments. In the twentieth
century they have been turned into new occasions for a general debate on
government policy, a possibility to take an urgent matter onto the agenda
or to allow the backbenchers to initiate a debate. But in these cases the
adjournment motion is no longer followed by a vote (Campion 1929, 91;
Griffith and Ryle 2003, 378–379).
The converse side of the use of time in parliament is the recogni-
tion that parliamentary time is always limited, and different devices have
been created to terminate debates within a reasonable time. The classical
Westminster rule is that a member can speak in the plenary only once.
As Henry Scobell explained, it is ‘the liberty that every member hath in
a Grand Committee, as well as in other Committees, to speak more than
once to the same Business […] which is not permitted in the House’
(Scobell 1656, 35). Although later replies and a final word from the mover
have been accepted (see May 1883 and the list of cases in Campion 1929,
167), this basic rule gives Westminster debates their own tone. Other time
limits are intra-parliamentary, such as the clôture to terminate debate and
the guillotine method of scheduling in advance the time reserved for an
item.
These examples illustrate how closely the parliamentary moves are
bound up with the use of time. This kind of intra-parliamentary time is sel-
dom made explicit in the rules of procedure or in the debates themselves.
It is rather presupposed in the practice of debate and well-known among
experienced parliamentarians, but easily missed by outside observers and
analysts. The examples of amendments and adjournments illustrate how
the same moves can be used for different purposes, and how they relate
to present, past and future. In order to judge parliamentary time politi-
cally, it is pertinent to identify the intentions of the movers and to evaluate
the consequences of the moves, which do not always correspond to each
other. We can speak of a temporal subtext of parliamentary politics. This
is often invisible to outsiders, though it is understood quite well among
the actors themselves. The temporal subtext can have unintended conse-
quences for the entire political constellation.
The idea of using time as part of the practice of argumentation in
debates accentuates the contrast of parliamentary politics to business or
administrative models of decision-making. In the contemporary world the
timeliness of parliamentary politics serves as a counterforce to the seldom
54  C. WIESNER ET AL.

questioned tendencies of speed and acceleration of time (see Koselleck


2000). Slowness and ‘bavardage’ have been the standard accusations of
anti-parliamentarians, such as Thomas Carlyle (1850) and Carl Schmitt
(esp. 1922, 1923). Today they rather indicate a sign of a willingness to
dedicate sufficient time to debating items from opposed perspectives.
Reservation of time for the deliberations does not, however, necessar-
ily mean a slow parliamentary process, although some provisions against
haste are contained in the procedure. Precipitous decision-making in
urgent matters is always possible—even if what is urgent and what is not
is itself a controversial question. Under the ever-growing agendas and the
corresponding pressures on time, the use of sufficient time as an argumen-
tative resource in debate no longer concerns single motions so much as
the prioritisation of what exactly is worth a time-consuming and thorough
debate, which can create a topic for a secondary debate.

2.3.6  The Scarcity of Time for Debate


To bring up questions of parliamentary time today is commonly to refer to
the contrast between parliamentary deliberations and everyday life outside
parliament. The accusation of wasting time by endless talking is reflective
of the fatigue of members in the face of endless debating and it under-
scores the contrast of the parliamentary rhythm with its session periods
and electoral terms to the ‘ordinary’ calendar of hours, days and weeks.
Bavardage is not just boring; it is carried out at a high cost to human
endurance, in marathon speeches and overnight sessions. Even if in recent
decades such sessions have declined in many parliaments, Westminster
included, the great number of hours worked by parliamentarian still shape
the picture of parliamentary debates.
Obstruction has always been used as a means to prevent the realisation
of legislation before the next elections. However, it was the Irish members
around Charles Parnell who in the 1870s and early 1880s applied obstruc-
tion as an instrument against the Ireland policies of both the Disraeli and
the Gladstone governments. The militant Irish members did not iden-
tify with the British Empire or with parliamentary government. They
learnt to use parliamentary resources to paralyse parliament as a whole
and attempted to delegitimise parliamentary politics as such. They used
speeches to merely consume time and wear down the other MPs. After
41 hours of speaking, Speaker Henry Brand introduced a state of par-
liamentary emergency. It marked an extreme event in the parliamentary
READING DEBATES POLITICALLY  55

calendar and was exercised for the sake of the affirmation and functioning
of parliament itself. The speeches by the obstructionists did not follow the
intra-parliamentary time of debate, but rather prevented it (see the inter-
pretations of Redlich 1905 and Vieira 2015).
With the increasing pressure on parliamentary time, its calendari-
sation has become a practical means to cope with the different items.
Parliamentary fair play has been turned into a question of the fair dis-
tribution of parliamentary time, both between members and between
the different types of items. The struggle to a large extent turned into a
struggle between the government and the opposition over time, as well as
one between the front and back benches. For the government the power
over parliamentary time was an instrument to realise its programme, while
the opposition and the backbenchers were afraid that this would lead to
government rendering parliament into a merely ratifying ‘rubber stamp’.
Empirical studies on the use of time consumption indicate, however, that
both the opposition and the backbenchers have retained an important
share in parliamentary time at Westminster (see Campion 1958; Griffith
and Ryle 2003).
An important background for the history of the struggle on parliamen-
tary time is that parliaments—with the exception of the US Congress—
were until after WWII assumed to be parliaments of amateurs. The
members received some compensation (in Westminster only after 1912),
but were not expected to live ‘off politics’, as James Bryce (1888/1914)
and Max Weber (1919) would say. The professionalisation of parliamen-
tarians on the full-time basis of salaries was resisted in the name of limit-
ing parliamentary powers (e.g. Bismarck) and out of envy amongst the
population against the idea of getting payment for ‘mere talking’. The
professionalisation of parliamentarians through the debates on compen-
sation, salaries, staff, pensions and so on has still never been properly
analysed in a broad historical and comparative basis (for some cases, see
Palonen 2012a). Since the mid-1970s the idea of full-time parliamentar-
ians has been institutionalised in the Western countries, although calls for
­returning to amateur parliaments can still be heard—with the implication
that powers would be transferred to the governments and administrations.
The political problem concerns how to coordinate intra-parliamentary
time with coping with time outside parliament. The parliamentary calen-
dar of today begins to resemble that of any ‘bourgeois’ workplace, and
the full professionalisation of parliamentarians has been used to justify this
trend. The tendency to subordinate the internal parliamentary rhythm of
56  C. WIESNER ET AL.

readings, amendments and adjournments, of the interplay between the


plenum and the committees, to the calendars of the outside world is,
however, questionable. The parliaments seem to be ready to give up the
rhetorical ideal of thorough debate due to the demands of everyday life.
An alternative would be to expand, in the style of Bagehot and others, the
parliamentary-type institutions of debate to a broader group of citizens.
The question also concerns the value of the parliamentary lifestyle based
on dissensus and debate as the best model for the political aspects of the
human condition.

2.3.7  Terminating Debates
We can imagine a number of ideal-typical ways of terminating debate.
According to the parliamentary paradigm a debate ends with a vote on a
resolution of the assembly. The transition to the vote follows when none
of the actors has anything more to say on the item on the agenda, when
debate has been ended through a vote by a qualified majority (clôture)
or at the end of the time reserved for it (guillotine, timetabling). In par-
liaments of today there are also debates without votes, occasions for an
exchange of opinions on suddenly arising matters or for backbenchers to
introduce items onto the agenda without a resolution. Such occasions for
brainstorming can become predecessors to future, proper debates or indi-
cators of whether a more thorough debate is needed.
The parliamentary model illustrates that there is no ‘natural’ end to
debate. When there is a resolution, debate is presupposed, in parliaments
in the committees above all, with the object of coming up with objections,
though in a single sitting it often happens that no proper debate spontane-
ously arises on an item and the resolution is then passed without a vote.
Even brainstorming sittings end when the members have lost interest or
are too exhausted to say anything more.
The ends that obstructionists have in mind, as discussed above, are not,
strictly speaking to debate. Rather, obstruction itself is the end of debate
in the sense that the speeches lose their debate character and are turned
into instruments for using up time in order to prevent parliament from
passing a resolution. A divide occurs among those who, in a session with
obstruction, still speak to the matter, and then when a sizeable number of
them no longer do. The course taken is not, of course, predictable at the
outset, nor can one necessarily refer to a definite point in time when the
change occurs.
READING DEBATES POLITICALLY  57

In less formally regulated situations, such as in academic debates, we


cannot speak of an end of debate, or do so only in historical i­ nterpretations
of specific, already terminated debates. Moreover, old debates can be
reopened, and such situations are also interesting topics of study in terms
of the recontextualisation of a former agenda-setting and the topoi of the
argumentation and so on. In these cases, we should notice the contextual
differences and ask whether a debate is new or whether it is an old one
reactivated. Both the intentions of the agents who actually contribute to
the debate and of the interpreters who claim that they do are also worth
a closer analysis.
We have also defended the view that it is not only scholarly works and
pamphlets, but also the more impersonal documents of constitutions,
peace treaties, party manifestos and government white papers that are
parts of debate. In a sense, the production of such documents represents
an end to debate in so far as they mark a transition from action to the fab-
rication of a product (in the sense of Hannah Arendt 1958).
At the same time as such documents contribute to a debate they might
also reorient the agenda-setting and thus open up new debates. A single
document, such as the rules of procedure of the EC, can be interpreted
from all of these perspectives. For the participants it may look like a
guideline for action, for jurists some specific problems of interpretation
might become visible in it, while for political analysts it might make cer-
tain interpretations of the political character of the EC obvious, and for
scholars it might suggest possible interpretive approaches (see Palonen
and Wiesner 2016).
Similarly, in a party manifesto, scholars might detect expressions of
intra-party majorities, whereas different factions might find their hobby-
horses or signs of inner-party lobbies in it, or intentionally diffuse formula-
tions, even glaring omissions, to hide disagreements or to help members
avoid taking a stand on conflicts inside the party. Obvious differences with
the party’s previous standpoints may become visible when comparing the
texts, and some formulations might be responses or adaptations to the
views of competing parties.
Debates exist almost everywhere, and their seeming absence is as inter-
esting as their presence. Here we have presented a number of issues related
to the range and forms of debate from different perspectives. As every-
thing cannot be analysed simultaneously, it is up to scholars to choose the
specific debates or problematic in debates as topics of analysis. To make
such choices, political literacy is a necessary prerequisite.
CHAPTER 3

Research Practices and Operations


in Studying Debates and Documents

This book treats the topic of how to analyse debates and documents as
political activity. More explicitly, we might say that the focus is on debates
and documents as parts, and as arenas and reflections, of political activity
and political processes, strategies and actions. This chapter is dedicated to
the research practices (or the methods and techniques, if you will) and the
core steps of this kind of analysis. Some general considerations are laid out
that will be valid and helpful for most interpretative and textual analyses,
with additional emphasis set on the question of how different approaches
can be used to analyse the political activity linked to the texts. This requires
finding out the relevant information about the moves, strategies, interests
and actors involved in the political processes in question. The research
interest and research question target these aspects, rather than simply the
contents of the text, and this, in return, crucially determines the setting of
the analysis, the material selection, the research questions, and the catego-
ries and course of the analysis.
In this context, the approach suggested in this book cannot be classi-
fied as following any definite school or methodological approach—and
we must add that we would not argue that such clear schools of, say,
discourse analysis or rhetorical analysis even exist. In our view, what all of
these interpretative approaches have in common is that they are based on
the assumption that linguistic action is political action, and speech acts are
seen as illocutionary acts. How the analyses then proceed is a m ­ atter of the

© The Author(s) 2017 59


C. Wiesner et al., Debates, Rhetoric and Political Action,
DOI 10.1057/978-1-137-57057-4_3
60  C. WIESNER ET AL.

specific questions and interests. The practices proposed here, ­nevertheless,


are inspired by rhetorical analysis, discourse analysis, conceptual history
and other interpretative techniques. These approaches can all be used in
studying debates and documents as political action, and debates as politics.
As said above, the aim of grasping the traces, elements and dimensions
of political activity embedded in a text, or related to a text, requires analyti-
cal steps besides the ones involved in the basic steps of an interpretative or
qualitative study. The main difference is that such a study does not stop
when it has found out the ‘what’ of a text, that is, the utterances, argu-
ments, topoi or concepts that are used in it. On the contrary: when those
are clear, the main work begins, and the researcher asks herself about the
‘how’ and the ‘why’ behind the ‘what’.
Our perspective here is similar to the ones opened up by Michel Foucault
and Quentin Skinner. Foucault in his general considerations on discourses
famously stressed that we should ask why a specific utterance is appear-
ing in a specific place rather than another utterance (see Foucault 1971).
And Skinner focuses on the interpretation of the moves contained in the
utterance, whether the prevailing conventions of the situation are affirmed,
rejected, modified, parodied or otherwise subverted to provide them with
a different ‘point’ (see Skinner 1988).
In addition to the perspective that prompts the researcher to look into
the ‘why’ or the reasons for an utterance (e.g. the background, the actor’s
interests, the political context or the strategic aims related to it), the sec-
ond decisive perspective regards the ‘how’, or the way things are put:
which figures, tropes and topoi are used? Which words and terms? How
do they play with established narratives, with fears and with positive and
negative connotations (Wiesner 2014a, 114–117)?
In the following, different practices and possibilities for enquiring
into the ‘what’, ‘how’ and ‘why’ of debates and texts will be sketched.
Before entering into this field, however, some considerations about the
­researcher’s perspective will be lined out, stressing that she needs to keep
a distance from the material while at the same time participating in the
debate, and underlining that sources have a ‘veto power’.

3.1   The Researcher’s Perspective on the Sources

In Chapters 1 and 2 we discussed the similarities and differences between


parliamentary and academic debates. As research is always a contribution
to debate and, therefore, contains a political aspect, how, then, is the study
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  61

of politics possible at all? What can and should be the relationship of an


interpreter of politics to politics, her subject matter?

3.1.1  Keeping Your Distance While Being Part of the Debate


One of the major aesthetic innovations of the twentieth century was
Bertolt Brecht’s ‘epic theatre’ and its central tool, the Verfremdungseffekt,
sketched in detail in his opera Der Aufstieg und Fall der Stadt Mahagonny
in the late 1920s. Its key point is to insist on the difference between life
and theatre:

To render an event or a character strange means, above all, simply refus-


ing to consider them as self-evident and self-illuminating and to produce
a wondering and curious look at them. […] To de-familiarise something
also means to historicise, that is to present the events and persons as transi-
tory (vergänglich) ones. (translation KP, quoted from Brecht 1967, http://
de.wikipedia.org/wiki/Verfremdungseffekt#Brechts_Idee_als_Gegenpart_
zum_aristotelischen_Dramenbegriff)

Brecht’s point has certain consequences for the study of politics and his-
tory. The scholar must always recognise that there is a gap between her
own situation and her object of research. This is a perspective that has
been mentioned by political thinkers and historians alike.
Reinhart Koselleck refers to Verfremdungseffekt in the Einleitung to
the Geschichtliche Grundbegriffe to emphasise the difference between the
past and the present meanings of concepts. He advises scholars to see
this as an advantage: ‘Exposure to experiences that once seemed distant
and unfamiliar may sharpen consciousness of the present; such historical
clarification may lead to a more enlightened political discourse’ (Koselleck
2011, 16; the German version in 1972, xix).
Quentin Skinner also insists upon the idea that historians should not
only treat the past like a foreign country, but also turn the distance to their
advantage:

It is the very impression of familiarity, however, which constitutes the added


barrier to understanding. The historians of our past still tend, perhaps in
consequence, to be much less aware than the social anthropologists have
become about the danger that an application of familiar concepts and con-
ventions may actually be self-defeating if the project is the understanding of
the past. (Skinner 1970, 136)
62  C. WIESNER ET AL.

The need for a Verfremdungseffekt is even more apparent in political sci-


ence. Nothing is more difficult to study than the politics of your own time,
your own country and conducted in your own language because so many
aspects will be so familiar that the researcher will have difficulties in mak-
ing out what is special about them.
Scholars cannot compete with journalists in just referring to what politi-
cal actions do; they must take a distance from the matters they are present-
ing. They must also profoundly understand what they are doing instead
of explaining their activities by just applying received models of analysis
to politics. A manner of rendering contemporary politics intelligible is to
situate it in a wider context, of its own history, for example, or of other
countries in which something comparable appears, or by analysing the
language of its political agents from both a historical and a contemporary
perspective (for contributions along these lines, see Haapala 2013).
Scholars first need to defamiliarise themselves with their ‘own’ political
context and then, as a second step, try to create a Verfremdungseffekt to
illustrate that they can, in principle, understand it better than the politi-
cal agents themselves, without, however, devaluating their activity (see
Koselleck 1983; also Kivistö 2013). A distance in time and space as well
as in language between the scholar and her subject matter can serve this
purpose. The need for such a distance is also important when analysing
forms of politics that are adverse to the scholar’s own views, such as the
Nazi (see Parvikko 2008) or contemporary extreme-right ideologies and
parties (see Vaarakallio 2013).
Frank Ankersmit illustrates these problems with yet another parallel to
politics, namely with the concept of representation: ‘Aesthetic pleasure
is not possible until we have learned to accept the radical rift between
the real world and the world of artistic representation’ (Ankersmit 1996,
46). The situation is exactly the same for political representation: ‘We can
only talk of representation when there is a difference—and not an iden-
tity—between the representative and the person represented’ (ibid.). This
perspectivistic character of representation justifies the autonomy of the
parliament as a deliberative assembly oriented to debating pro et contra.
‘Just as the work of art has an (aesthetic) autonomy with regard to what
it represents, so has the representative a certain independence or auton-
omy with regard to the voters which sent him to Parliament or Congress’
(Ankersmit 2002, 112). Furthermore, Ankersmit sees ‘historical represen-
tation’ as an ‘explosion’ of points of view (Ankersmit 2001, 16).
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  63

The arguments above lend further support for a perspectivistic view


of all knowledge, discussed in Chapter 1 in relation to Nietzsche, Weber
and others. As Weber further emphasised, such perspectivism leads us to a
complete rejection of the standard textbook view of scholarly ‘objectivity’.
A close attention to what Weber wrote in his 1904 essay on ‘objectivity’
shows quite clearly that, for him, ‘objectivity’ cannot mean something
that is ‘already there’ and could be reached were the scholar able to get
beyond her own ‘subjectivity’ (see Weber 1904, 192, 196, 208). He also
rejects the possibility that ‘objectivity’ could ever be attributed to indi-
vidual scholars, as research itself forms a debate, and in the scholarship
there cannot be any neutral judge or official to serve as the incarnation of
a Hegelian ‘objective spirit’ against the subjectivity of politicians, seen as
somehow above debate (ibid., 157; see also Skinner 2002a, 6–7). Neither
does the ‘objectivity’ refer to a midpoint between differing points of view
or a consensus among scholars (Weber 1904, 153). Weber’s perspectivism
is compatible with the proliferation of different points of view and allows
the full use of the ‘subjective’ ideas for this purpose.
Weber regularly speaks of ‘objectivity’ in quotation marks. He refers to
the academic context of his 1904 essay: the quarrel in German-language
economics between the historical school of Gustav Schmoller and the
Austrian marginalist economics of Carl Menger. Weber quotes a Viennese
student who was lost between the ‘two kinds of economics’ (zwei
Nationalökonomien) (Weber 1904, 160–161). For Weber, the ‘conflict
about methods’, ‘basic concepts’ and presuppositions, the constant shift
of ‘viewpoints’ and the continual redefinition of the ‘concepts employed’
(ibid.; translated in Weber 2012, 108) are the real problems with ‘objec-
tivity’. The conceptual struggles in scholarly debates will, in principle,
never end (Weber 1904, 206–207).
In other words, Weber recognises that scholarly controversies cannot
be terminated without causing stagnation in research, and this is particu-
larly important in matters relating to politics and culture (Weber 1904,
153). This led him to reconsider the concept of ‘objectivity’ in terms of
regulating, instead of terminating scholarly disputes. My (KP) interpreta-
tion, based on connecting his essay on ‘objectivity’ with his 1918 essay
on the parliamentary control of the knowledge of officials (Weber 1918,
esp. 235–237), claims that Weber’s regulative idea for scholarly controver-
sies is fair play and British parliamentary procedure provides the historical
model for such fair treatment (see Palonen 2010a, b).
64  C. WIESNER ET AL.

The ideas proposed by Brecht, Ankersmit and Weber are immediately


relevant for the analysis of texts and debates. As no scholar can claim to
have found ‘the philosopher’s stone’, but must admit to being one debater
among others, she does better by constructing a highly one-sided profile
of interpretation. Only then has the scholar a chance to say something new
and original about the subject matter. Ideal types of interpretation are nei-
ther true nor false, but each has its strengths and weaknesses. They remain
a matter of debate, during the course of which they may be revised, and
the criteria for judging strengths and weaknesses can also be contested.
In the course of debates, scholarly perspectives not necessarily converge;
consensus can rather be regarded as a sign of stagnation. There are strong
reasons not to conform to the ‘mainstream’ of the discipline, but rather to
stick to one’s own profile and to strengthen it further.
There are some further general considerations on the role of the
researcher in interpreting texts that should be kept in mind. Firstly, it
is paramount that she does not overinterpret a text. A common pitfall
is to jump to conclusions that are not justified by the sources. A way
to avoid this is to explain the context in which the text is embedded. A
detailed explanation of the context makes it easier to analyse the text in a
way the actors themselves would have experienced it. Second, the limita-
tions of interpretation affect the position of the researcher. She should
be aware of the limits of what is possible to do with the available sources
or with the approach used in the study. Being aware of one’s own posi-
tion as a researcher is one of the most important things to bear in mind
while conducting research. Finally, it is no less significant to keep in mind
that a researcher participates herself in debate as she conducts research.
Sometimes she takes part in debates intentionally but, of course, there
might be some debates she is not aware of at the time. It is, however, good
to follow closely the debates that touch upon her own research area.

3.1.2  On the Veto Power of the Sources


Studying political activity in debates and documents does not just consist
of making a plan for the study in advance and applying it to the sources
to be analysed. This is the classical design of the hypothetico-deductive
method, sketched for example in Karl Popper’s Logik der Forschung
(1934), in which he presents a hypothesis and the empirical study gives
a ‘yes or no’ answer. The hypothesis is then either falsified or not. For
Popper ‘yes’ does not, however, mean ‘verified’ but merely ‘corroborated’
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  65

for the further tests. Such a design is not applicable for sources, which are
studied with the aim to follow the moves of the agents and their acting
politically.
The sources never speak for themselves, and what is more, they need to be
taken seriously. Reinhart Koselleck for historical studies used the political
metaphor of the ’veto power of the sources’ (das Vetorecht der Quellen;
see Koselleck 1982, included in Koselleck 2010). Analogously to the veto
power of presidents, supreme courts or member states in the European
Council in certain situations, the words of the acting politicians, as they
appear in the sources to the scholar, are assumed to present a veto (in
Latin: ‘I forbid’) against her all too simple research design.
In other words, Koselleck’s metaphor of veto power of the sources—of
the political actors whose words and deeds are present in the sources—is a
strong argument against the danger of presenting results of a study, which
would be known a priori, before conducting the study. To apply a reper-
toire of analysis used in this book, if we find in a certain text or debate
aspects of polity, policy, politicisation and politicking, we still have to ask
whether they are equally relevant in all types of texts. For example, in stud-
ies on party manifestos or legislation the policy aspects might dominate,
but the polity is more or less given, whereas in constitutional debates the
dimensions of polity and politicisation play a more prominent role than
policy questions. Still, even this we cannot know for sure in advance but
must modify our analytical schemes according to the subject matter and
the sources we are analysing.
The veto power of the sources thus requires a judgement from the
scholar herself. In practice the result seldom is neither the ignorance of
the veto nor the abandonment of the original plan, but rather a revision
of the research design in a manner that the study can account for the pres-
ence of adverse voices in the sources. Research plans thus are no Soviet-
style ‘five-­years plans’ (although those presented to the funding agencies
tend to resemble them). Even if successfully funded, no plan should be
followed as faithfully as possible, but its realisation requires both impro-
visation and willingness to revise. A closer acquaintance with the sources
available for studying the research topic frequently obliges the scholars to
leave out something of the plans, to narrow down the focus or to intensify
the study of those aspects which turn out to be the main topic.
Furthermore, the political agency of the persons forming the ‘subject
matter’ of the study regularly contains important aspects that transcend
the plan. Sometimes the author does wisely to ignore them or to postpone
66  C. WIESNER ET AL.

their analysis to her next study, but in other cases she must recognise
that the plan must be revised to include them, sometimes even making
of them the main point of the story. ‘If you write as you have planned, you
are missing the point of your study’, could we formulate such a veto (on
changes of the research plan as according to the findings in the sources,
see examples below and in Chapter 4).
But the researcher should neither remain too close to what is said in
the sources. A close account of what each party said in last years’ budget
debate is, of course, an important part of political journalism. But a clas-
sification of a ‘what’, that is, the utterances, themes or topoi the scholar
‘found’ in the sources, forms only one step of research. They often explain
‘how things actually were’—‘wie es eigentlich gewesen’, as was the slogan
of the nineteenth-century German historian Leopold von Ranke. As has
been said in the beginning and is further explained in the following, when
studied with the purposes to line out the political in debates and docu-
ments, it is equally important to ask after the ‘how’ and the ‘why’ of an
utterance, a topos or a theme as well.
Taking the example of analysing a Westminster debate, it is important
that it includes a number of possibilities both within the ongoing debate
and even in the course of an ongoing speech to intervene. Making an
amendment to the motion, an adjournment of the debate or a motion of
clôture, that is, a termination of the debate, interrupts the debate. The
‘Order! Order!’ cries are addressed to direct Speaker’s attention to the
claim that the currently speaking member does not speak ‘to the question’
or, for example, uses ‘unparliamentary language’. ‘Order’ in these cases
refers to the parliamentary priority of the procedural questions over the
motions currently on the agenda.
In such situation the scholar must include also the interruptions as
important turns of the debate. However, they are not interesting as such,
but it is the political point of the different moves that matters. Moving
an amendment in the middle of the debate is in Westminster-style parlia-
ments a major political tool for opposition or backbenchers to alter the
government-supported motion, whereas moving for closing the debate is
a frequent tool in the majority’s repertoire. Raising the question of ‘unpar-
liamentary language’ concerns the political respect for the parliament and
its members. But it is neither sufficient for the scholars to know what such
moves in general might serve at, but they should also focus on this spe-
cific move at this distinct situation of debating this singular motion. Only
then they are able to offer a proper interpretation of what is the point
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  67

of such moves in the context of current debate and within the broader
parliamentary constellation. The point might be hinted at by the debaters
themselves, but must still be constructed and formulated by the scholar
in order to arrive at a judgement of what is characteristic and distinctive
of the debate.

3.2   Approaches to Linguistic Political Action:


Discourse Analysis, Rhetorical Analysis
and Conceptual History

Included among the many pitfalls of political analysis is that it is easy


to accept prejudices at face value and make conclusions based on them.
However, what has been said cannot be analysed without its proper con-
text (Skinner 2002a, 42). As aforementioned, we understand politics as a
contingent and controversial activity, of which debate is one of its major
forms, and parliamentary debate the paradigmatic style. Accordingly,
understanding the political situation and the conceptual, procedural and
institutional framework of a debate should precede discussion of the rhe-
torical tools used in the debate. Thus, the analysis of the context of what
has been said should always be part of the interpretative analysis of the
debates and texts, even if there are different approaches and sources for
studying the context.
Debates, as has been said, are to be considered as politics that is
language-­based. Political scientists have become increasingly aware of the
linguistic aspects of politics (Martin 2014, 91–92). The ‘linguistic turn’
now dates some time back, and since then the role of language in the study
of politics has been more or less generally accepted. Especially discourse
and rhetorical analysis have become popular ‘methods’ to interpret politics
(cf. e.g. Charteris-Black 2014; Atkins et al. 2014).
As has also been said, this book pays special attention to the rhetorical
character of politics (Vaarakallio and Haapala 2013a, 4). It is important
to focus on the role of rhetoric in political analysis because it is essen-
tial to pay attention to how words are used in a variety of contexts. One
should ask: What aim does an utterance, or a text, seek to achieve? What
kinds of audiences are targeted? When and under what circumstances was
the text produced? This does not, of course, mean that politics can be
completely reduced to its linguistic elements. Political action has material
and immaterial forms of expression, for both the agents and the other
68  C. WIESNER ET AL.

people it concerns. But the acts and the emotions involved are not neces-
sarily political in themselves; they must be interpreted as such, that is, con-
structed as political by means of language. This concerns even war. Carl
von Clausewitz’s famous formula on war as ‘the continuation of politics
by other means’ contains also his interpretation that war has its own gram-
mar, if not its own logic (von Clausewitz 1832, 675).
Two prominent approaches in studying the linguistic aspects of politics
are rhetorical analysis and discourse analysis. As a third useful approach
we add the one of conceptual history, which focuses on the politics and
linguistic actions related to concepts.

3.2.1  Discourse Analysis
The label discourse analysis describes a field of various approaches that
sometimes differ considerably. Some interpretations concentrate on lan-
guage, while discourse-theoretical approaches like the ones of Michel
Foucault (Foucault 1969, 1971) or Ernesto Laclau and Chantal Mouffe
(Laclau and Mouffe 1985) interpret discourses as networks of interrela-
tions between different discursive events. Most discourse-theoretical or
discourse-analytical approaches share the following ontological and meth-
odological theses:
–– Language is a social practice and is analysed as such: ‘When you say
something you are doing something’ (Johnstone 2008, 230). This
is also Quentin Skinner’s point in Meaning and Understanding
which he wrote in 1969. Language is a social practice that consti-
tutes meaning and, as according to the school of critical discourse
analysis, also dominance (Fairclough and Wodak 1997; van Dijk
2001).
–– A discourse is a setting of practices or events that constitutes
meaning and that can be distinguished according to its subject,
special institutional setting or context.
–– A discourse is also a central element for creating and circulating
distinct world views and ideologies. The base for this assumption is
the idea that the choice of words and definitions in discourse always
represent at the same time an interpretation or evaluation concern-
ing the events and practices that are the subject of the discourse.
–– Discourse does not happen by accident, but is structured accord-
ing to distinct rules that influence what can be said and which
meanings can be assigned.
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  69

–– The general aim of discourse analysis is to find out the rules of


the discourse, to distinguish which factors enable statements to be
made, and which factors enable statements to constitute meaning.
Besides these methodological assumptions, the field of discourse
a­nalysis  can vary considerably according to further methodological
­assumptions, subjects and methods (for overviews, see Johnstone 2008;
Wodak 2008). Discourse-analytical approaches can be differentiated
according to:

–– Their analytical perspectives. There are micro-, meso- and macro-­


analytical approaches. As a general tendency it can be said that the
more an approach has a linguistic background, the more micro-­
analytical it is. The poles of the continuum are built by conver-
sation analysis (CA), which is the most micro-analytical, and by
critical discourse analysis (CDA) as well as by approaches based on
the methodological ideas of Foucault, Mouffe and Laclau, which
are the most macro-oriented.
–– Their subjects: while linguistic approaches will mainly concentrate
on communication, critical and post-structuralist approaches will
mainly analyse written text.
–– Their methods: discourse-analytical approaches use different meth-
ods, which in practice mostly are qualitative.
–– Their definitions of the term ‘discourse’: the definition named ear-
lier is interpreted differently, sometimes because the authors are
not very exact, but also because they follow different method-
ological directions.
–– The role of context: some authors emphasise that aspects outside
the discourse must not be analysed (Wood and Kroger 2000, 64).
An opposing view is that discourse must be analysed in ­relation
to its context: discourse is shaped by the world outside, language,
media, previous and future discourse, and also shapes these
(Johnstone 2008, 10).

Similar to this second perspective which emphasises the decisive role of


a discourse’s context, the core argument in this book is that to understand
debate as political action, its contexts must also be taken into account. We
also support the view of discourse analysis that linguistic action is political
action, but concentrate on the notion of debate rather than discourse. This
leads to some different emphases and foci in what we look at compared
70  C. WIESNER ET AL.

to most discourse analyses. Besides the linguistic aspects and d


­ imensions,
we are interested primarily in analysing the activity of debate itself or its
procedures, practices and forms.
This does not, however, mean that a discourse-analytical approach
excludes such a perspective as ours, on the contrary. A discourse-analytical
perspective can be combined with all of the research approaches, strate-
gies and dimensions that are discussed in the following. As we emphasised
in the beginning of this chapter, we oppose the view that there are clear
schools related to the different approaches, and would argue in favour of
their commonalities: we assume that linguistic action is political action,
and speech acts are seen as moves in debates. The exemplar study that is
presented from Section 3.2.2 onwards when research steps and practices
are explained is a discourse analysis that focuses on the political actions
related to a press discourse.

3.2.2  Rhetorical Analysis
Rhetorical analysis is one approach that helps in interpreting linguistic
action as political action. While rhetoric rather refers to practices of speak-
ing or writing, rhetorical analysis means the study of a debate that uses the
rhetorical tools as interpretative categories.
Traditionally, rhetoric is related to persuasion. It is a form of knowledge
that tells us how most effectively to make others see your point of view.
The ancient Greek assumed that rhetoric belongs to public assemblies.
However, this type of oratory is just one form of persuasion today. It is our
contention that rhetoric is a form of political language. As an approach,
rhetoric is not limited to certain spaces or times only.
The long history of rhetoric underlines that it has had a major impact
on Western political culture. One can point to the institutionalisation of
the rhetorical paradigm of speaking pro et contra in parliamentary practice
(cf. Mack 2002). The procedure entails the construction of arguments for
and against, and provides an opportunity as well to discuss proposals from
different angles (Palonen 2008, 200). The invention of arguments, or the
construction of new alternatives, is a key step of classical rhetorical theory.
It is the phase in which an orator chooses the most persuasive ideas for
the occasion. The political relevance of rhetoric lies precisely in the idea of
finding alternatives.
In twentieth-century scholarship there are numerous examples of
rhetorical turns or the rediscovery of rhetoric. Famous among them is
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  71

the work of US literary scholar Kenneth Burke (1897–1993), who in A


Grammar of Motives (1945) and A Rhetoric of Motives (1950) presents
a ‘pentad’, or five-­term scheme, for classifying human actions and—fol-
lowing Giambattista Vico—a system of four master tropes—metaphor,
metonymy, synecdoche and irony. This system has also been applied by
US history theorist Hayden White to nineteenth-century history writing
in his Metahistory (1973) and to other works. It could also be applied to
other fields of study, including political science.
A major theorist of new rhetoric is the Polish-Belgian philosopher
and legal theorist Chaïm Perelman (1912–1984). In 1945, he published
Justice et raison, in which he defended the role of a rhetorical, rather than
purely logical, form of argumentation for understanding legal thought.
His main work, published together with Lucie Olbrechts-Tyteca in 1958,
is Traité de l’argumentation. La nouvelle rhétorique. This book contains a
massive repertoire of rhetorical strategies of argumentation, as opposed to
reducing rhetoric to a genre of writing. A short version of this programme
is contained in L’empire rhétorique (1977), and Perelman’s articles are
included in the collections Rhétoriques and Ethique et droit. Perelman’s
rhetoric is shaped by the priority of the forensic genre. Its central idea is
a wider concept of ‘reasonable’ as opposed to the narrowly philosophical
concept of ‘rational’.
In an Iowa conference on the rhetoric of human sciences (see Nelson
et al. 1987), the philosopher Richard Rorty launched the concept of ‘rhe-
torical turn’ to mark the new interest in rhetoric in the human sciences
(see Simons 1990). This turn was largely based on the US tradition of
teaching public speaking—connected in part to the speech competitions
between schools and universities—and it reached also into such fields as
economics and political science. Iowa political scientist John S. Nelson has
been a major proponent of a rhetorical approach in political theory (see
his Tropes of Politics 1998).
The later work of Ludwig Wittgenstein, in particular his Philosophische
Untersuchungen (published posthumously in 1953) and, inspired by it, the
linguistic philosophy of J.L. Austin, especially his How to do Things with Words
(1962), have also contributed to the recovery of rhetoric. Already in the late
1960s the Cambridge historians/political-theorists John Dunn (1968) and
Quentin Skinner (1969) began to analyse the history of political thought in
the Austinian terms of ‘speech acts’, assisted by a number of Wittgensteinian
tools. Skinner presented his insight into the historically important role of
rhetoric in political language already in The Foundations of Modern Political
72  C. WIESNER ET AL.

Thought (1978), and he systematically incorporated the speech-act terminol-


ogy with classical and Renaissance rhetoric into his studies from the 1990s
onwards, with Reason and Rhetoric in the Philosophy of Hobbes (1996), and,
most recently, Forensic Shakespeare (2014). Especially the scheme of paradi-
astole—of devaluating virtues and extenuating vices, rehabilitated by Skinner
(see below)—and the practice of speaking in utramque partem contain rhe-
torical principles useful for studying conceptual changes and debates. They
are of major importance for the present volume.
More recently, a revival of rhetorical analysis among British political
scientists has produced numerous studies, for example, on persuasion,
performance and political speeches (see e.g. Atkins et al., eds. 2014; on
rhetoric in contemporary British politics, see Finlayson 2014). Moreover,
the Rhetoric and Politics Specialist Group of the UK Political Studies
Association formed by Alan Finlayson and James Martin aims at increasing
research on debate and politics. The group brings together political sci-
ence scholars based in the UK interested in persuasion and rhetoric and is
very actively involved in fostering rhetorical analysis of politics in general.

The Rhetorical Tools: Topoi, Tropes and Figures as Arguments


The classic approach of rhetorical analysis did distinguish between differ-
ent genres of rhetorical action, namely deliberative, epideictic and forensic
rhetoric. In rhetorical terms, debate is an inherent part of the deliberative
genre of politics, of considering the strengths and weaknesses of proposals
or judgments. It differs from the epideictic rhetoric of acclamation, from
the forensic rhetoric of the plea, and from the negotiative rhetoric aiming
at agreement, characteristic of partners in diplomacy or in the labour mar-
ket, for example. The different media of rhetorical analysis, such as topoi
(places to look for concept and arguments), figures (strategies by which
the agents try to persuade the adversary) and the specific rhetorical tech-
nique of using tropes, but also other dimensions discussed below, present
useful categories of study.
The starting point for the following considerations is to understand all
use of language as rhetorical in the elementary sense that it is addressed
to some audience and can be accepted, rejected or modified by it. As
Nietzsche and others have insisted, there is no difference between figu-
rative and non-figurative speech (see Nietzsche’s lectures on rhetoric in
Volume II of the Kritische Gesamtausgabe [1995] and the contributions in
Kopperschmidt and Schanze 1994). The question is what kind of figura-
tion is a debater using as an instrument of persuasion: a highly figurative
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  73

grand style of oratory, or a seemingly matter-of-fact ‘low profile’ rhetoric,


to appeal to the audience by its impression of being detailed, concrete
or near to reality. Adversaries disputing a stand can understand that the
matter-of-fact style is as partisan as every other style.
For example the academic style of argumentation frequently relies on
appealing to ‘matter-of-factness’ (Sachlichkeit in German), just in order to
better persuade the audience to accept the arguments as unproblematic,
as ‘facts’, keeping silent on the wider perspectives on which their ‘factual’
character depends (see Weber 1904). To make this visible requires, how-
ever, illustration of these kinds of assertions from other perspectives, which
can always be constructed, although some effort is required in order to
discard commonplaces.
In certain contexts even a strict denial of driving an agenda can be used
as a rhetorical strategy for doing just that. We already mentioned how for
Carl Schmitt a highly intensive way of doing politics (Politik treiben) was
to accuse the adversary of politicking and present one’s own standpoint as
objective and impartial (1932, 21; see also Chapter 1).
Since antiquity we can speak of rhetoric as being a powerful language
of politics (in the sense of Pocock 1971), for it is conscious of its specific
conceptual and strategic instruments. According to a ‘rhetorical theory
of knowledge’, indebted to the ancient Sophists, it is always possible to
invent plausible arguments or interesting perspectives against any proposal
or analysis of a situation. Parliamentary rhetoric is, as we quoted from
James De Mille (1878) above, a rhetoric of debate (see Palonen 2016).
In parliamentary debates, agenda-setting, rules of procedure as well
as their relationship to parliamentary time correspond to the moments
of the classical rhetorical terms inventio and dispositio. In relation to
the Westminster procedure, this means that the parliamentary inventio
includes the question of what the Westminster parliament can do, that
is, the rhetoric used in parliamentary agenda-setting. The repertoire of
dispositio, presented in the current chapter, deals with the parliamentary
modes of acting politically according to the rules of debate on items on
the agenda. Indeed, to make clearer the political resources included in the
two principles, we can apply two steps. At first we can discuss the general
principles of agenda-setting and of laying down the rules of debate, then
analyse their specific connections to parliamentary time and its history in
Westminster (Palonen 2014c).
In the more literary interpretation of rhetoric, inspired by the Ramist
movement of the sixteenth century, the focus has been on the moments
74  C. WIESNER ET AL.

of elocutio and pronunciatio. The linguistical studies even of parliamentary


rhetoric tend to focus on the more technical sides of rhetoric (see articles
in Ilie, ed. 2010).
Another classical scheme for speech is the division of it into the phases
of exordium, narratio, confirmatio, refutatio and peroratio. In other
words, public speeches—in the courts, in popular assemblies and at cer-
emonial occasions—were prepared in advance in a detailed manner with
a view to their favourable reception by the audience. Either the audience
remained passive, or a rule was set whereby another speech could be held
only after the termination of the first one, such as in the succession of the
speeches of the prosecutor and the defendant in a criminal court. In such
prefabricated speeches the beginning (exordium), the presentation of the
thesis (narratio), the support for it (confirmatio) and disputation of pos-
sible objections (refutatio) as well as the artistic performance (peroratio)
could be held separately. In parliamentary debates between members, in
which the rotation of speeches pro and contra is the main feature, the
speeches are in principle more spontaneous and connected to the speeches
of the previous speakers and, therefore, contain replies and polemics. Thus
they are much more to the point and cannot follow all of the canons of
classical oratory.
With the exception of some programmatic ‘set pieces’, such as a
prime minister’s declaration of a new government programme or a
finance minister’s annual budget speech, parliamentary speeches no
longer are hour-long expositions of a position. The increasing role
of replies and spontaneous interjections from the floor has made pre-
written declarations anachronistic. The speeches are judged by strictly
political criteria, such as their persuasive power, the reputation of the
speaker or their capacity to shift the political agenda (see the ­discussions
in Palonen 2016).
The ancient and Renaissance rhetorical tradition possesses a broad array
of resources for parliamentary argumentation (see Peltonen 2013). The
Greek topos and the Latin locus originally referred to the place to search
for arguments. In modern usage the terms frequently refer to typical argu-
ments themselves, to ‘commonplaces’, which appear so self-evident and
well-known that no further arguments are thought to be needed.
Typical topoi, or the recurrent argumentative principles inherited
from the ancient rhetoric, were honestas, utilitas and necessitas. These
concepts  are not simply equivalent to their counterparts in modern
­
­languages: ­honestas refers to matters of moral or other principle, ­utilitas
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  75

to questions of political expediency, necessitas to a situation in which some


action was required rather than to the availability of a single course of
action. All of these can still be detected in contemporary debates, and
they still play a role in the nineteenth-century Oxford and Cambridge
Union Societies, together with some newer topoi (Haapala 2012; see also
Haapala, forthcoming 2017).
William Gerard Hamilton’s Parliamentary Logick, a collection of max-
ims that he had noted as used in the House of Commons from 1754
to 1796, relies on classical rhetoric, including the rhetorical uses of such
logical tools as reductio ad absurdum. The use of topoi, tropes and figures
as well as the schemes of argumentation are also a part of parliamentary
debate and have been joined by the specific tools of the ‘new rhetoric’ of
Kenneth Burke and Chaïm Perelman. Here some illustrations of the use
of each type of rhetoric will suffice (see also the examples below and in
Chapter 4).
Among the critiques of parliamentarism we can distinguish at least four
commonplaces that have been regularly repeated during the last 150 years:
the traditionalist argument against the threat to the established order,
the anti-rhetorical argument against ‘bavardage’, the populist argument
against professional politics and the Bonapartist argument against ‘for-
malism’ in favour of rapid reforms from above. To defend parliamentary
politics requires their replacement with different topoi, models of argu-
mentation that could be rendered persuasive to the audience. The power
of the words to change the world, the necessity of procedure to prevent
arbitrary power, and the importance of professional politicians in control-
ling government and administration may be cited as examples of such
topoi.
The trope is a basic rhetorical technique of arguing with figures of
speech, and there exist long lists of them (see e.g. Lausberg 1962). For
example, amplificatio refers to the extension of the range of a concept,
reductio to its limitation. Many tropes are so deeply embedded in everyday
language as dead metaphors that their figurative character is no longer
obvious to the speakers, and just, therefore, could be made explicit in the
analysis in order to analyse some of the tacit commitments.
Kenneth Burke (1945) and Hayden White (1973) among others speak
of four ‘master tropes’: metaphor, metonymy, synecdoche and irony.
To simplify matters: metaphor refers to comparison, metonymy refers
to contextual reduction, synecdoche marks a ‘representative a­necdote’
(K.  Burke) and irony inverts the normative tone. As an example of
76  C. WIESNER ET AL.

­ etaphor, an opposition politician might speak of the budget deficit as


m
an earthquake.
In this volume, we use the term ‘Westminster’ as a metonymy for the
British parliament, but it can also be seen as a synecdoche in the sense of
incarnating the parliamentary ideal type. Contemporary oppositional writ-
ers in Russia might ironise the politics of Vladimir Putin by connecting it
to the politics and the person of Rasputin, an obscure figure having lived
during the last years of the czarist regime. In a wider sense, all of these can
be seen as metaphors.
Paradiastole is, as Skinner writes, a ‘scheme’ or ‘figures of speech’ reac-
tivated by Renaissance rhetoric (see Adamson et al. 2007). It refers par-
ticularly to the rhetorical procedure for altering the normative colour of a
concept, to the devaluation of virtues and the extenuation of vices, but it
may also concern the renaming of concepts or the amplification or reduc-
tion of their range of reference (Skinner 1996, Chapter 4; 1999, 2007).
In Aristotle’s rhetoric every virtue has a definite vice as its ‘neighbour’.
The paradiastolic arguments operate with the inversion of this relationship
but not directly claiming that a vice actually should be regarded as a virtue,
or vice versa. A classical pair of naming consists of the opposition between
‘liberty’ and ‘licence’. Even today elementary principles such as ‘freedom
of speech’ are still denounced by authoritarian governments as ‘licence’,
for example by the new right-wing Polish government.
One example of a paradiastolic move is the interpretation of the accords
of the ‘third basket’ of the 1975 Conference on Security and Cooperation
in Europe (CSCE) in Helsinki. This summit provoked in Eastern Europe
the rise of ‘Helsinki groups’, which cited the principles to which their
communist governments had agreed in the conference resolution. When
these governments considered ‘the human rights’ of the third basket in a
manner similar to the lip service paid to human rights in their Soviet-style
constitutions, oppositional actors, such as Charta 77 in Czechoslovakia,
amplified the range of reference of the human rights as defined in the
Helsinki Accords. The rise of such paradiastolic extension was not planned
by the Western governments signing the resolution, but was a rhetorical
invention of the dissidents.
As an example of the argumentative schemes we can further refer to the
‘philosophical pairs’ of Perelman and Olbrechts-Tyteca (1958, 550–609).
A typical model of argumentation is to appeal to such oppositions as ‘appar-
ent vs. real’, ‘surface vs. depth’, ‘anormal vs. normal’, ‘rare vs. frequent’
and so on. The pairs are presented as if the hierarchy of the terms was
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  77

s­ elf-evident. Who would be content to move only on the ‘surface’ instead


of looking for deeper ‘essences’?
Clever rhetoricians might, however, always question, modify or even
invert the use of such pairs. Is it not open to doubt to assume that phe-
nomena always have an ‘essence’? If nobody has the authority to know
what is ‘real’, then appearance is all that matters (e.g. this argument was
used by Machiavelli). What is ‘rare’ may also be precious and, worth care-
ful study, that is, infrequent phenomena are, not necessarily secondary
in importance. When rhetoric is based on arguing in utramque partem,
nothing remains a priori beyond dispute, and the famous ‘there is no
alternative’ (TINA) of Margaret Thatcher and others challenges parlia-
mentarians to invent perspectives from which plausible and realistic alter-
natives become visible.
From a historical and comparative perspective it is also important to
regard what has been shared in a specific debate. We can speak of the
phenomenon of uncontested presuppositions among agents. The post-­
Marxist discourse analysis of Laclau and Mouffe looks into various situ-
ations for the presence of ‘hegemonies’ that remain undisputed (see e.g.
Mouffe 2013). We can, however, also speak of the limits of imagination
for the agents, in the sense of their perceived horizons of the possible or
thinkable. Such limits may become visible to outsiders, historians, anthro-
pologists or archaeologists, or to politicians involved in breaking down
the old horizons in a changing situation. Some ideas that were considered
unthinkable in Eastern Europe in 1988 had become almost common-
places two years later.

3.2.3  Concepts as Nodal Points of Debates


Sometimes less immediate objects of research may also be politically highly
interesting. One such aspect is to direct attention to the use of concepts as
political nodal points of debates.
As arguments in a debate we can also detect references to ‘basic con-
cepts’, Grundbegriffe in the sense of Reinhart Koselleck (e.g. 2006), which
are used across the political spectrum but are at the same time inherently
contested and controversial. The famous German lexicon Geschichtliche
Grundbegriffe (GG) was based on a query among historians in the 1960s,
concerning what can be regarded as the key concepts of German politics
and history since the mid-eighteenth century. Koselleck called the period
Sattelzeit, or saddle time. This was a time in which a systematic change
78  C. WIESNER ET AL.

in the character and the meaning of key concepts took place, for which
Koselleck proposed his hypotheses of temporalisation, democratisation,
politicisation and the tendency to ideologisation (Ideologisierbarkeit) (see
Koselleck 1972, 1979, 2006).
One of the key sources of the GG volumes was the multivolume
Konversationslexika, which a number of editors began publishing more
or less regularly in Germany in the eighteenth century. The various lexica
competed with each other and both revised the existing ‘definitions’ and
introduced new items, which the GG has analysed in a long-term perspec-
tive as a kind of indirect debate, based on the increasing recognition of
the controversial character of the key concepts (on the GG, see Palonen
2014b, Chapter 6 in particular). The focus on concepts can also be applied
to other types of sources that have been built up over long time periods,
including parliamentary debates from different countries (see Ihalainen
and Palonen 2009).
Freedom and democracy are in our times perhaps the most prominent
key concepts, which though disputed by a few extremists, are also fiercely
controversial among their adherents. If we speak of the parliamentary
freedom of the members of parliament—referring to free speech, the free
mandate for representatives, free elections, freedom from arrest—these
can be closely connected to the contrasting freedom of dependence on
arbitrary power, as in the neo-Roman conception of freedom (in the sense
of Skinner 1998). In this sense we can interpret the lack of women’s vot-
ing rights as an expression of dependence, as in Harriet Taylor’s classical
defence of the enfranchisement of women (1851), whereas opponents of
the extension of suffrage regarded female suffrage, above all, as an increase
of state interference in the lives of citizens, thus representing a different
concept based on freedom from interference (see Lowe 1867).
The parliamentary freedom of MPs relates to two ways to analyse the
political concepts used in parliamentary debate. One way uses sources that
are extensive and publicly documented, such as debate situations, in which
the revision, modification, and revaluation and devaluation of concepts are
an inherent part of the debate (as already Hamilton noted). The other way
searches for conceptual controversies in the debates, for example, when
the compensations and salaries of the MPs are on the agenda. Then the
concept of representation and the acceptance of the professionalisation of
politicians are tacitly presupposed in the debate, even if they might not be
mentioned in it at all (see Palonen 2012a). Debate is seldom a yes-or-no
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  79

quarrel, but rather various indirect modes of paradialistolic re-evaluation


of concepts are preferably used (see also Skinner 1996).
There are numerous other ways in which concepts can be seen as mat-
ters of political struggle in debates. Here only a few types can be taken up.
There are slogan-like concepts that in certain situations have become
politically contested. In April 1917 the Bolsheviks in Russia adopted the
old anarchist slogan, ‘All power to the soviets’, but soon after their coup
in November they suppressed both the soviets and the parliament. Still,
the slogan united much of the European left beyond the Social Democrats
for a few more years, before the subordination of the soviets to the party
and the state, after which the anarchists and left-wing communists still
defended them. Or to take a recent example: Barack Obama’s 2008 elec-
toral slogan, ‘Yes We Can’, contained a tone of optimism rare in the USA
of recent years. It was soon adopted for quite different purposes, for exam-
ple, by the Spanish Podemos party.
Slogans of this kind are difficult to debate: they may be ridiculed, paro-
died or paraphrased, but they signal a turn in the debate, a claim to re-set
the agenda (see Palonen 2006, 217–224 on the discussion, modification
and inversion of the Bismarckian slogan, ‘Politics is the art of the possi-
ble’—eine Kunst des Möglichen—in later political and academic debates). Of
course, not all slogans are successful, and it is impossible to know in advance
whether a slogan will work. For scholars, it is important to identify the
momentum of successful slogans, their persistence and their fading away.
Important for the study of parliamentary debates are quasi-technical
concepts, such as ‘bill’ or ‘parliamentary immunity’. For new MPs and for
parliamentary scholars, the first thing to learn is their lexical sense, that is,
their ‘locutionary’ meaning, but equally important is their political sense,
or the ‘illocutionary act’ included in them, to use John Austin’s (1962)
speech-act terminology.
In the parliamentary sense a ‘bill’ refers to the main instrument of leg-
islation, developed in Westminster during the late medieval period. It is
a special case of the ‘motion’, but nowadays proposing a bill requires a
legislative apparatus in which the government is superior to the individual
members. The ‘Reform Bills’ of 1832 and 1866/67 to enlarge the fran-
chise and redistribute seats are examples of bills that were not only highly
debated, but also altered the political composition and character of the
parliament as such.
80  C. WIESNER ET AL.

The demand for ‘freedom from arrest’ in Westminster and for ‘parlia-
mentary immunity’ in continental parliaments after the French Revolution
are both directed against the arbitrary powers of government. Under
the Tudor and Stuart kings, MPs could still be arrested during House
of Commons sessions (see Colclough 2005). Against such actions the
demand was raised that if a parliamentarian is accused in a court of law,
the parliament must agree to withdraw his ‘parliamentary immunity’.
From early on, some members under pressure from the courts of law got
themselves elected to the parliament not in order to practise politics, but
in order to gain protection from arrest, trusting that the parliamentary
majority would not remove their parliamentary immunity (see Hexter,
ed. 1992). Thus the parliamentary freedom of members from arbitrary
power was mixed with partisan interpretations of the immunity principle.
The debates in the Italian chambers on the parliamentary immunity of
Antonio Negri, accused of supporting terrorism, and of Silvio Berlusconi,
for whom parliamentary immunity was a major reason for starting his par-
liamentary career, provide excellent modern cases for debate analysis of a
seemingly technical principle with strong political consequences.
Another concept that refers to a general principle, ‘universal’, as in
‘universal suffrage’, has been subject to different kinds of debates, both
before its widespread acceptance and afterwards. In the nineteenth-­
century debates and in the histories of democratisation, ‘universal’ was
long associated with male suffrage, and the vote of women was regarded
as concerning only a detail in the more finely grained application of it
(see Kurunmäki 2015). Moreover, the de jure or de facto exclusion of
not only underage persons and foreigners, but also, for example, criminal
offenders from suffrage hint at debates not only about suffrage and vot-
ing, but also about citizenship, representation, democracy and political
agency in general.
Of special importance are concepts directly related to the debating pro-
cedure itself. ‘Unparliamentary language’ is not just a question of the care-
ful and polite use of language, but refers to the principle of respect for other
members and for the parliament itself as a necessary condition for debate
(see Ilie 2001, 2004). Also the rhetorical distance created in the manner
of speaking including addressing the Speaker and referring indirectly to
other members (as in ‘the right honourable member of Cambridge’) are
Westminster practices of respect, though nowadays sometimes ridiculed
(Flynn 2012). Historically, however, these were not merely rituals, but an
important part of the procedure. ‘Parliamentary’ is in such contexts not a
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  81

reference to the existing parliament, but a principle that is presupposed, if


seldom formulated, in the actual rules of parliament.
‘Fair play’ marks another important, although tacit, parliamentary
principle that can be used to judge the actual rules and conventions (see
Campion 1958; Redlich 1905). In an analysis of the procedural debates
of 1882, reference to fair play or fairness was something accepted by all,
although by some Irish members with an ironical distance to contrast
with the ideal of the English gentleman. Nonetheless, we can detect in
the use of the concept two opposed interpretations, a substantive and a
procedural. In the substantive use, some models of conduct are declared
to be fair—as in the self-declaration of ‘fair trade’ nowadays—and others
condemned accordingly. Such a use—frequently connected with the noun
‘fairness’—is clearly partisan, whereas in the procedural interpretation ‘fair
play’ is a regulative idea for judging the actual procedures, not the quality
of the person, but as a criterion to which the parties refer when criticising
or justifying a given measure. As a regulative idea fair play is a precondi-
tion for a debate open to new and unconventional ideas, a procedural and
parliamentary reinterpretation of ‘objectivity’ (on the 1882 debates, see
Palonen 2014a; on Max Weber’s recasting of ‘objectivity’ as fair play, see
Palonen 2010a). Having thus lined out approaches, practices and research
dimensions for studying debates and documents more generally, in the
following the steps of analysis, will be discussed in more detail.

3.3   Analysing Debates and Documents:


The Essential Steps
When enquiring into the essential steps of an analysis that serves to find
out what is political and what constitutes political action in a debate or a
text, it is important to underline two points: (a) there is not one single
way of doing the analysis, but there are several possible approaches that
can vary according to the research dimensions and the proceeding; and
(b) there are, however, some essential basics and standards that should be
kept. In the following, a typology of the essential steps of such a qualita-
tive and interpretative analysis will be sketched.
First, qualitative text analyses are to be distinguished from quantitative
text analyses by a major difference: quantitative text analyses are done
with the purpose to solely count frequencies of arguments or other coded
(tagged and annotated) text parts, while qualitative analyses interpret the
texts. This point requires some further elaboration.
82  C. WIESNER ET AL.

To take a caricature: the content analysis practised in the post-war


decades tended to assume that the more frequently a party manifesto, for
example, used the term ‘democracy’ or ‘democratic’, the more ‘demo-
cratic’ the party in question was. In contrast to this approach, the inter-
pretation of texts is a necessary condition for understanding the political
moves, intentions and strategies that are linked to it. To continue the
example, differences in the respective interpretations of the term democ-
racy, in the evaluation of democracy, or in regarding why the agents used
the term ‘democracy’ in their political speech acts can be made, and if they
are made, they are informative. The purpose of an interpretative qualita-
tive text analysis will be to find out more about these differences. The
distinction between quantitative and qualitative analyses, therefore, leads
to several differences in the setting and the course of the analyses, which
will be discussed in more detail below.
What has been said so far, however, does not speak against all quan-
titative elements per se. On the contrary, it is just a clarification of the
limitations of the counting method: numbers and frequencies can give us
useful information on a phenomenon that is interesting with regard to the
research question (see below), but the numbers will not tell the researcher
what the reasons, strategies and political interests related to the phenom-
enon are. To get closer to those, the material needs to be interpreted.
Second, qualitative text analyses are a research method consisting of a
systematic, theory-based and rule-based way of interpreting and under-
standing texts (Mayring 2008, 56) that is based on a clearly defined and
distinct research question. In the cases discussed in this book, ‘theory’
relates to the perspective on the concepts of debate and politics discussed
in the previous chapters and sections, and to the modes of acting politically
through speaking and writing. The goal of analysis accordingly relates to
the political dimension in the speech acts—the how and why of an argu-
ment; the reasons, strategies and interests that guide an intervention; and
also the related power struggles and the contexts.
‘Rules’ relates to the fact that the researcher will not ‘just read’ the
texts when she analyses them, but will proceed in her analysis by follow-
ing a certain order, using codes, categories or types that she will set down
explicitly (see below), and also by ordering her results so that it becomes
clear how they refer to her question.
It may be disputed here whether and to what extent qualitative research
needs to be reproducible. One might answer with a certain right that qual-
itative research can never be fully reproduced, as it will always be linked
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  83

to the researcher’s interpretation. But this does not mean the researcher
cannot or should not explain how she arrived at her results. Appropriate
means to help structure the research and to allow others to follow how the
results were obtained are codes, categories, and coding schemes, and also
protocols that explain the research steps and the intermediate findings are
helpful. If software for qualitative analyses is used, the material may be eas-
ily stored when it is coded and annotated. This stock of material can also
be given to other researchers to allow them to see how the material was
ordered and what information was deemed important for analysis.
Third, for all kinds of qualitative analyses the research material, that
is, the texts, needs to be carefully determined, collected and assembled.
Here, again, the purpose is different from quantitative analyses: the mate-
rial needs to be collected for its theoretical relevance, for its capacity to
help answer the research question. Qualitative studies hence do not follow
a logic of representativeness or average, as is often the case in quantitative
data collection. What is untypical is frequently more interesting than what
is typical, and indeed, we can only analyse the typical expressions and argu-
ments when we consider their untypical alternatives.
Fourth, after having assembled the material, that is, after material
sampling or collection, the interpretative analysis proper starts, in order
to answer the research question. At this stage, many different ways of
proceeding are possible. In this step the material is marked or notes are
taken, or those parts of the texts that are apt to help answer the research
question are extracted. The most frequent move in this phase is to work
with the material by coding it, that is, by marking and/or extracting a part
of the texts (see below). In any case, now the important details for analysis
are retrieved from the texts (the fragments, tropes, arguments that interest
the researcher). This process of extraction, analysis and interpretation can
be split into several successive steps that are all part of the working with
the material and its interpretation.
Fifth, qualitative analyses not only relate to texts as research material:
they always also relate to a kind of communication (for our purposes,
debate) from which the texts originated. The texts, therefore, are not ana-
lysed as such, but as part of a communication process or a debate, and they
are analysed precisely because the researcher wants to draw conclusions
about the communication and the contexts behind them.
Sixth, there is not one simple or single way of working with the texts.
This should happen with the purpose of answering the research question,
and according to well-reflected theoretical bases and rules, as explained
84  C. WIESNER ET AL.

above. And there are different possible outcomes of this interpretative and
analytical work, for example, the building of typologies, the assembling of
core themes, the categorisation of patterns of arguments and so on. The
outcome here essentially depends on what information is sought for in the
texts and what was of interest in the research.
In the following, this ideal-typical steps of a qualitative analysis will be
discussed in detail. Each time, an exemplar study will be used to serve in
highlighting what the respective step means in practice. The example used
here is a comparative analysis of the German and French quality press
discourse around the ratification votes on the EU Constitutional Treaty in
2005. The study was carried out by Claudia Wiesner (2014a).

3.3.1  Finding a Research Topic and Fixing the


Research Question
When a researcher analyses debates, documents or other texts, she does it
with the purpose to find answers to a distinct research question. But how
does she find her research question to begin with? Most studies do not
begin with an explicit research question. More often the research starts
with a wondering about a certain phenomenon the researcher finds inter-
esting or intriguing, or with the idea of a certain topic she would like to
enquire into, or a certain case she would like to study.
In many cases, it would even be unwise to bind oneself to a distinct
research question too early. The researcher may have an intuitive idea of
what would be interesting, politically important or provocative in relation
to the previous research, but she cannot say whether the sources can offer
the information or hints towards answering such questions until she has
taken a look at what she can find in the sources for her questions’ point of
view. A crucial point to reiterate is that the sources never ‘speak for them-
selves’; it is the scholar herself who poses the questions to the sources, and
who also must interpret what she thinks to have found there.
It may also happen that the sources give hints indicating that not the
original questions are interesting, but something different. Researchers
never know what they will find in the sources until they have started to analyse
them. For this reason, it is always advisable to start the analysis of the mate-
rial with at least a short tentative analysis of the material (or a p ­ re-­test).
The researcher might thereby find sources that refer to, for example,
political struggles that have never been noticed by scholars, but which are
nevertheless significant.
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  85

All of this means that a scholar has to understand that setting ‘the
research question’ is a multi-step process of revising one’s own question,
returning to the material, going through the steps of the analysis and
thereby developing and confirming on further concrete steps to take,
and the going through the same once or several times more. It usually is
important to base the research on the existing literature and previous find-
ings—but this knowledge should be used in fit with the research design
and support, not limit the researcher’s own perspective on the material.
The existing scholarly literature may also provoke new ideas or thought
experiments that challenge ‘the established truths’ in the field. One should
remember that surprises and new ideas are the fuel that drives analysis. It
might be said that the discovery of provocative and unconventional per-
spectives is a necessary condition of any studies worth doing. This kind of
questioning of the received wisdom is one major aspect that distinguishes
academic research from administrative questions.

3.3.2  From Research Topic to Research Question


in the Exemplar Study
In the case of the exemplar study discussed here, the initial event that trig-
gered the research interest was a brochure distributed in France inviting
people to attend an informational meeting for the ‘No’ campaign in the
referendum on the EU Constitutional Treaty in 2005. The researcher was
attending a conference in France at that time, being herself of German ori-
gin, and she was struck by the peculiar claims of the brochure: it claimed the
EU to be ‘ultraliberal’ and urged people to vote ‘No’. Such claims, and such
a campaign, were not to be imagined in Germany because, firstly, no ratifica-
tion referendum existed (in Germany the Treaty was ratified by a parliamen-
tary vote alone), but also because such EU criticism was unknown there.
The researcher, being quite familiar with France and the French political
culture, having lived and studied there for over a year, became interested
in the question because the event signalled such strong differences with
regard to the EU between the political cultures of France and Germany.
When later the ‘No’ camp triumphed in the 2005 referendum, this impres-
sion was confirmed. She decided that this case was worth a study.
But this did not mean her research question was in its final form yet, far
from it. The decisive task now was to put her initial interest (Why did the
French vote ‘No’ in the referendum? And why were things so different in
Germany?) into the form of a research design and a research project.
86  C. WIESNER ET AL.

Defining a suitable research question in this case necessitated consid-


erable work. It first required finding out more about the context of the
French referendum. There are several ways to proceed in such a case, the
usual one being reading, that is, doing a review of the literature and find-
ing out what others have had to say about the topic. If possible—and in
this case indeed it was—it is recommendable also to discuss the research
project with experts on the topic.
After thus having explored the context, one core finding was that there
had been a large public discourse before the referendum. It took place in
different kinds of media and also in public fora, and apparently had strongly
contributed to turning the general opinion on the EU Constitutional
Treaty into the negative. A second core finding was that EU criticism was
much more common in France than in Germany.
A second task was to clarify what exactly the researcher wanted to study:
it is an important point to underline that a formula, such as ‘How did the
French referendum discourse go about’, would in this case have sufficed
to define a topic or an area for research, but it did not yet constitute a
research question. So what exactly was the question that the researcher
wanted to answer?
The theoretical background and a link to earlier research proved deci-
sive here, that is, the scholarly debate on building a European identity
and an EU demos. Studying EU citizenship, the researcher had been left
with the puzzle of ‘EU identity construction’. Previous research results in
this area claimed that the development of an EU identity was a necessary
condition for the democratisation of the EU, but the results also under-
lined how complicated this construction process might be. The results also
hypothesised that public debates on the EU and EU referenda would help
constructing an EU identity. Therefore, an important finding of this aca-
demic debate, and also of the researcher’s previous work, was that public
debates on the EU could potentially have a crucial impact in helping to
construct an EU identity (see Wiesner 2007). Thus the researcher found
that her question was addressed in the scholarly literature and that an
analysis of the French EU referendum might yield valuable answers to the
identity construction process while it was underway in reality.
The research interest that arose from both this finding and the informa-
tion gathered so far was then to find out how different kinds of discourses
influenced the construction of an EU identity in different ways. This was
closer to the proper research question, but not quite there yet.
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  87

Third, for clarifying the research question and the research design
it was helpful to decide on whether to do a comparative study, and if
so, what to compare. The researcher decided to compare the French
­referendum ­discourse with the German discourse preceding the treaty rat-
ification process in Germany, which concluded also in spring 2005. This
case was different to the French in numerous respects: the Constitutional
Treaty was only ratified in parliament and not by referendum, the dis-
course was much less intensive and much less controversial. There was
some discourse however, and there was ratification, which was the com-
mon denominator between the two cases and permitted the comparative
perspective.
Defining the final research question required a number of further steps
and a going back and forth between the initial parts of the analyses, the
materials and the project, until it was finally settled: ‘How and to what
extent do national European discourses serve as a means of constructing
an EU identity?’ It was given further detail by two sub-questions: 1. ‘How
do national European discourses of political, academic and economic elites
construct the EU and Europe via the quality press?’ 2. ‘How are national
European discourses shaped by specific national contexts?’

3.3.3  Determining the Research Material


As briefly sketched in Chapter 1, depending on the research interest, texts
of any kind can be of interest if useful for carrying out the research and
answering the research question. Whether the material comes from inter-
views (which are then recorded and transcribed), parliamentary debate
(also recorded and transcribed) or policy documents, the important point
is that the material be well analysed and help the researcher answer her
question. The following list, therefore, is non-exclusive; the documents a
researcher wants to use in her research should be determined according to
her research interests and her research question:
–– Parliamentary protocols and rules of procedure, nowadays widely
available online (see Sections 4.1, 4.2 and 4.4)
–– Party programmes, flyers and brochures
–– Policy documents and papers (see Section 4.5)
–– Laws and judgements (see Section 4.5)
–– Governmental proceedings (see Section 4.5) and media articles
(see Section 4.6)
88  C. WIESNER ET AL.

–– Written or transcribed interviews (interviews differ from the mate-


rial types discussed in this book in that they involve direct discus-
sion between the researcher and the interviewed)
But how can the researcher determine which kind of material is the
ideal one for her purposes? The second step of a qualitative analysis always
consists in selecting a corpus of theoretically relevant material. Qualitative
content analysis works with texts that have been selected: (1) according
to the research question and (2) based on a systematic selection process.
The main criterion for the selection of the material is in any case not repre-
sentativeness, but theoretical relevance: Does the material selected contain
theoretically relevant information for answering the research question?
There are three classical types of material selection: (1) case contrasting
(in the tradition of the Chicago school, see Kelle and Kluge 1999, 40), (2)
theoretical sampling following Glaser and Strauss (1967) and (3) selection
plans. These types of material selections can also be used during the analy-
sis to reduce the amount of material subject to detailed analysis.
The first step in selecting the material will usually be closely linked
to the determination of the research question. It is imminently impor-
tant that, while determining what to research and what exact question
to answer, the researcher also starts reflecting upon, and trying out, the
kinds of research material that will help her answer the research question.
Already at the very beginning of this process it is important to develop
a general idea of the material that could be appropriate for the purpose.
For instance, to analyse the public discourse on the EU referendum,
several different kinds of material could have been used: booklets, pam-
phlets, and flyers that underline the political claims made in the discourse;
posters, websites, blogs or other more visual campaign material; party
protocols and information material; TV programmes; discussion proto-
cols of public meetings; newspaper articles; books on the matter; expert
interviews; and research articles. If the focus had been on the governmen-
tal and parliamentary levels rather than on the citizens and the public,
parliamentary protocols and government documents would have been of
interest, too.
Therefore, it is important to determine what exactly the research inter-
est is before selecting the material, and also to concretise this ­eventually
during material selection, as all materials are not equal in helping to answer
the question. Moreover, it is critical to know that the material in question
will contain the information sought. In this context, a tentative test analy-
sis of the material is strongly recommended.
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  89

The researcher should start sampling the material with seemingly


simple, but fundamental questions: Can I find the material? Is it gener-
ally accessible? If, for instance, only some of the TV shows needed are
accessible to the researcher (e.g. due to copyright prohibitions of the TV
networks), it might be unwise to select them as primary research mate-
rial. If the websites of a discourse have not been archived (which is very
rarely the case), it might not even be possible to do a proper analysis based
on the websites. Party protocols and materials usually are more systemati-
cally stored, but how can they be accessed? Does the researcher have to
dig into the archives herself to find what she needs? Can someone help her
with that? Or, are the documents in question perhaps stored online also,
so that she can simply download them?
Even in cases where the material is well stored, more or less complete
(to the extent that can be known at this point in the research) and easily
accessible, information retrieval can still be a problem. Can the material
be downloaded? In what data format? If the researcher uses a software
tool that is only compatible with PDF or Word format, for instance, this
might not be a major problem, though it may require considerable work
to convert the websites into the appropriate format.
The next question the researcher should ask herself is whether there
is enough material to answer the question—or whether there is rather
too much? If she finds, say, only ten texts on her topic, it may be insuf-
ficient for answering the research question properly. A discourse of sev-
eral months can hardly be analysed when there are only ten texts. At this
point, the researcher can either change her question and approach—or her
choice of material.
If, instead, she finds over 8000 texts in her material sampling, as was
the case with the exemplar study, she can be sure that it will be enough
to analyse a discourse lasting several months—but she will also know that
she will never be able to conduct a serious detailed analysis of all the texts.
Rather, she will have to find a way to limit the number of texts that are
finally analysed in detail.
The next step that should be carried out might help with this pur-
pose. The researcher should ask herself whether the material she has pre-
tested contains theoretically relevant information. Does it, in other words,
­contain the information she is looking for and that should help answer the
research question?
This step usually entails doing more intensive pre-tests in the material,
that is, more readings, drafting overviews of the material, and conducting
90  C. WIESNER ET AL.

first analyses of the texts. It might become obvious in the readings that
the material does not contain information relevant to the original research
question, but it might also contain information previously unknown to
the researcher, and that can be important to include in the research.
If a comparative analysis is planned, it is important to make sure the
materials can be compared, that is, whether they show commensurate fea-
tures at least to a certain extent. For instance, it can be more complicated
to compare very different types of protocols or media.
The last two decisive questions for the selection of the material are how
long it will take to assemble and how it can be stored (and eventually pro-
cessed). Especially in cases where the material is difficult to access (e.g. if
it is stored only in paper archives), it is important to estimate how long it
might take to assemble the necessary corpus, but this is also the case when
extensive online searches have to be carried out. Storage is a question
especially in the case of paper copies, but also with regard to electronic
documents: Shall they just be organised and stored in a computer file? Or
shall they be analysed and stored with the help of an analytical software?
In brief, all the information that has just been sketched can be found
out in the pre-test, and how long each step will take can also be tested. On
this basis it is possible to establish a realistic work plan.

3.3.4  Material Selection in the Exemplar Study


In the exemplar research design described here and after pre-testing the
material, two selection methods were used.
(1) A selection plan has been used to define the cases to be analysed as
well as the material. This was possible and useful because the researcher
(CW) had previous knowledge in the field which served at determining
cases and material that were theoretically relevant. In interpretative analy-
sis, ‘previous knowledge’ means any kind of previous knowledge, ranging
from everyday knowledge (i.e. if the researcher knows what is decisive in
the French discourse because, e.g., she is French, or present in France, and
reads the newspapers) to previous research results (e.g. most studies on a
topic agree on certain core findings) (Kelle and Kluge 1999, 14–17). Based
on the researcher’s previous knowledge, the following decisions were made.
Concerning the material, the selection was limited to the quality press,
for a number of reasons. The first reason was comparability across the cases.
As the study aimed at comparing the German and the French discourse,
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  91

first it would have been difficult to analyse very different types of m


­ aterials.
If, for instance, leaflets and brochures were analysed, those would only
have existed for the French case, as in Germany the Constitutional Treaty
was ratified by a parliamentary vote, and there was no public campaign
that preceded it. The second reason was similarity within the material cor-
pus. Visual elements, as they are contained in TV programmes, require
different analytic strategies than do texts, and would therefore have added
a different layer to the analysis at a minimum. For this reason, a text-based
corpus was determined.
This could, however, have been an argument for including different
types of texts, which had been the researcher’s (CW’s) original idea,
including a variety of daily and weekly newspapers as well as books. A
decisive bit of information that changed the selection parameters in this
respect came from French colleagues. In the beginning of the material
sampling (i.e. already after the pre-tests) and during a guest stay in France
the researcher discussed her project with numerous French colleagues
who made clear to her one important factor she previously had not been
aware of, as one needed to be in the heart of French political discourse to
realise it: a decisive split in the referendum campaign went through the
centre-left, that is, the socialist and the green parties and their adherents.
Were the researcher to focus, as originally intended, on various weekly
and daily newspapers that were more or less in the political centre, that
is, the biggest newspapers, she would not find the decisive split fully
reflected in the material. In these discussions with the French experts,
the researcher therefore received crucial information for her case and the
criteria of theoretical relevance for it: if she wanted to understand the
dynamics of the discourse, she needed to mirror the conflict lines already
in her material, and so sample the material in such a way as would reflect
these conflict lines.
Similar discussions with experts on the subject are always to be recom-
mended, especially if the researcher is not herself yet a seasoned expert in
the field, which is most often the case when dealing with the details of a
political culture that is still partly unfamiliar. Deeply rooted and unspoken
national traditions or historical conflicts are difficult to discover if one is
not an insider. But even if one feels that everything possibly important has
been considered, a discussion with experts is still to be recommended just
for confirmation. A stay in the country in question is always helpful, too,
if possible.
92  C. WIESNER ET AL.

As a consequence of the decisive insight gained with the help of the


French colleagues, the researcher decided to change her plans about the
corpus and concentrate on daily quality newspapers alone, leaving out the
books and the weekly papers. With the goal of enabling a perspective on
the full range of the discourse and including all the relevant conflicts, four
newspapers were analysed in each country case, covering the political spec-
trum from conservative-right to liberal-left and extreme-left. The newspa-
pers chosen were thus: Le Figaro, Le Monde, Libération and L´Humanité
for France; Frankfurter Allgemeine Zeitung, Süddeutsche Zeitung, taz and
Neues Deutschland for Germany.
The relevant time period that should be subject to analysis also had to
be determined. The decision here was taken in relation to crucial dates
in the ratification process. On 1 January 2005, Jacques Chirac, then
French president, had announced he would hold a referendum on the
EU Constitutional Treaty. The referendum took place on 29 May. The
German Bundestag and Bundesrat had ratified the Treaty just before, on
12 and 27 May. To include a discussion of the French referendum and its
consequences, the analysis was extended until one week after the meeting
of the European Council on 16–17 June. The period of the analysis hence
was set from 1 January to 25 June 2005.
Having thus determined the media and the time period to be studied,
the material needed to be retrieved. The researcher used web-based press
databases for these purposes. For the French case, she assembled all arti-
cles that contained the search word ‘referendum’ in the four newspapers.
For the German case, the search words had to be varied, as the constitu-
tional treaty was ratified in parliament. Here the search terms ‘EU UND
Verfassung*’ (EU AND Constitution*) and ‘EU UND Referendum’ (EU
AND Referendum) were chosen. Other possible search words were tested,
too, but the tests showed that they did not lead to different material. In
total 6373 articles were sampled for France and 2152 for Germany. After
a control of the material, part of it had to be corrected, as the search had
returned some duplicate and erroneous hits. Ultimately, 8145 relevant
articles were included in the analysis.
This was obviously too much for a detailed analysis, even for a deter-
mined researcher. The researcher therefore took some time to reflect
upon and test sampling strategies she could use to reduce the corpus,
but they all proved unsatisfactory: a random selection would have pro-
duced a random outcome, for it would have mixed relevant articles with
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  93

ones containing only marginal information, such as short notes of ten


lines. A search using topics or new search words was also unsatisfactory
because without some actual detailed insight into the discourse, it was not
­possible to know what had been the decisive topics. The same was true
for key events.
The successful strategy for dealing with the large number of articles
proved to be theoretical sampling.
(2) Theoretical sampling was used throughout the coding of the mate-
rial. In theoretical sampling, relevant units of material are determined on
the basis of the findings of the previous steps of the analysis and the cod-
ing. The selection process ends when no further theoretically relevant dif-
ferences and findings can be detected in the material (Glaser and Strauss
1967; Kelle and Kluge 1999, 44–46). This meant that the researcher first
flipped through all of the material once, which is easy with newspaper
articles as they have headlines and it can be quickly grasped what they are
about. The results of this first overview led to two further steps (see below
in detail): a sample of articles potentially relevant for detailed analysis was
assembled, and an overview on the course of the discourse was noted
down.
The researcher then proceeded on a month-per-month basis with these
pre-assembled relevant articles, reading and coding them one by one.
Thus she found out more and more, and step by step, about the themes
and arguments, and the actors and their strategies in the discourse. In the-
oretical sampling, a point of theoretical saturation is eventually reached,
namely when an argument pops up for the 37th time and the researcher
has already analysed it in three possible forms, it is not necessary to do this
again for the 37th instance.
The researcher thus was able to limit the analysis of the material to
those texts that added new information. Nevertheless, all in all 2247 arti-
cles were manually coded, using software for qualitative data analysis.

3.3.5  Analysing the Material
When analysing the material, it has been emphasised above that the
researcher needs to follow an individualised proceeding that is custom-­
designed for the needs of her research. It is important to underline once
more that there are practices and techniques that should be common to
all kinds of interpretative analyses, but there is by no means a single way of
94  C. WIESNER ET AL.

carrying them out. The following considerations therefore are mainly to


be interpreted as possible approaches and strategies, or options and ideas
that can be adapted to the needs of the respective research plan. However,
some generalisations can be made.
First, the material needs to be treated and/or prepared for analysis.
This concerns some points previously mentioned: the format of the texts
when they are retrieved, the format they need to have for analysis and
the type of material (on paper or in electronic form). The material needs
to have a form that the researcher can use for her analyses. If they are
computer-based, this means she will probably need Word, RTF or PDF
files. If she uses interviews, they need to be fully transcribed in order to be
used as a written text. Once the material is thus prepared, the researcher
can start the analysis.
One possibility, and this often is a productive one, is to ‘just start some-
where’. Whether the researcher starts with a single text or with an over-
view of all or part of the material, she will read through her text(s), jot
down her impressions and ideas, and develop first hypotheses on her mate-
rial, her categories and the findings. Such a ‘just start’ strategy is often a
good way to obtain clues about how to continue with the analysis.
In what can be described as the opposite strategy, the researcher pro-
ceeds with a predetermined plan for the analysis by defining, say, the order
of material analysis, the steps of analysis and the categories as much in
advance as possible. In reality, most researchers will combine parts of both
approaches.
Another thing that can be said is that, independent of how the
researcher proceeds, it is essential that she be open to what comes out and
can be learned from the analysis, especially to unexpected findings. It is
essential for the researcher to let herself be surprised by her findings and
learn from them.

3.3.6  Ordering, Coding and Extracting


In the social sciences, coding refers to a process of text markup. Based
on the coding system, the relevant information is marked in the text or
extracted from it. Codes can be understood as intermediate categories of
analysis that build a basis for the final results, but they do not themselves
represent the final results.
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  95

Coding systems can be either defined before coding the material or


established during the coding process, that is, develop continually in the
course of the analysis based on intermediate findings (Gläser and Laudel
2004, 41; Mayring 2008, 74). This distinction is decisive, albeit somewhat
ideal-typical.
In the first case, when a coding system is defined before analysing the
material, the researcher puts together a list of codes, or categories, that she
supposes to be of importance for the material analysis. The coding scheme
hence is based on an expectation of what will be found in the material and
is relevant for answering the research question. The expectation is usu-
ally based on previous knowledge (see above). Especially if the researcher
wants to do some more standardised analyses, she will rely strongly on her
past knowledge in setting up the coding scheme.
The other strategy is to proceed in a way that is purely inspired by
theoretical sampling, in other words, that is strictly based on the infor-
mation and patterns that the researcher finds in the material. The strong
advantage of this approach is that it guarantees the highest level of open-
ness, while the other approach runs a higher risk of being closed to new or
unexpected findings. This distinction can again best be explained using an
example from the press discourse analysed.
In the exemplar case, at first, the researcher started with a predefined
coding scheme to analyse the French referendum discourse. She had read
a lot of secondary literature on the French discourse and its context, and
she had discussed the research design with colleagues. She seemingly had
all information together to know what she was looking for and what she
could expect to find. But when she started coding, she saw that several of
her previously defined categories (and which had been presented as well-­
established findings in the literature) did not work. In other words: there
were no parts in the texts that corresponded to codes that she strongly
had expected to be significant, based on her reading of other research
findings.
The most flagrant example here was the theme of ‘le plombier polonais’,
the Polish plumber. It was a well-received and well-discussed theme out-
side France that the stereotype of a Polish plumber who came to France
and took away jobs that could have been done by the French had been
decisive in the referendum discourse. Now, the material analysed defi-
nitely showed that this was not the case, at least not based on what had
96  C. WIESNER ET AL.

happened in the press discourse. The Polish plumber barely appeared at


all, and when it did appear, it had no decisive effect on the discourse,
that is, people did not react to the theme, and it did not trigger any con-
flict. The analysis proved, on the contrary, that a number of other themes
that were decisive for the discourse had been previously unknown to the
researcher, and unmentioned in previous research on the topic. Those
new codes were added to the code system, which finally consisted of more
new codes than codes that had been previously set.
This experience underlines why it is always useful and probably essential
in interpretative analyses to include an element of deliberate openness and
the possibility to learn from the material: if the researcher does not do this,
she may well exclude unexpected findings, which often can be the most
telling. It underlines also what has been described as the ‘veto power of
the sources’ in Section 3.1.
This does not mean that a predefined category is always a bad thing.
It rather is a plea for a mixed strategy. In practice it can often be recom-
mended that both approaches be combined, that is, predefined codes are
used and completed with new codes that are derived from material analysis.
Researchers will start their analyses with a certain set of predefined codes or
categories, and add to them in the course of their analysis what they newly
learned. The only important point is that the complete coding system that
is finally developed must be consistently applied to the whole material.
In her coding of the material, the researcher proceeded as follows
(Table 3.1):

Table 3.1  Steps in the coding

Preparation of (1) Definition and sampling of the corpus


coding
Coding (2) Definition of a basic coding system (previous knowledge)
(3) First phases of coding and further development of the system
(theoretical sampling)
(4) Further selection of texts to be coded (theoretical sampling)
(5) After first phase: a fully developed code system
(6) Application of fully developed code system to all of the coded
material
Analysis (7) Interpretation and further analysis of the results

Source: Wiesner 2014a, 113


RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  97

3.4   Categories for Analysing Debates


and Documents: Explaining the ‘How’
and the ‘Why’

It has been outlined above why it is essential in an interpretative analy-


sis, especially when wanting to study a debate as political action, that the
researcher does not only search for the ‘what’ in a debate or document,
but also for the ‘how’ and the ‘why’. The ‘what’ in a very general sense can
be outlined by the analytical steps presented above. When the researcher is
done with her coding, she is more or less done with the ‘what’ questions.
She will know which utterances, topoi, and arguments are to be found in
her textual material. In most cases she will already have developed sev-
eral theses and ideas that open a way for further analysis of her findings.
Now she enters the next stage: adding further steps to the analysis which
explore interrelations of the ‘what’ findings with the interests, strategies,
actors, their interests and the contexts. This enables to explain the ‘how’
and ‘why’ behind the texts. To understand the ‘how’ and ‘why’ of a debate
and of linguistic actions, it is often helpful or even necessary to be familiar
with the context, or to include it systematically in the analysis.
In the following, some useful possible research dimensions for get-
ting behind the ‘how’ and ‘why’ will be sketched. The dimensions and
approaches that are explained help to link the ‘what’ (i.e. what is being
said) to the ‘how’ (how it is said) and the ‘why’ (why it is said).

3.4.1  
Numbers and Quantities
An approach to debates that is often helpful and simple is to use numbers
and quantities. But, as said above, the researcher needs to be aware of the
limits of such an approach: quantitative analyses cannot give the researcher
complete answers to her ‘how’ and ‘why’ questions, since the quantities of a
‘what’ do not explain a ‘why’. But changes in the quantities of words, themes
or topoi do hint at possible aspects and developments that invite further
study, even if they do not tell the researcher yet what happened and why.
Despite the limits of quantitative content analysis, thus, there is no reason to
strictly avoid counts and numbers altogether. The researcher simply needs
to be aware that numbers and frequencies only give a general idea on what
was happening in the debate and indicate phenomena worth further study.
98  C. WIESNER ET AL.

For instance, with regard to terms that are used more often than oth-
ers in a debate or a text, it has to be noted that, while the number is not
an explanation of anything in itself, the result provides an opportunity to
ask why one expression is more frequent than another. In the example
in Section 4.2, the numbers set the basis for further interpretative steps,
which are strongly based on a background knowledge of the political cul-
ture of the 1940s Great Britain.
This first example concerns an analysis of House of Commons
debates from the 1940s, and the analysis starts with overviews and
counts. These can well be carried out using the Hansard search engine,
looking for usages and terms that start with ‘polit’. Such an analysis
indicates several avenues for further study, for example, which words
and terms are used, and how often. This search for the terms actually
used shows that, besides ‘political party’, which is the most frequently
used term, such a term as ‘political life’ was also frequently used in the
1940s parliamentary debates. The analysis also indicates a number of
other terms in use.
In the discourse-analytical exemplar study described above, numbers
were also used. The articles retrieved from the database were counted on a
per-day basis, and the count showed several crucial results concerning the
course of the discourse. An overview of the key findings on the develop-
ment of the discourse, its intensity and significant events are presented in
the figure below. It was shown that:

(1) The numbers of articles per day continually increased between January
and May, and they did so in both countries.
(2) The French discourse was much more intensive, at least in terms of
output.
(3) Peaks, that is, an extraordinarily high number of articles per day, were
usually linked to key events in the discourse.
(4) In France, throughout the whole discourse, many more articles were
published than in Germany.
(5) The German peak was not linked to the Bundestag ratification vote,
but to the French referendum. This finding, along with other findings
from the discourse analysis, underlines that the French discourse was
followed in Germany.
(6) The leading function of the French discourse ends after the French
referendum: the German discourse continues after the referenda while
the French one ebbs off (Figure 3.1).
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  99

total of sampled articles per day


180
France
Germany
160

140

120

100

80

60

40

20

.0 05
.0 05

4 5

.0 05
.0 05
.0 05
.0 05
.0 05

.0 05
.0 05
.0 05

.0 05
.0 05

.0 05
05
.0 05
.0 05

.0 05
.0 05
.0 05
.0 05

.0 05

.0 05
.0 05

.0 05
.0 05
.0 00
09 .20

30 .20

11 .20
08 1.20
15 .20
22 20
29 .20

12 2.20
19 .20
26 .20

12 .20
19 .20

25 .20
20
26 .20
02 .20

07 .20
14 .20
21 .20
05 .20

05 .20

18 .20
23 .20

28 .20
04 .20
16 4.2
1.

6.
4

6
1

2
2

3
3

6
3
3

4
5
5
1

6
5
.0
01

Fig. 3.1  Number of articles per day. ​Source: Wiesner 2014

These findings suggest that the German discourse apparently followed


the French one, but they do not explain why this was the case, and how it took
shape. To learn about those points, an interpretative study of the further
dimensions indicated above was necessary. The interpretative steps that
were then carried out underlined that the French discourse clearly had
a leading function with regard to its themes and arguments, which were
taken up, or literally imported, into the German one. To put it bluntly: the
French discourse was one decisive topic of the German discourse.

3.4.2  Finding Out the ‘How’ and ‘Why’ in the Exemplar Study


For the purposes of the comparative study of the two discourses, the three
key questions on the what, how and why were concretised into: (1) What
happened in the discourse in terms of its contents? (2) How was meaning
constructed? (3) Why was meaning constructed in the way it was? A num-
ber of analytical steps were taken in relation to each question.
100  C. WIESNER ET AL.

The discourse context was systematically studied for the purpose of this
research with regard to five dimensions: (1) the political systems, (2) the
political parties and their reaction to European integration, (3) the citizens
and their views on EU integration, (4) concepts of the national identities
and (5) previous discourses on the EU. The core findings were (see in detail
Wiesner 2014a, 398–415):

(1) France’s political system (a presidential system, in which protest move-


ments traditionally have a strong role and parliament is rather weak
compared to Germany) sets other conditions for discourse than does
the German system (a parliamentary system with a rather strong parlia-
ment and a strong culture of consensus). Moreover, France held a ref-
erendum, which was not the case in Germany. This means that in
France the chances for an intensive debate were much greater, and an
occasion was created for interactions between the elite discourses and
the opinions and attitudes of the general population. And indeed, for
six months, the European Union was a central topic of public discourse
across nearly all social classes and groups. As a consequence of the lack
of a referendum in Germany, there was only little public discourse.
Moreover, this discourse rarely cut across the limits of the level of the
political and media elites. Finally, France experienced significant changes
in its political system due to Europeanisation, but Germany less so.
(2) France’s political parties are strongly influenced by European integra-
tion, or more precisely, by the fact that they have to take a stand on it.
In particular following the Maastricht debate in 1992, diverging
actors left mainstream parties and founded new, often EU-critical par-
ties or movements, while the official positions of the centre converged.
In Germany, most mainstream parties, except the Left party (in 2005
the PDS, Partei des Demokratischen Sozialismus), agreed in an elite
consensus in supporting European integration.
(3) The citizens played a different role in the discourses. In France they
voted on the Constitutional Treaty, hence they were decisively con-
cerned by the referendum discourse, and they also were actors in the
discourse. In Germany, the Constitutional Treaty was ratified in both
chambers of the parliamentary and federal system, and hence the citi-
zens intervened much less in the discourse. The citizens’ opinions
with regard to the EU, on the other hand, were rather similar in both
countries. In both there was an EU-critical potential of up to 50 %,
but only in France did this play a role in the discourse.
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  101

(4) The narratives of national identity are different, as is well known, in


particular with regard to EU integration. In Germany, EU integration
is part of the raison d´état of the new federal republic; it was a means
to become sovereign again; and it became a cornerstone of the new
narrative of national identity. The French national identity narrative,
on the other hand, is based on specific interpretations of the state, the
republic, the nation and sovereignty (unified, impartible, special) that
are rather contradictory to European integration.
(5) In France, finally, considerable groups among the leading national
politicians traditionally have a complicated or distanced position to
European integration. This also showed in previous EU discourses. In
Germany, both the relationship of political and media elites to the EU
and the previous EU discourses were much more harmonious. There
exists a broad consensus among leading politicians in favour of the
EU, German EU membership and Germany’s role in the EU (see also
Section 4.6).

All of these contextual factors influenced the discourses, sometimes


directly, when there was a reference to the context, and sometimes indi-
rectly, when for example a context setting influenced the behaviour of the
actors.
In the coding of the two discourses, the researcher used eight analytical
dimensions that she developed for the purposes of her analysis and used as
key categories (Table 3.2):

Table 3.2  Eight dimensions of a discourse

1. Course The course and development of the discourse, regarding


topics, intensity, and key events
2. Actors Key persons and institutional actors that shape the
discourse
3. Rules They structure the discourse and the possibilities for
utterances to appear
4. Levels of reference Political levels (EU, foreign countries, domestic politics)
the discourse relates to
5. Topics Topics the discourse relates to
6. Themes Typical ascriptions of meaning
7. Arguments Typical patterns of argumentation
8. Interrelations Interrelations between themes, topics, levels of reference,
rules, actors and contexts constructed in the discourse

Source: Wiesner 2014a, 116


102  C. WIESNER ET AL.

In the later phases of her research, when she was enquiring after the
‘why’ and the ‘how’ of the discourse, the researcher also proceeded
based on this typology. As explained above, she first studied the course
of the discourses through an overview of all articles and protocols. The
result was (a) an overview of the course, actors, rules, levels of reference,
topics, themes, arguments and interrelations of the discourse, and (b) a
theoretical sampling of theoretically relevant articles, based on the first
round and the protocols. Next, and based on the preselection of the
theoretically most relevant articles explained above, the discourses were
coded with regard to the eight discourse dimensions sketched above.
In the synthesis phase, the results obtained so far were synthesised and
analysed with regard to the ‘how’ and ‘why’. Based on the synthesised
findings,

(a) the research questions were answered;


(b) the results regarding the working hypotheses were summed up;
(c) the contexts were compared;
(d) the discourses were compared and
(e) further hypotheses and models were developed.

Table 3.3  Studying the what, how and why in the discourse
What happened in the How was meaning constructed? Why was meaning constructed
discourse? in the way it was?

Survey and analysis Proceeding as according to the Proceeding following the


of the course of the principles of qualitative principles of qualitative research
discourse research design: coding, survey design: coding, survey of relevant
– Overviews of relevant findings, building findings, building of types and
(protocols) of types and categories, categories, building of theories
– Key events building of theories and and models
– Intensity (article models – Which rules are shaping the
count) – Which rules of the discourses discourse?
– Actors can be distinguished? – Which interrelations with the
Survey and analysis – Which arguments are context shape the discourse?
of discourse contents dominant, and how and why – Which combinations of
– Themes do they succeed? discourse contents and which
– Arguments – Which interrelations between constructions of interrelations
– Interrelations themes, arguments and context are particularly successful, and
– Topics can be found? why?
– Levels of reference

Source: Wiesner 2014a, 116–117


RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  103

Table 3.3 presents an overview of the ways the research proceeded in try-
ing to answer the ‘what’, ‘how’ and ‘why’ questions.

3.4.3  Further Useful Dimensions for Analysing Debates


The dimensions presented above can serve as examples, or a set of tools,
for a similar study. It is however important to underline that such dimen-
sions can never be transferred directly without modification, and can never
be used as simple ‘cooking recipes’. It is of the utmost importance that the
researcher develops or at least adapts the categories of the analysis to her
own research question and material.
The dimensions of the research thus need to be defined and varied
according to the demands of the research, that is, in accordance with the
research interests and the research question. It may be that the researcher
is more interested in other dimensions than the ones sketched above, or
that she wants to focus rather on the moves in a debate that use and shape
the rules and structures. In that case, the forms and dimensions sketched
in Chapter 2 can indicate useful dimensions for studying the how and
why of a debate, that is, the political criteria, institutions and conditions
for the debate practices. The parliamentary ideal-type allows us to clas-
sify debates by different criteria described in Chapter 2, such as by topic,
parliamentary speech act or participant, as well as by rules and procedures
of debate, debates on agenda and debates of agenda, the actors in debate,
the regulators of debate, the temporality of debates, and the terminations
of debate.

3.5   Building Typologies, Drawing Conclusions


and Answering Questions

When the researcher has analysed her texts with regard to her research
dimensions in several steps of analysis, going back and forth between her
material, her questions and the intermediate findings, the ultimate goal is
to answer the research questions by drawing conclusions from the results.
Intermediate steps that will be helpful in this task include ordering the
findings, building typologies and creating systematic links to the context.
A typology in this context can refer to any kind of order that struc-
tures the findings into groups according to shared characteristics. The
researcher might build typologies of topoi, of concepts or conceptual
clusters, or of rhetorical figures, for example. The criteria that guide the
104  C. WIESNER ET AL.

building of the typologies must always relate to the characteristics that are
relevant with regard to the research interest. The building of typologies is
thus, once again, a distinct stage of the research and must be carried out
individually.
In the exemplar study, the typologies that were thus built were then
summed up in overview tables, which allowed for a compact and concise
answer to the question. The following tables show the main findings of
the discourse analysis.

3.5.1  Key Themes, Topics and Rules of the Discourses


in the Exemplar Study
The main themes of the French discourse can be separated into themes of
arguing for a ‘yes’ or for a ‘no’ in the referendum (Table 3.4).
To illustrate what the categories above mean, one brief example shall suf-
fice. The most decisive theme in the French discourse was anti-­liberalism,
which can be summarised as: ‘The EU threatens France’s welfare state, the
EU is ultra-liberal—and we have had enough of this!’ This was definitely
one of the most dominant themes of the discourse. It was coined on the
political left and influenced by the actors in the political centre and on the
right in their actions and their statements. The theme used traditional con-
cepts and themes of French political culture like services publiques, Égalité,
a strong state, protest, elite criticism and France’s special role in the world.

Table 3.4  The main themes and levels of reference in the French discourse
Main Yes-Themes (Oui) Main No-Themes (Non)

1. Internal motives of the discourse 1. Internal motives of the discourse


Reference to opposing actors Reference to opposing actors
Yes, but… Populist theme
Internal debates in the yes-camp Left alliance (Non de Gauche)
2. France´s interests 2. EU criticism
Responsibility Anti-liberal theme
Pragmatism Criticism of constitutional treaty
France´s role in the EU Sovereigntist theme
Europe puissance
3. Normative Europe

Source: Wiesner 2014a, 248–250


RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  105

Table 3.5  The main rules, topics, themes and levels of reference in the German
discourse
External to Germany Internal to Germany

1. Imported discourse 1. Main rule


France and its discourse Silencing strategy
EU-level 2. Particular German motives
Imported criticisms and reactions EU enlargement (especially Turkey)
Discussion concerning EU development and Assertions regarding EU politics
contents of the Constitutional Treaty Criticism of the EU and of the
2. Fundamental debate on the political Constitutional Treaty (Anti-Militarism)
principles of the EU after the ‘Non’ vote Classical themes of German–EU discourse
The citizenry/the demos (Western integration)
European identity New themes of support for the EU and the
What kind of Europe do we want? Treaty

Source: Wiesner 2014a, 372–373

It was used strategically by left-wing and centre-­left actors (Trotskyists,


communists, and dissident ­socialists like Laurent Fabius), and it matched
the mood of the citizenry, which marked a disenchantment with politics
and a strong feeling of social insecurity.
The main themes, topics and rules of the German discourse, on the other
hand, were rather differentiated, according to themes internal to or external
to Germany (Table 3.5).
A key rule of the German discourse was the silencing strategy (see also
Section 4.6). It was used to silence EU criticism (a) in the bigger main-
stream parties and (b) in the PDS.  The silencing strategy proceeded in
three steps:

1. EU criticism was silenced in the discourse: when criticism arose, it was


simply ignored, or it was marginalised (discussion was confined to only
small articles).
2. EU criticism and the critics were trivialised. (‘Anyway, it is impossible
to take him seriously’.)
3. The last step was to threaten the critics with sanctions. (‘Obviously, the
party may need to remind the MP about such events when the lists for
the next elections are set up’.)
106  C. WIESNER ET AL.

Table 3.6  A comparison of the French and the German discourse


Germany France
An EU discourse with a national base A national EU discourse

1. Openness 1. Closedness
Continual references to the EU and its Self-referential
member states France is ‘us’
EU and France are both described as ‘us’ EU and member states are ‘the other’
2. Intensity 2. Intensity
Two phases Very intensive
Until May scarcely any discourse High level of mobilisation: EU the most
From the end of May, very intensive important topic in April/May

Source: Wiesner 2014a, 432–434

3.5.2  The Discourses in a Comparative Perspective


The main findings of the comparison of the two discourses can be summed
up like this (Table 3.6).
The German discourse can be characterised as an EU-discourse with a
national base: its key features were shaped by its openness and continual
references to the EU and other member states, in particular to the French
discourse. Issues of the EU and its member states were constructed as
issues of European domestic politics, or as pertaining to ‘us’.
On the other hand, the German discourse was not very intensive.
Before May, it barely constituted a discourse at all, but rather several
smaller debates arose from time to time. After May, and in particular from
the middle and end of May onwards, the discourse was very intensive.
The French discourse can be characterised as a national EU discourse.
Its key features were shaped by its closedness. It was rather self-referential,
EU and EU member states were more rarely referred to than was the case
in Germany, and the EU and member states were constructed as foreign
politics, or as ‘the other’.
The French discourse, on the other hand, was very intensive. This was
due to the high level of public interest, the EU being the most important
discussion topic in April and May.

3.5.3  Computer-Supported Analysis
To conclude this chapter, some words shall be said about the usefulness
of computer-supported analysis. In a nutshell, it can be resumed that such
tools are very helpful, but that they should only be regarded as simple
RESEARCH PRACTICES AND OPERATIONS IN STUDYING DEBATES...  107

tools that have a greater or lesser degree of sophistication. They only reg-
ister, support and complement the researcher’s work, but do not replace
it. The use of software, then, is a useful tool to support the researcher’s
political literacy, but no substitute for it.
Most studies that start with a retrieval of their material in electronic
form will have used computer-support right from the beginning, by using
search engines, search words, databases or the Internet. Most research
material belongs to what is now often called the ‘digital humanities’. It is
digitally registered and stored in large, often freely accessible databases.
To use software like MaxQDA or Atlas.ti for the further steps of analysis
may be helpful especially for two reasons. First, such software is designed
for registering and supporting qualitative and interpretative analyses of
large corpora. A corpus of several hundreds or thousands of texts can be
handled and analysed easily with such specialised software. Second, the
software will help to document and register the course of the analysis, the
intermediate steps and the results. The advantage especially in the cod-
ing phase is that the software registers the coding system as the system
develops, and it stores the coded texts, offering also different possibilities
to sort out the coded parts of the texts and hence to re-manipulate them
at a later stage. In the days before these software tools were developed,
research was often carried out on paper, using pens, scissors and glue to
mark, cut out and re-organise the intermediate findings.
Most of the software in question, in addition, offers possibilities for
electronic analyses to be run. This can also be useful at certain stages of
the analysis, especially in the beginning when the researcher is interested
in an overview of the keywords or topics.
All in all, there are a number of arguments that speak in favour of using
software, but again, the limits of what it can do must be stressed. The ana-
lytical software helps to record the researcher’s own interpretative analysis,
but it does not carry out the analysis itself. In brief: the software is only as
clever as the questions that are posed by the researcher.
CHAPTER 4

Examples of Analysing Debates as Politics

In the preceding chapters some preliminary commitments regarding our


understanding of what can and should be considered as politics or political
have been made. Chapter 3 added a discussion and suggestions about how
this topic can be studied in practice and the steps one might take in such
research, with an exemplar study to illustrate the way one research project
was carried out using these steps. This chapter presents concrete examples
of research into the presence of politics or the political in actual texts and
debates. Before coming to the examples, we will discuss how they relate to
our understanding of political literacy.
One major point in our thesis on political literacy is that every issue
potentially has a political aspect, and that political literacy is required to
discover these aspects. Politics can be searched for by examining the vari-
ous forms it takes and by employing a variety of approaches—the example
cases in this chapter shall illustrate this point. First, politics is related to dif-
ferent actors, strategies, issues and media, all of which can be researched.
Second, depending on the research question and the research interest, as
outlined in Chapter 3, the researcher can choose different types of sources
to answer her question. And third, the scholarly interests and interpreta-
tive tools and strategies used in the analysis will differ from study to study,
and from researcher to researcher. The examples thus highlight how the

© The Author(s) 2017 109


C. Wiesner et al., Debates, Rhetoric and Political Action,
DOI 10.1057/978-1-137-57057-4_4
110  C. WIESNER ET AL.

analytical tools of interpretation can be used with regard to different cases


and forms of the political and different types of material, along with the
scholar’s own political literacy.
The case examples represent analyses of debate, starting with a House
of Commons debate. The first example shows three ‘readings’ of a
­parliamentary debate. Our intention is to provide three separate analy-
ses, presenting three different perspectives on one and the same debate.
Analysis like this is now possible thanks to the digitalisation projects of
parliaments, which have made parliamentary documents available for
everyone. The purpose of giving three interpretations of a debate stems
from the idea that there is not one ‘right’ way to conduct political analysis.
The different competences and backgrounds of scholars lead to different
interpretations and outcomes, even of the same texts.
Section 4.2 presents examples of a conceptual analysis of politics. It
concerns the actual use of the ‘polit-vocabulary’ among political agents—
in this case British MPs. The intention is to identify differences in the
grammar, in the rhetorical tones as well as in the mode of conceptualisa-
tion of politics, both as a sphere and as an activity. First, a set of parliamen-
tary debates is studied in a synchronic way—that is, parliamentary debates
of one and the same time period in the 1940s. Second, a long-term analy-
sis of parliamentary uses of the expression ‘politicking’ from the post-war
period is carried out. The aim is to identify ‘paradiastolic’ usages of the
term politicking, which expresses the doing of politics.
Section 4.3 shows an analysis of the use of rules of debate in debating
societies. The example is taken from the Oxford and Cambridge Union
Societies, which adopted the parliamentary procedure and terminology of
the nineteenth-century House of Commons.
The fourth example analyses a document about regulating debates. The
case is from the European Parliament. Its rules of procedure, approved
after the last European elections (2014), are studied from the perspective
of the historical transformations that the European Parliament has under-
gone. The rules are also compared to the procedural styles of the British as
well as the French parliamentary tradition and discussed from the perspec-
tive of three conceptions of parliament: the deliberative, the representative
and the legislative.
The fifth example concerns an analysis of policy documents and legal
texts as both a part and an indicator of political processes and struggles.
The case in question concerns European Union (EU) citizenship rights.
These rights were introduced and shaped largely behind closed doors
EXAMPLES OF ANALYSING DEBATES AS POLITICS  111

and in a rather top-down manner by policy documents and laws that


were drafted by the EU Commission, voted upon by the Council of the
European Union (i.e. the EU member states’ ministers),1 and interpreted
by the Court of Justice of the European Union (CJEU). The related pro-
cesses and conflicts, even if they occurred largely behind closed doors, are
by no means unpolitical. On the contrary, different interests and strategies
with regard to citizenship rights and their implementation were expressed,
and related conflicts were actively carried pursued by the agents, that is,
the EU Commission, the Council, and the CJEU, as the analysis of the
documents demonstrates.
In Section 4.6, an example of press discourse is analysed. The example
derives from the exemplar study presented already in Chapter 3, the com-
parison of the French and German press discourses on the Constitutional
Treaty in 2005. The question followed is how the rules of the German
discourse worked in order to silence certain statements in the discourse.
Finally, the football example illustrates the extension of the concept of
text to something more implicit. It can be called a ‘subtext’, where politics
is read from the game, not from the rule books alone. Football is politics
insofar as it is a controversial and contingent activity. The analysis uses
knowledge of the history of football together with studies of the changes
in football tactics.

4.1   Three Interpretations of a Parliamentary


Debate: The Case of Constitutional Renewal
in Westminster, 9 June 2009

In this chapter we want to illustrate the mode of analysing the practices of


debate with a more contemporary case, also from the House of Commons,
namely, a debate on ‘Constitutional Renewal’ initiated by Prime Minister
Gordon Brown on 9 June 2009 (http://www.publications.parliament.
uk/pa/cm200809/cmhansrd/cm090610/debtext/90610-0004.
htm#09061062000005).

4.1.1  Procedure, Concepts and Time as Topics of the Debate—


Kari Palonen
This debate is not one of the fiercest exchanges of opinions and argu-
ments between the government and the opposition(s), but rather a debate
that concerns questions central to the British political order. Or, in our
112  C. WIESNER ET AL.

language, a debate on changes of the polity rather than of policy. At the


end of debate, Peter Lilley (Con) characterised it as ‘a rag-bag of constitu-
tional reforms’ used ‘simply a form of displacement activity by the Prime
Minister’ in order to ‘divert the agenda’ (813). As we will see in Section
4.2, this also relates to the observation that it is historically Labour rather
than Conservative governments that initiate constitutional reforms.
The debate is not a classical legislative motion with a resolution to be
debated, but rather a governmental declaration of intent by the prime
minister. The Constitutional Renewal Bill was at the time of the debate
already on the agenda of both Houses of Parliament and would, according
to Prime Minister Gordon Brown, ‘come before the House of Commons
for its Second Reading very soon’ (801). For the sake of our illustration,
we shall not trace the origins of the bill, but limit our discussion to the
declaration of Mr Brown in the sitting.
The debate on the declaration does not follow the traditional House
of Commons practice in which the Speaker will ‘catch the eye’ of mem-
bers rising to speak and give the floor to them, with the aim of rotating
between speakers pro and con. The prime minister’s declaration was fol-
lowed by speeches of the opposition leaders David Cameron and Nick
Clegg. Unlike the classical rule, according to which a member can speak
on a motion only once in the plenary, the prime minister replied after
almost every intervention, his replies becoming to some extent explana-
tions and clarifications of his original declaration.
After the speeches of the ‘official opposition’ leaders the debate was
free. In contrast to the constitutional reform debates of the Attlee govern-
ment (see Section 4.2), the majority of speakers were Labour members,
who did not just applaud the prime minister or criticise the opposition
speakers, but rather introduced additional items to the debate on consti-
tutional renewal. This illustrates the presence of the second main divide
in Westminster, namely, that between the frontbenchers and backbenchers
across of parties.

 he Reform Agenda
T
The specific background for why the prime minister, at a time when the
next elections were less than a year away, still made a major declaration on
constitutional renewal was the MPs’ so-called expenses scandal, which had
raged the previous months. Gordon Brown opened his speech with the
need to respond to this scandal:

The public require, as an urgent imperative, higher standards of financial


conduct from all people in public life and an end to any abuses of the past.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  113

There is no more pressing task for this Parliament than to respond immedi-
ately to this public demand … (795). [He assured the parliament:] We must
reflect on what has happened, redress the abuses, ensure that nothing like
this can ever happen again and ensure that the public see us as individual
MPs accountable to our constituents … [Then he proposed to redress the
situation a] move from the old system of self-regulation to independent,
statutory regulation … [For the members’ finances:] The House will be
asked to agree a statutory code of conduct for all MPs, clarifying their role
in relation to their constituents and Parliament. (795, 796)

Brown also mentions that, due to the urgent need for ‘modernisation of
the procedures of the House of Commons’ (797):

I am happy to give the Government’s support to a proposal from my hon.


Friend the Chairman of the Public Administration Committee that we will
work with a special parliamentary commission comprising Members from
both sides of this House, convened for a defined period to advise on neces-
sary reforms. (Ibid.)

This ‘hon. Friend’ was the Labour backbencher and political scientist
Tony Wright, who later in a book reported on some of the reforms that
the committee, with cross-party support, managed to realise in spring
2010, just before the elections, in which Wright no longer was a candidate
for re-election (see Wright 2012, esp. 195–203, referring to his speech in
the House of Commons on 22 February 2010).
Although these motions involved a major revision of the Westminster
procedure, Brown does not regard them as constitutional renewal. Here
his agenda was focused on five other reforms:

First, we will move forward with reform of the House of Lords … Secondly,
setting out the rights that people can expect as a British citizen, but also the
responsibilities that come with those rights, is a fundamental step in bal-
ancing power between Government, Parliament and the people … Thirdly,
there is the devolution of power and the engagement of people themselves
in their local communities. (797–798)

For the fourth reform, concerning the relationship between the MP and
the constituency, he was more cautious:

We should be prepared to propose change only if there is a broad consensus


in the country that it would strengthen our democracy and our politics …
[and] fifthly, we will set out proposals for increasing public engagement in
politics. (798)
114  C. WIESNER ET AL.

This agenda set the terms for the following debate, in which opposition
leader Cameron attempted to delegitimise the reform proposals and to call
for immediate elections: ‘Is not the answer to our discredited politics, to
our disillusioned country and to our desperately weak Government a gen-
eral election?’ (799). Alternative proposals to reform came largely from
Liberal Democrats and Labour backbenchers.

 ordon Brown’s Politics of Limited Times


G
A striking feature of the prime minister’s declaration is the intense sensitiv-
ity to the politics of time. As the declaration was presented less than a year
before the five-year electoral term terminates, Brown did not use the tra-
ditional prime ministerial privilege of declaring new elections but, on the
contrary, declared an extensive programme for constitutional reform to
be realised within a relatively short time span. Still, he said he was willing
to debate the reforms: ‘The Government will look to consult widely. All
proposed reforms will be underpinned by cross-party discussions’ (798).
From an opposite point of view, this meant avoiding trying to push the
constitutional reforms through with the help of the impressive House of
Commons majority, which Tony Blair as Labour premier gained in the
2005 general elections, before being replaced by Brown in 2007.
The second temporal figure was the attempt to get rid of the expenses
scandal as soon as possible, with the implicit belief that otherwise even
more damage would be done to the reputation of parliament and the
Labour government’s way of dealing with it. As we have seen from the
quotations above, Brown speaks of urgency, of a ‘pressing task’ to which
the government has to ‘respond immediately’, even by legislating the
reforms with less than two months before the summer recess. It seems
as if he experienced the crisis as a highly personal one, as though he were
being unjustly blamed for not having done anything on the matter earlier.
Therefore, he now was responding with rapid reforms, hoping to give
the impression of a resolute government and firm parliamentary majority.
To some extent this analysis of the scandal was shared by other parties,
as Brown declared: ‘All parties have committed themselves to accept the
further recommendations of the independent Kelly committee’ (795), a
statement neither Cameron nor Clegg disputed.
The situation can, of course, be interpreted differently. The expenses
scandal looks, from a few years distance, more like a media bubble at
the margins of the established practices, which inadvertently left open
some loopholes and the potential for misuse. Brown’s comment that
EXAMPLES OF ANALYSING DEBATES AS POLITICS  115

‘the battered reputation of this institution cannot be repaired without


fundamental change’ (ibid.) marks an uncritical acceptance of the over-
dramatisation of the affair by the media.
He could well have said that the accusation that the parliament is self-­
serving and corrupt is an old anti-parliamentary topos that has been used
in Westminster at least since the Walpole era in the 1730s (see e.g. Turkka
2007). He could have also said that giving ample time to reflection and
debate, including time for parliamentary committees, is a major advantage
of parliamentary politics, and this argument could have been used against
the popular outcry, stirred up by the tabloid headlines. Other parliaments
have brought about greater transparency in members’ salaries, expenses
and extra-parliamentary incomes and revenues without switching ‘the old
system of self-regulation’ for an independent, extra-parliamentary com-
mission (for a critique of the IPSA—Independent Parliamentary Standards
Authority—commission, see Labour backbencher Paul Flynn’s book from
2012). Brown’s rhetoric of urgency can be seen as a way of getting out
of the scandal then threatening the Labour government by any means
necessary, even means that are inappropriate to the parliamentary style of
politics of the Westminster tradition.
From this perspective we can discuss the constitutional reforms and
their time frame. With his announcement of an extensive programme
of constitutional reforms for the final year of the incumbent parliament,
Brown wanted to show that Labour is able to do more than just crisis
management to get rid of the expenses scandal. At the end of his speech
he declared: ‘In the midst of all the rancour and recriminations about
expenses, let us seize the moment to lift our politics to a higher standard’
(799). By doing so he shows no regret that ‘the subject of politics itself has
become the focus of our politics’ (795), but attempts instead to use this
situation as a chance to achieve constitutional reform. The classical topos
of ‘seize the moment’ is often used in turning a crisis into an opportunity
for change. The point for Brown was that the months before the elec-
tions could now be dedicated to a major constitutional debate, with the
unstated expectation that Labour’s reputation, by updating features of the
British constitution, would be strengthened as the party of reform.
In short, Brown wants to appear in his speech as a parliamentary
politician applying time as a political resource. His politics is based on
­interpreting the immediate situation as one that requires a more or less
radical break with the past, and his government is able to master both
the short-­term ‘urgent’ reforms and combine them with the long-term
116  C. WIESNER ET AL.

­ rogramme for constitutional reform, to be realised within the limited


p
time frame before the elections.
To the opposition leader David Cameron the situation at hand looked
completely different. Its distinctive feature for him was Labour’s loss of elec-
toral support, as illustrated in a recent by-election that the Conservatives
had won. His remedy for it would be, as quoted above, to dissolve the par-
liament and hold new elections, which Brown had failed to do during his
two years as prime minister. Therefore, he regarded Brown’s entire reform
agenda with contempt: ‘He has spoken a lot about constitutional change and
innovation, but is not the real change we need not much of an innovation
at all’ (799). He refers to Brown’s unkept promises: ‘The Prime Minister
has promised constitutional change countless times before’ (800) and paro-
dies the new board planning the reform: ‘The Prime Minister’s latest brain-
child … is a National Council for Democratic Renewal. That sounds like
something out of North Korea’ (ibid.). He combines the admission ‘there is
much in the statement that we support’ with the accusation that the govern-
ment’s proposal is an imitation of an earlier proposal by the Conservatives:
‘The Government have at least mastered the art of copying things’ (799).
Cameron’s main point regarding the constitutional reform is to reject
all plans for proportional representation: ‘That it is a recipe for weak coali-
tion Governments’. Instead he proposes equal-size constituencies and
a reduction of the House membership (ibid.). Claiming that Brown’s
reform plans are an excuse to direct attention from the government’s oth-
erwise failed policies, Cameron asks: ‘Are these proposals not a pretty sorry
attempt to distract attention away from a Prime Minister who has lost his
authority, a Cabinet full of second preferences, and a Labour Government
who have led this country to the brink of bankruptcy?’ (801).
This speech makes the opposition leader appear as a self-confident
politician who is sure about his party’s victory in the next elections and
who, therefore, refuses to accept the prime minister’s reform agenda, and
even more blatantly ignores his time schedule. He seems sure that in a
country such as Britain constitutional reforms tend to be unpopular, and
he does not believe that the Labour government will be able to realise
much of them even with their House majority, and he seems well aware
that the reforms could be revoked after a Conservative electoral victory.
In his reply to Cameron, Brown hints that the Conservatives are alone
in their resistance to the constitutional reform: ‘We are not going to turn
our back on any discussion of reform; I suspect that only the Conservative
party—not the Liberal party, nor other parties—want to do that now’
EXAMPLES OF ANALYSING DEBATES AS POLITICS  117

(802). He specifies points in the reform plan, referring, for example, to


removing some governmental prerogatives in foreign policy:

The Constitutional Reform Bill removes the royal prerogative in key areas:
the ability to declare peace and war is no longer a matter just for the Prime
Minister or the Executive; it is a matter for the House of Commons, as
are the declaration of treaties, the selection of people and pre-appointment
hearings. (801)

In other words, Brown response to Cameron’s ridicule involves insisting


on the seriousness of the reforms by making their content more precise.
Liberal leader Nick Clegg shares with Cameron the same expectations
regarding a future electoral loss by Labour, and he also accuses the Labour
government trying desperately to compensate for what it failed to do when
it had time to do it:

I welcome this deathbed conversion to political reform from the man


who has blocked change at almost every opportunity for the last 12 years.
Everyone knows that the Labour party will lose the next general election, so
any reforms must be in place before the election if they are to mean anything
at all (802). He even speaks of Brown’s ‘trademark of timidity’. (803)

Unlike Cameron, Clegg pressures the government to do the reforms


before the next election:

Does the Prime Minister not see that this is no time for more committees,
more reviews and more consultation? We have been debating these issues for
decades; is it not now time to get things done? (802)

This is an appeal to another old anti-parliamentary topos, of ‘acting’ instead


of debating. This shows little appreciation for the parliamentary manner of
holding thorough deliberations in several rounds and its committee-style
focus on details and formulations. The point is to act and pass the reform, on
a broad-lined basis if need to secure the support of the parliamentary major-
ity—that is, the support of both the Labour government and the Liberals.
Clegg harshly rejects the simple-majority electoral system in particular:

I welcome any movement away from our discredited system: a system that
gives the Prime Minister’s Government untrammelled power when only one
in five people voted for them; a system that gives MPs safe seats for life. (803)
118  C. WIESNER ET AL.

Whereas Brown left the content of electoral reform open and Cameron
affirmed his categorical opposition to the proportional representation,
Clegg sees a window of opportunity for passing the reform in a propor-
tional direction with the combined forces of Labour and the Liberals,
since the former’s loss of the majority is imminent. He surely knows that
his appeal to Brown to change his mind in favour of the Liberal plan is in
vain; still, it could be used to strengthen the Liberal cause (804).
In the rest of the debate the prime minister’s politics of time was hardly
discussed. Only the Conservative Malcom Rifkind supported Cameron’s
polemics against the last-minute introduction of the large reform package.
He cast doubt on the idea of a written constitution:

Although much of our existing law is written, one thing that has made our
democracy evolve in such a vibrant and straightforward fashion has been
the conventions that have enabled change to be made without the rigidities
associated with a written constitution. (807)

In contrast to Clegg, he asks ‘the Prime Minister whether he agrees that it


would be premature for his Government to commit themselves to a writ-
ten constitution until some proper deliberation has taken place’ (ibid.). In
short, he is expecting a debate on the desirability of a written constitution
in the next parliament with a different majority.
Brown agrees that debate is needed, but declares that the present forms
have become obsolete: ‘A gentlemen’s club, operating with its own rules
and its own powers of discipline, has proven unsatisfactory and inadequate
to meet the needs of the times’ (ibid.). With the announcement of the con-
stitutional reform, and the lack of any guarantee it could be passed before the
next elections, Brown might also have alluded to reform as a change in the
time frame of the politics: whereas parliamentary supremacy at Westminster
sometimes led to a short-sighted politics, binding the parliament to a writ-
ten constitution would enable longer-term politics, extending beyond one
electoral term. This is an old social democratic topos in favour of more long-
term planning, which, for its part, has also been criticised from a republican
understanding of Britain’s (unwritten) constitution (see Tomkins 2005).
The lack of detailed debate between the party leaders is not uncommon
in this kind of situation, where the prime minister’s statement is simply a
declaration of intent without any specific motion or resolution, though it
could be analysed, amended more or less radically and submitted to a vote.
In the Westminster procedure of the twentieth century, such informal
occasions for debating matters of principle have increased (see e.g. Griffith
EXAMPLES OF ANALYSING DEBATES AS POLITICS  119

and Ryle 2003). In the present case we can see that the interpretations of
the parliamentary situation, less than a year before the elections, diverged
so radically between the prime minister and the leaders of the two opposi-
tion parties that no substantial debate was possible. Neither Cameron nor
Clegg rose again after their initial response to Brown.

 ew Items on the Agenda from the Backbenchers


N
The ensuing debate was not, however, uninteresting. Again, it did not
include detailed comments on the prime minister’s proposal to resolve the
expenses crisis or to initiate constitutional reform. The matter of debate
dealt more with suggestions for additional constitutional reform. As indi-
cated above, Labour members were more active in this debate in raising
new possibilities for constitutional reform.
The debate after the opening statements provided an occasion for back-
benchers to speak (as no ministers did) and make their special concerns
known, not merely to the House of Commons, but to the general pub-
lic. Due to the vague open-endedness of Brown’s reform agenda, Labour
backbenchers saw an opportunity to add new items perhaps more easily
realisable than the issues Brown mentioned. Passage of the backbenchers’
item could improve the position of the party and the political careers of
the members themselves. Here are two examples.
The first speaker after the party leaders was Tony Wright (Lab), who
did not, however, speak as the chair of the procedural reform committee
mentioned above. Instead he took up the issue of fixed-term parliaments:

It is an item on which I was elected in our manifesto in 1992, and I intro-


duced a Bill on it eight years ago. It is an item that most people in the House
now seem to have signed up to, and it is the proposition that we should have
fixed-term Parliaments. (804)

This has been a long-term issue in Westminster parliamentarism, debated


in the hearing of the Whittaker committee for procedure in 1914/14,
for example (House of Commons Parliamentary papers, 1914-070470-­
p1to345.pdf).
Brown, however, gives no encouragement to Wright on this question:

I know that my hon. Friend … has strong views on these issues, including
fixed-term Parliaments, which would be part of the discussions on a written
constitution. He will understand why I am making no specific announce-
ments today, and I do not propose to do so. (ibid.)
120  C. WIESNER ET AL.

Another major issue dealing with the forms of parliamentary government


was taken up by Robert N. Wareing (Lab) who spoke of the unduly grown
of powers of government over parliament:

Does the Prime Minister agree that … we have reached a situation where,
far from the Government being accountable to the House of Commons,
the House of Commons is now accountable to the Government? The pro-
gramme is determined outside the legislature, and, if we are to look at the
renewal of the constitution, perhaps we should be even more sweeping.
Perhaps we should even consider—I did not believe in this at one time—a
separation of powers between Parliament and the Executive. (807–808)

In other words, Wareing seems to challenge the main doctrine of


Westminster parliamentary government, Walter Bagehot’s famous ‘fusion
between the legislative and executive powers’, where the cabinet is the exec-
utive committee of the parliament (Bagehot 1867, 11). Brown provides a
Bagehotian answer to discard the argument: ‘My hon. Friend is proposing
the American constitution for Britain’ (808) and refers, à la Bagehot, to
the ‘deadlock’ of the US system in comparison with the greater flexibility
of the British parliamentary practice (ibid.). Here Brown’s answer seems
to miss Wareing’s point, as Wareing is referring (as did the Whittaker
Committee) to a problem unknown to Bagehot in the 1860s: the whole-
sale transfer of the parliamentary initiative to the government. Though the
exchange between Wareing and Brown ended there, there can be possibili-
ties between the fusion and the separation powers worth discussing.
More relevant to Wareing’s question is the second aspect: the strength-
ening of the independence of select committees as a counterweight to
the government’s power of initiative. Brown refers to the suggestion
of ­political science professor John Mackintosh (a Scottish Labour MP
1966–1978) for the planned reforms:

I remember the debate that John P. Mackintosh, who was a Member of this
House in the 1960s, started about the role of Select Committees and how
they could play a big role in the management of this House. (808)

Again he mentions this topic as one of Wareing’s favourite hobby horses,


but cautions that the issue might be more difficult than Wareing realises:

The Member … has great expertise in this matter to see what can be done. I
am open to these discussions, as most Members are, but we must recognise
EXAMPLES OF ANALYSING DEBATES AS POLITICS  121

the background. We have been trying to reform the Select Committee sys-
tem and to make it more relevant for 40 years. (Ibid.)

Unlike in the case of Wright, he gives Wareing at least a bit of hope that
the reform might be discussed in the current parliament.
The select committees became the most discussed topic in the sub-
sequent debate, which crossed party lines. The members realised that
they had become probably the most efficient tool for revising govern-
ment motions, and the committee chaired by Wright also pushed through
a reform in 2010 that removed the powers of party whips to nominate
members to select committees. This change was a major boost for the
backbenchers (see also Crewe 2015).

4.1.2  Analysis of a House of Commons Debate—Claudia


Wiesner
When the researcher looks into this debate, there are a few questions she
should raise at first in order to be better able to understand the debate,
what is said, and why:
–– What is the context of the debate? To which events does it
relate?
–– What is the issue in question and why is it important?
–– What have been the earlier debates, and are there later or related
debates?
–– What are the consequences of the debate: what happened after-
wards, were any decisions taken or laws proposed, or was the con-
stitution indeed changed?
To answer these questions, some additional research will be necessary,
which could in this case be carried out via an Internet research of newspa-
per and information websites, as well as by reading secondary literature.
In this way, the researcher will be able to find out what has been the
context.
Such a context research cannot be carried out here. However, the
main background information has already been given by Kari Palonen
in his exemplar analysis: the background for why the prime minister, a
year before elections, made a major declaration on constitutional renewal
involved the MPs’ ‘expenses scandal’.
122  C. WIESNER ET AL.

Research on the consequences of the debate and the decisions taken


can also be carried out via the Internet and the Hansard search engine.
Research on earlier and related debates can be carried out somewhat
more easily by using Hansard, although the search engine presents one
major problem: it does not allow a search by topics. Both these steps can-
not be illustrated here in more detail.
Having carried out such steps of context research, it is important to
define the research question that is the basis for analysing the debate. One
possible research question is: ‘What are the actors, rules, arguments and
themes that characterise the debate?’
A slight variation to the research dimensions presented in Chapter 3
would lead to the following research dimensions (Table 4.1).

 ourse, Actors and Rules of the Debate


C
With regard to the first dimension, the course of the debate, it is apparent
that it is split into two parts. The first was marked by the prime minister,
who made an appearance in the symbolic arena of parliament to present
and justify the major reform proposal he wanted to carry out. The prime
minister was followed by the leaders of the two opposition parties, the
Conservatives and the Liberal Democrats.
The exchange between the leader of the government party and the
opposition leaders is a classic in parliament and highly symbolic: each of
them takes a stand and presents arguments, as according to their role, and
as much for the sake of presenting them to the audience inside the parlia-
ment and as for the audience outside.
In such highly symbolic exchanges, everyone tries to substantiate their
claim and make their story pass. Often a crucial indicator of success is
whether the speaker gets verbal support and applause from their camp and
with what intensity.

Table 4.1  Six dimensions in the constitutional reform debate

1. Course The course and development of the debate


2. Actors Key persons, institutional actors and their strategies that shape
the discourse
3. Rules They structure the debate and the possibilities for statements to
appear
4. Topics The main topics referred to in the debate
5. Concepts The main concepts used or referred to in the debate
6. Themes Typical ascriptions of meaning
EXAMPLES OF ANALYSING DEBATES AS POLITICS  123

This first part of the debate was followed by what could be termed the
business part: the prime minister responded to questions and remarks of
backbenchers. The speaking times seem to have been much shorter, and
the content of the remarks less heated and less symbolic (see below).
These reflections directly lead to the second research dimension in the
table, the actors and their strategies. The prime minister and the opposi-
tion leaders in such a debate, as has been explained above, are assigned as
speakers because of their political weight and their symbolical function,
and the way they appear and speak is also to some extent determined by
their role. The interest of the prime minister is to gain approval for the
government’s ideas, and the interest of the opposition leaders usually is to
prove these wrong, explain why the implementation of the ideas would be
impossible, or criticise particular points in the proposal.
This is in accordance with the classical parliamentary function of the
opposition. To be critical of the government and its proposals is a dimen-
sion of government control. Opposition speakers publicly point out weak
aspects in government proposals and offer suggestion for how they can be
improved. Often this is done with polemic effect. It may happen that the
aim of the opposition speakers is less to show what is problematic in the
government proposal and much more to attack the government in gen-
eral, without addressing the proposal as such. It would be a subject for a
detail analysis to see how the opposition function has been carried out in
the case at hand here.
The second part of the debate after the exchange of the prime minister
and the opposition leaders also plays a classical role. Backbenchers speak
up and raise questions in the debate. This gives them the opportunity to
represent their constituencies’ interests and also to show the plurality of
opinions in the parliamentary group. By having several speakers, a group
can also show that there is broad expertise. The amount of support (or
lack thereof) for the proposals of the party leaders or the government can
also be displayed by the backbencher speakers. The individual interest of
the backbenchers in speaking in reply is to be present in the debate, to
have the stage, to speak, and to be seen and heard.
The third aspect, the rules, has already been discussed in part in the
considerations above. The classical roles in debates like the one discussed
here, that is, the oppositional camps and their players, work as a rule and
determine the range of what can and what must be said. The parliamen-
tary rules of speaking structure the debate, as well as the parliamentary
124  C. WIESNER ET AL.

culture. While the rules of speaking are fixed by the rules of procedure
and kept by the Speaker of the House (whom we see intervening in the
debate repeatedly), parliamentary cultures are a different matter, as they
are largely unwritten, that is, MPs learn by and by in their position how to
behave and what culture to follow.
Differences in national parliamentary cultures become immediately
apparent when watching parliamentary debates. When comparing a
debate in the House of Commons to one in the German Bundestag, the
differences are striking: while the Bundestag debates in a very calm and
ordered way, the debaters are nearly peaceful, and in any case quite polite
for the most part, the House of Commons resembles more a boxing
arena. This is partly related to the locations: the Bundestag is large, and
the interior space is enormous, while the House of Commons is small
and the government and opposition factions sit directly opposed to each
other with the MPs sitting closely side-by-side in their respective groups.
In the House of Commons, hooting and booing is a standard means of
expression in the debate, as well as the use of certain established expres-
sions, the atmosphere is often heated, and the debate may resemble a
fight.

 oncepts, Themes and Arguments in the Debate


C
The expense scandal debate was fully coded (see below) in a proceeding
that was entirely inductive, that is, no codes were fixed beforehand, but all
of them were drawn from the material itself. It turned out that the main
topics regarded, unsurprisingly, questions on polity organisation and con-
stitutional reform, and the measures to be taken in this respect.
The main concepts used or referred to, as well as the main themes
used, can be grouped into six categories. The first category involved the
structure and constitution of the polity, that is, the British political sys-
tem, and the reform measures to be taken. The second category relates
to a classical feature of debate: the attack against the political opponent.
The third category relates to concepts of democracy. The fourth concerns
the responsibility of MPs and the role of parliament. The fifth relates to
the prime minister’s claims of leadership and displays of his willingness
to act, including appeals to future-oriented measures. The sixth category
includes proposals made by the backbenchers. Most are very specific and
could have been included in the code ‘measures’, except for the fact that
most of the measures suggested by the backbenchers were rejected in the
prime minister’s response.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  125

Table 4.2  Overview on


Code 197
the codings in the consti-
tutional reform debate Backbencher proposals 42
Polity 11
Attack 30
Sovereignty of parliament 9
Leadership 9
Future 9
Willingness to act 5
Measures 23
MPs’ mission 7
Responsibility 26
Democracy 26

The Table 4.2 shows: (1) the codes that were developed and (2)
the number of text passages categorised under the respective coding.
These numbers, as underlined in Chapter 3, are not to be understood as
­quantitative proofs of anything. As has been explained, the codes as well
as the codings (i.e. the coded parts of the text) cannot be regarded as
results in themselves; they simply help to structure the findings. The num-
bers can also be read as indicators for the themes and concepts that were
used the most frequently in the debate. An interpretative analysis of the
speech should use the codings to build categories or types of arguments
and themes.
The codings in the analysis of the debate underline an impression that
the researcher already noted in her first reading of the debate. The first
part, the debate between the prime minister and the opposition leaders, is
marked by the symbolical exchange between the two leaders. Brown tries
to be neutral and quasi-presidential in his arguing style, while facing the
opposition leaders’ attacks. This first part of the debate is hence marked by
the usage of appeals, normatively loaded concepts and themes, and future-
oriented arguments.
Gordon Brown in his entry speech uses a mixture of appeals to the
MPs’ sense of responsibility, their mission, and their duty with regard to
safeguarding democracy:

There is no more pressing task for this Parliament than to respond immedi-
ately to this public demand … Like every Member here, I believe that most
Members of Parliament enter public life so that they can serve the public inter-
est … But all of us have to have the humility to accept that public confidence
126  C. WIESNER ET AL.

has been shaken and that the battered reputation of this ­institution cannot be
repaired without fundamental change … It will be what we now do, not just
what we say, that will prove that we have learned and that we have changed
… I believe also that the vast majority of MPs work hard for their constituents
and demonstrate by their service, whatever party they belong to, that they are
in politics not for what they can get but for what they can give. (795)

Brown also presents measures that have already been taken or that are
planned:

So first, all MPs’ past and future expenses should and will be published on
the internet in the next few days. Home claims submitted by MPs from all
parts of the House over the past four years must, as we have agreed, be scru-
tinised by an independently led panel … First, we propose that the House
of Commons—and subsequently the House of Lords—move from the old
system of self-regulation to independent, statutory regulation. That will
mean the immediate creation of a new Parliamentary Standards Authority,
which will have delegated power to regulate the system of allowances. The
proposed new authority would take over the role of the Fees Office in
authorising Members’ claims, oversee the new allowance system, follow-
ing proposals from the Committee on Standards in Public Life, maintain
the Register of Members’ Interests, and disallow claims, require repayment
and apply firm and appropriate sanctions in cases of financial irregularity …
Secondly, the House will be asked to agree a statutory code of conduct for
all MPs, clarifying their role in relation to their constituents and Parliament,
detailing what the electorate can expect and the consequences that will fol-
low for those who fail to deliver. It will codify much more clearly the differ-
ent potential offences that must be addressed, and the options available to
sanction. These measures will be included in a short, self-standing Bill on
the conduct of Members in the Commons, which will be introduced and
debated before the summer Adjournment. This will address the most imme-
diate issues about which we know the public are most upset, but it will be
only the first stage of our legislation on the constitution. (795–796)

Last, but not least in importance, Brown emphasises his willingness to


act as well as his leadership position in the country, appealing to a better
future:

At its first meeting yesterday, the Government’s democratic council decided


to bring forward new legislative proposals before the summer … As we come
forward with proposals, in each case the Government will look to consult
widely. All proposed reforms will be underpinned by cross-party discussions.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  127

Our proposals will also be informed by leading external figures, including


academics and others who command public respect and have a recognised
interest or expertise in the different elements of democratic reform … In the
midst of all the recriminations, let us seize the moment to lift our politics to
a higher standard. (796,798–799)

When David Cameron, then the Tory opposition leader, takes over,
he immediately attacks Brown and the Labour government. In com-
parison, it is apparent that there were no such attacks by Brown against
the opposition (and accordingly, no attacks were coded for Brown’s first
speech).

I thank the Prime Minister for his statement, but I have to say that he read it
so quickly that I am not sure he convinced even himself … The Government
have at least mastered the art of copying things … The problem is that the
Prime Minister has promised constitutional change countless times before.
He promised it when he launched his campaign for leadership of the Labour
party two years ago, he promised it in his first statement to the House as
Prime Minister, and he promised it in his speech on liberty in the autumn
of 2007. He promised it in two Queen’s Speeches, when he specifically
pledged the delivery of a Constitutional Renewal Bill. That Bill was first
mooted in November 2007, then again in December 2008. Why has it
taken so long? (799).

Moreover, Cameron mixes appeals to democracy (in a rather populist vein)


with a critique of the government:

How many more excuses will they come up with before they recognise that
it is time for people to have their say?
Let me turn now to the proposals themselves. The country is too cen-
tralised, Parliament is too weak, and the Government are too top-down, too
secretive and too unwilling to give up power.
Above all, is not the real problem the fact that people feel shut out of
decision making and unable to control the things that matter to them? (799)

Even where Cameron expresses support for government proposals, he


emphasises his critique:

There is much in the statement that we support, not least because it is taken
from the comprehensive case for reform that I made to the Open University.
The Government have at least mastered the art of copying things. We agree
128  C. WIESNER ET AL.

with giving more power to local government, but let us not stop there. Why
not abolish the regional quangos that have taken so much power away from
local government? (799)

In sum, thus, Cameron’s speech consists mainly of a series of attacks,


mixed with appeals for a better democracy (or what he understands as
such).
In his immediate reply to Cameron, Brown for the first time that day
also switches into attack mode, and as did Cameron, he mixes attacks with
appeals to what he understands as democracy, emphases of his responsi-
bility, explanations of the measures he is going to take, and themes that
underline his leadership and his willingness to act.

All MPs sitting in this Chamber must know what their constituents are tell-
ing them, but I do not think the Leader of the Opposition’s statement
reflects what they are saying to him. Nearly 70 per cent of people did not
vote in the elections of last week, and 40 per cent voted for parties other
than those represented here, and we have to accept that people want us first
to clean up our politics quickly. I would have thought he would make more
mention of that in his response to the statement … We want all-party sup-
port for the new code of conduct and the parliamentary standards commis-
sioner, and also for the way we deal with issues of exclusion from the House
of Commons, which is something that has not been faced up to before. The
precondition of any debate about the future of our democracy must be our
determination to recognise, with humility, that this House has got to clean
up its affairs as a matter of urgency, and every Member of Parliament shares
a responsibility for doing that. (801)

Nick Clegg, the leader of the Liberal Democrats, starts his speech with a
series of attacks as well:

I thank the Prime Minister for his statement. Of course everyone agrees
that the political crisis requires big changes in the way we do things, so I
welcome this deathbed conversion to political reform from the man who
has blocked change at almost every opportunity for the last 12  years …
Everyone knows that the Labour party will lose the next general election, so
any reforms must be in place before the election if they are to mean anything
at all … Does the Prime Minister not see that this is no time for more com-
mittees, more reviews and more consultation? We have been debating these
issues for decades; is it not now time to get things done? (802)
EXAMPLES OF ANALYSING DEBATES AS POLITICS  129

Similar to Cameron, he then goes on mixing attacks with appeals to


democracy as he understands it.
Interestingly, Brown’s reply to Clegg begins with expression of sup-
port. It is a long statement that underlines his claim to leadership:

First, let me say where we agree. We agree—I am glad the right hon.
Gentleman has said this explicitly—that we will all support the new
Parliamentary Standards Authority; we will move from self-regulation
to statutory regulation. We can therefore do that very quickly; it can be
enacted very quickly to start almost immediately. We will all agree to the
code of conduct, which means that the conditions under which MPs may be
excluded from the House of Commons will be set down for them. We will
modernise the means by which we deal with those issues where exclusion
or recall is a possibility; I think there should be a debate on that over the
next few weeks. We have got to make sure first of all that the country sees us
dealing with the changes that are necessary, and I think that the mood of the
House today still does not sufficiently recognise the gravity of the problems
we face with our constituents and that we have got to deal with as a matter
of immediacy. (803)

With regard to the backbencher’s questions, it is apparent that there are


two types. First, there are polite questions and remarks that get back to
concrete, technical proposals and ideas for reform that the MPs of all
the parties have worked on. They are answered more or less politely,
or sometimes also negligently, by Gordon Brown, who in his answers
also ­concentrates on the point in question. The following example comes
from a Labour backbencher who should hence be supportive of Brown,
and he is:
Dr Tony Wright (Cannock Chase) (Lab):

My right hon. Friend has made an important announcement today that


may even turn out to be historic. Will he consider adding one item to the
list? It is an item on which I was elected in our manifesto in 1992, and I
introduced a Bill on it eight years ago. It is an item that most people in the
House now seem to have signed up to, and it is the proposition that we
should have fixed-term Parliaments. May I ask him to signal his commit-
ment to fixed-­term Parliaments by announcing now the date of the next
general election? May I suggest that 6 May 2010 would be an excellent
date? (804)
130  C. WIESNER ET AL.

The Prime Minister:

I know that my hon. Friend, whom we are asking to undertake new respon-
sibilities in this review, has strong views on these issues, including fixed-term
Parliaments, which would be part of the discussions on a written consti-
tution. He will understand why I am making no specific announcements
today, and I do not propose to do so. It is more important to get on with
the work that we have set out, both in cleaning up the politics of this coun-
try and in making the reforms that he, to his credit, has been proposing for
years in this House. (804)

Second, there are comments by prominent backbenchers who have played


a decisive role in their parties. Their remarks are much more in the form
of an attack against the prime minister and the Labour government, and
they are also taken as such by the prime minister:
Sir Malcolm Rifkind (Kensington and Chelsea) (Con):

May I, in the kindest possible way, suggest that today’s statement is a rag-­
bag of ill-considered proposals brought forward in the last 11 months of this
Parliament by an exhausted Prime Minister. That is no more apparent than
in regard to his proposals for a written constitution. He knows that although
much of our existing law is written, one thing that has made our democracy
evolve in such a vibrant and straightforward fashion has been the conven-
tions that have enabled change to be made without the rigidities associated
with a written constitution. May I ask the Prime Minister whether he agrees
that it would be premature for his Government to commit themselves to a
written constitution until some proper deliberation has taken place—not
just on what a written constitution might say, but on whether it is desirable
in the first place? (807)

In his reply, Brown once again uses the big themes and concepts, empha-
sises democracy and a future-orientation:
The Prime Minister:

The problems of this House of Commons and the workings of our political
system being ad hoc and evolutionary have been revealed in the expenses
scandal in the House of Commons. It is absolutely clear that a gentlemen’s
club, operating with its own rules and its own powers of discipline, has
proven unsatisfactory and inadequate to meet the needs of the times. I
believe that there are other areas in our constitution whereby our inability
EXAMPLES OF ANALYSING DEBATES AS POLITICS  131

to be straight about what we are trying to do and to put that down in legis-
lation means that we sometimes fail the public. I said that there is a debate
to be had about a written constitution, and I know that the right hon. and
learned Gentleman took my words carefully. There is a debate to be had on
it. It is a major decision for our country, and he is clearly against it. Given
that so much of our constitution is now written for the different parts of
the United Kingdom, for different areas of policy and for the relationship
between individuals and the state, it is worth considering putting that into
one written constitution. (807)

Further Perspectives
After these first steps, the analysis could be continued in a number of
directions. For example:
–– The following debates on the issue of constitutional
reform  could  be  studied, possibly establishing a comparative
perspective.
–– The interests and the backgrounds of the actors could be exam-
ined more closely, to better understand the themes and arguments
used and the strategies behind.
–– A study could focus in detail on what followed from Brown’s pro-
posals, which ones were realised or abandoned and when.
–– The argumentive strategies used by the different speakers could be
analysed in more detail and compared.
As underlined above, the research question to be answered is the main
determining factor in how the researcher will continue.

4.1.3  A House of Commons Debate Analysis—Taru Haapala


My analysis pays closer attention to the use of rhetoric in the House of
Commons debate. The rhetorical analysis involves focusing on the use of
language aimed at persuading the parliamentary audience. I will discuss,
for example, the use of the rhetorical genres in this context. To come
back to what was explained in Chapter 3, the classical genres of rhetoric
comprise of the epideictic, the forensic and the deliberative. The epideictic
genre means the kind of persuasion that aims at finding approval or disap-
proval for an action in the present. Forensic rhetoric refers to persuasion
that describes or re-describes past events. The deliberative genre is action
to persuade in relation to the future, which often means efforts to turn the
course of events to one’s own advantage.
132  C. WIESNER ET AL.

I will also deepen the rhetorical analysis by pointing out some topoi
that were used in the debate. Topoi, or rhetorical commonplaces, are
expressions that form a certain argumentative pattern and are used to
add persuasive power to the claims made in debate. In taking a closer
look at the kinds of persuasive means used, I will highlight key con-
cepts or turns of phrase that I call topoi. Through these argumentative
elements, the individual moves, or attempts to intervene or direct the
course of the debate, will be more easily identified. These are important
in establishing what kind of political activity was practised in the case
study.
The analysis will also show the politics of agenda-setting that Gordon
Brown, as prime minister, and his supporters and opponents conducted
during the debate. The aim is not to determine their effectiveness, as,
firstly, that it is not the point of a political-activity analysis, which seeks to
find out the ways in which politics is conducted and is not interested in
the outcomes as such. Secondly, effectiveness as a research object would
require much broader material than this. Finally, I will also discuss the
contrast between parliamentary sovereignty and popular sovereignty,
between appeals to the people and referenda.

 hetorical Genres and Commonplace Expressions


R
In his statement the prime minister used both forensic and deliberative
rhetoric when he declared: ‘The past few months have shown us that
the public require, as an urgent imperative, higher standards of financial
conduct from all people in public life and an end to any abuses of the
past. There is no more pressing task for this Parliament than to respond
immediately to this public demand’ (House of Commons, 10 June 2009,
col. 795). At the same time, Gordon Brown was making a statement that
included an interpretation of the past and that tried to persuade the audi-
ence to accept a course of action based on his judgement of what was
lying ahead in the future. He appeals to the necessity of creating ‘higher
standards’ of public life, not only for correcting things done in the past,
but also urgently for future purposes.
The rhetorical commonplace of necessity can be defined in terms of a
topos of expediency that has been part of the British parliamentary culture
of debate at least since the mid-nineteenth century (cf. Haapala, forth-
coming 2017, esp. Chapter 4). It was used as an argument in the 1830s,
for example, when members of the House of Lords demanded a proce-
dure reform for the lower chamber.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  133

In 1837 Lord Brougham argued that it was necessary to revise the


procedure of the House of Commons to expedite legislation. The rhetori-
cal commonplace of necessity was used to add more argumentative force
to the suggestion of making an inquiry into the business of the House of
Commons. He suggested that the House of Lords conduct inquiries into
the restriction of private business in the lower chamber, as ‘an inquiry
would shew the necessity of some legislative provision to expedite the
business’ (House of Lords, 5 June 1837, cols. 1181–1182). This can be
compared to Gordon Brown’s speech, where the topos was used at a time
when Parliament was under pressure to change its procedures with the
general feeling that some reform of the legislative system was desperately
needed. Like Brown, Brougham also blamed ‘the system’ rather than indi-
vidual parties or members of parliament: ‘It was the system, and the sys-
tem alone, of which he complained, and which by long usage had grown
up into an inveterate practice’ (ibid., 1176).
In Brown’s speech, the epideictic genre of rhetoric (shown as an
effort to find support for action in the present) was used as he praised
the morale of the House of Commons members for serving ‘the public
interest’: ‘I believe also that the vast majority of MPs work hard for
their constituents and demonstrate by their service, whatever party they
belong to, that they are in politics not for what they can get but for
what they can give’ (House of Commons, 10 June 2009, col. 795).
After acknowledging the good intentions of MPs, he suggested that it
was time to ‘accept that public confidence has been shaken and that
the battered reputation of this institution cannot be repaired without
fundamental change’ (ibid.). The quote contains two additional rhetori-
cal commonplaces, the topoi of the vote of confidence and of character
(see Haapala, forthcoming 2017, Section 4.2). These indicate the use of
deliberative rhetoric.
Brown argues that immediate action is needed to restore public confi-
dence, which has been badly shaken. In this manner, he shows that there
is a problem that has to be resolved regarding political representation.
The topos of the vote of confidence is related to the representative aspect
of parliamentary politics (Haapala, forthcoming 2017, esp. Section 4.2.4).
Proposing that representation is not working as it should is a way of
demanding action on House of Commons practices. Brown further argues
that the public take such a low view of parliament that the reputation of
the institution is at stake. He, therefore, also uses a topos of character that
refers to the ethos of parliament.
134  C. WIESNER ET AL.

‘Character’ is closely connected to ‘vote of confidence’ as both con-


cern the relationship between parliament and what is expected of the
institution on the outside. They point to the question of the legitimacy
of parliamentary politics and the power to act on behalf of the elector-
ate: ‘Each of us has a part to play in the hard task of regaining the coun-
try’s trust, not for the sake of our different parties but for the sake of
our common democracy. Without that trust, there can be no legitimacy;
and without legitimacy, none of us can do the job our constituents
have sent us here to do’ (Brown: House of Commons, 10 June 2009,
col. 795). In other words, Brown’s message is that there is a problem
related to parliamentary sovereignty itself, and this urgently needs to be
amended.
Another element of deliberative rhetoric in use in Brown’s speech is the
topos of principle. This classical topos of honestas has been used for centuries
as a counterpart of utilitas. All Roman rhetoricians agreed that delibera-
tive speeches aimed at honestas or utilitas (Peltonen 2013, 70). ‘Principle’
is equivalent to honestas, the classic topos of morality as opposed to the
political ‘expediency’, or utilitas. The use of these ancient topoi becomes
apparent, for example, when Brown remarks: ‘This is not a modern and
accountable system that puts the interests of constituents first. It needs to
change’ (House of Commons, 10 June 2009, col. 796). ‘Principle’ refers
to what is desirable and of value and ‘expediency’ to the interests and ben-
efits. While the former points to what ought to be, or what is considered
to be good or bad, the latter to what would be convenient in the circum-
stances. Here Brown emphasises the value of ‘a modern and accountable
system’ and the need for the present system to change in order to benefit
the constituents.

 olitics of Agenda-Setting in the Debate


P
Setting the agenda refers to rhetorical invention, the act of formulat-
ing expressions and of determining their order of appearance. Prime
Minister Gordon Brown used his prerogative power to set the agenda
of the debate by laying out an impressive array of arguments for reform.
Based on one debate, it is not possible to determine the entire context
of the discussion. What can be done, however, is analyse some of the
reactions in the parliamentary audience. Were there new formulations of
Brown’s proposals? Or were they simply set aside for some other item on
the agenda?
EXAMPLES OF ANALYSING DEBATES AS POLITICS  135

Leader of the opposition David Cameron took the first turn to speak
after Brown. He did not reply to Brown’s statement as such, but took the
opportunity to mock the prime minister:

I thank the Prime Minister for his statement, but I have to say that he read
it so quickly that I am not sure he convinced even himself. He has spoken a
lot about constitutional change and innovation, but is not the real change we
need not much of an innovation at all? Is not the answer to our discredited
politics, to our disillusioned country and to our desperately weak Government
a general election? (House of Commons, 10 June 2009, col. 799)

Cameron dismissed Brown’s statement and argued that the simplest and
surest answer to the country’s political crisis was to call a general election.
By doing so he turned the attention away from the substance of Brown’s
statement and was able to introduce onto the agenda an alternative course
of action. In fact, Cameron claimed that Brown’s proposals had been
taken directly from his own ‘comprehensive case for reform’, which he had
made to the Open University and which gave him the chance to congratu-
late the Government for mastering ‘the art of copying things’ (House of
Commons, 10 June 2009, col. 799).
Cameron’s strategy was to use forensic rhetoric to show the weaknesses
of Brown’s government.

Let us consider the things that the Prime Minister has proposed in the past.
He has proposed a British day, an institute of Britishness, a Bill of Rights,
a written constitution and reform of the House of Lords. They are all end-
lessly launched and relaunched, but nothing ever happens. It is not so much
a Government strategy as a relaunch distraction strategy, intended to give
the Prime Minister something to talk about when he is in desperate straits.
(House of Commons, 10 June 2009, col. 800)

He claimed that what the prime minister was proposing was just a way to
camouflage the government’s failure to execute its policies. But he also
continued to use deliberative rhetoric when he directed his ridicule at the
prime minister’s latest innovation:

The Prime Minister’s latest brainchild—you could not make this one up—is
a National Council for Democratic Renewal. That sounds like something out
of North Korea, but let us be clear about what it is. It is not some outward-­
looking convention that is open to the public. It is not even cross-party. It
136  C. WIESNER ET AL.

is just a bunch of Ministers talking to themselves. (House of Commons, 10


June 2009, col. 800)

Cameron portrayed the prime minister’s project as a joke played on the


principle of democracy, a feeble attempt to make an elite discussion group
seem credible. By mocking the prime minister’s efforts he was able to cast
them in an unfavourable light and make them look ridiculous.
He then turned to use honestas when he suggested: ‘At the heart of
any programme of constitutional reform should be proper taxpayer trans-
parency. Is it not time to publish all public spending, national and local,
online, so that each taxpayer can see precisely how his or her hard-earned
money is being spent?’ (House of Commons, 10 June 2009, col. 800).
In this way Cameron redefined what was the desirable and valuable aim
of constitutional reform. His own aim, however, was to depict Brown’s
reform proposals as ancient and ineffectual, thus counterfeiting utilitas:

Are these proposals not a pretty sorry attempt to distract attention away
from a Prime Minister who has lost his authority, a Cabinet full of second
preferences, and a Labour Government who have led this country to the
brink of bankruptcy? (House of Commons, 10 June 2009, col. 801)

Cameron’s message was that Brown’s proposals should be dismissed as


failed attempts to save the reputation of the government. He clearly was not
willing to discuss the reform but drove the discussion in the direction of his
own political agenda. Brown replied by discrediting Cameron’s arguments:

All MPs sitting in this Chamber must know what their constituents are tell-
ing them, but I do not think the Leader of the Opposition’s statement
reflects what they are saying to him. Nearly 70 per cent of people did not
vote in the elections of last week, and 40 per cent voted for parties other
than those represented here, and we have to accept that people want us first
to clean up our politics quickly. I would have thought he would make more
mention of that in his response to the statement. (House of Commons, 10
June 2009, col. 801)

He made clear that Cameron’s allegations of ‘what people want’ were


based on a rather weak turnout in the elections. He then tried to reset the
agenda as he had originally intended:

The precondition of any debate about the future of our democracy must be
our determination to recognise, with humility, that this House has got to
EXAMPLES OF ANALYSING DEBATES AS POLITICS  137

clean up its affairs as a matter of urgency, and every Member of Parliament


shares a responsibility for doing that. (House of Commons, 10 June 2009,
col. 801)

Brown reiterated his argument, which was based on both honestas and
utilitas. While he emphasised the need to establish a new code of conduct
that could get the support of all political parties (utilitas), he also referred
to the value of responsibility (honestas).
Both Cameron and Brown used the topos of principle in trying to per-
suade their audience of what was the right and desirable course of action.
Whereas Brown used it to emphasise responsibility and reconciliation,
Cameron concentrated on the discontent with and the failures of the pres-
ent government. The prime minister was unable to appeal to reason and
goodwill as the leader of the opposition portrayed him as insincere and
meaning something else than what he was actually saying.

 arliamentary versus Popular Sovereignty


P
As Cameron tried to shift the attention away from Brown’s reform pro-
posals, he was simultaneously reaching out to the electorate: ‘Our view
is clear: we should not take away from the British people the right to get
rid of weak, tired and discredited Governments’ (House of Commons,
10 June 2009, col. 800). By appealing to the British people, he was try-
ing to show that the parliamentary sovereignty Brown was attempting to
save by reform was not working because of the dysfunctionality of the
government.
Cameron’s argument can be contrasted with the views of the Liberal
Democrat leader, Nick Clegg, whose claim was that only his party was on
the people’s side:

We cannot afford to wait for a cross-party consensus because the


Conservatives will never want to change this cosy Westminster stitch-up. We
do not need to wait for the Cabinet to make up its mind; it is not up to it
to decide how our democracy works. People should now be given a say, so
will the Prime Minister now call a referendum this autumn to give people
a choice—a choice between the bankrupt system we have now and serious
proposals for reform which finally put the people in charge, not politicians?
(House of Commons, 10 June 2009, col. 803)

Clegg was appealing to popular sovereignty instead of parliamentary sov-


ereignty. His claim was that a referendum would give the people the power
138  C. WIESNER ET AL.

to make decisions about the reform. His message was that ‘the people’ are
not represented at Westminster at all, which also basically undermines the
idea of parliamentary sovereignty.
In Brown’s reply, Clegg’s proposal was dismissed as a way of making
politicians look like they did not work hard enough:

I must ask the leader of the Liberal Democrats not to perpetuate the myth
that for 12 or 14 weeks during the summer and autumn MPs do absolutely
nothing. MPs are in their constituencies, and working there, and let us not
perpetuate a myth that is not the correct story. Most MPs I know are work-
ing very hard indeed in their constituencies. That deserves to be said so
that people understand that that is the case. (House of Commons, 10 June
2009, col. 804)

While denouncing Clegg’s argument as ‘a myth’, Brown did not really


confront the issue of parliamentary sovereignty. In this manner he let the
actual problem remain untouched.
As Brown did not address the issue and defend parliamentary sover-
eignty, the matter was picked up by Labour backbencher Robert Wareing
(Liverpool West Derby), who asked the prime minister:

Does the Prime Minister agree that the authority and power of Parliament
have been diminishing for decades under successive Governments and that
in fact—programme motions have been mentioned in this context—we have
reached a situation where, far from the Government being accountable to
the House of Commons, the House of Commons is now accountable to the
Government? The programme is determined outside the legislature, and, if
we are to look at the renewal of the constitution, perhaps we should be even
more sweeping. Perhaps we should even consider—I did not believe in this
at one time—a separation of powers between Parliament and the Executive.
(House of Commons, 10 June 2009, cols. 807–808)

Another Labour backbencher, Geraldine Smith (Morecambe and


Lunesdale), also suggested a referendum on European Union membership:

One democratic reform that I feel that the British public would like to see is
a referendum on the principle of whether we remain in the European Union.
No one under the age of 50 has had a chance to vote on the question of
Europe and I think that, as a matter of course, we should have a referendum
once every 15 to 20 years. (House of Commons, 10 June 2009, cols. 809)
EXAMPLES OF ANALYSING DEBATES AS POLITICS  139

The backbenchers’ suggestions clearly harmed Brown’s efforts here as


they directed attention away from his attempts to reaffirm parliamentary
sovereignty.
Cameron’s intention of creating discontent was proving fruitful. He
was also supported by his own party members, who did not miss any
opportunity to create further distrust against the prime minister:

In the discussions that I have had, many backbenchers have agreed that
one of the biggest problems is that the Executive have become part of the
problem rather than part of the solution. Is not the point today that the
Prime Minister has come forward with a statement in which the Executive
tell the rest of us what we shall now reform? Should not the rest of us be tell-
ing the Executive what they should or should not do? (Iain Duncan Smith
[Chingford and Woodford Green] [Con]; House of Commons, 10 June
2009, col. 809)

The argument here was that the prime minister did not represent the
views of the majority of the House of Commons at all. This suggestion
prompted Brown to react strongly, and he was forced to defend his posi-
tion as the prime minister:

First, I am talking about powers that the Executive are surrendering to


Parliament or powers that Parliament should have. I put forward proposals
two years ago for a whole range of areas, such as pre-confirmation hearings,
where the Executive should surrender some of their powers to Parliament. If
the right hon. Gentleman is going to perpetuate the myth that somehow the
problems of the past few weeks are not the problems of Parliament and are
not problems that people consider to have been caused by mistakes made
by Members of Parliament and this House, I do not think that he will get
an echo in the country for what he is saying. (House of Commons, 10 June
2009, col. 809)

However, his response did not satisfy the backbenchers on either side of
the floor. The Labour party backbenchers kept on calling for a referen-
dum on staying in the European Union. The Conservatives reiterated
Cameron’s argument that the Government was not addressing the more
pressing issue of the economic downturn:

Is the Prime Minister aware that what really matters to our constituents
at the moment and what fuels their anger over parliamentary allowances
140  C. WIESNER ET AL.

is the state of the economy and their fear for their jobs and livelihoods?
They see the attempt to divert the agenda to consideration of a rag-bag
of constitutional reforms as simply a form of displacement activity by the
Prime Minister. Is he aware of the definition of ‘displacement activity’?
It is defined as follows: ‘A pattern of behaviour believed to be a means’—by
which animals relieve—‘tension resulting from two contradicting instincts’.
The definition continues by stating that this activity ‘often involves actions
such as scratching, excessive grooming or chasing one’s own tail’. (Peter
Lilley [Hitchin and Harpenden] [Con]; House of Commons, 10 June
2009, col. 813)

It seems that the Conservatives were making good work of discrediting


the government and sowing distrust among Labour backbenchers. The
situation left Brown on the defensive:

I cannot really understand the statements that are now coming from
Conservative backbenchers. We must face up to the expenses issue; the right
hon. Gentleman seems to suggest that we do not need to face up to that
issue, but we do. That is not displacement activity; it is essential activity in
order to restore the reputation of politics. I happen to agree with him about
the economy, but the action that we have taken is to move the economy as
quickly as possible through the downturn. The Leader of the Opposition
fails to ask any questions about the economy at any time we meet. (House
of Commons, 10 June 2009, col. 813)

Brown was finally cornered on the economic downturn. Although he had


the reputation of being strong on financial issues, the way he defended
himself against accusations of ‘displacement activity’ did not support that
view. Brown’s reform policy now looked beside the point.
The attacks on the prime minister in this debate were first fuelled by
Cameron’s suggestion that parliamentary sovereignty was not being served
by the incumbent government. Then Clegg raised the idea of popular
sovereignty and remarked that the government should give the people a
‘choice’. Labour backbenchers were also eager to demand for referenda as
a means of staging a rebellion against the government.

In Conclusion
The analysis shows that the House of Commons debate has different lay-
ers of persuasion, including the genres of epideictic, forensic and delibera-
tive rhetoric. Also identified was the use of rhetorical commonplaces that
have been recognised and used since ancient times. These topoi allowed
EXAMPLES OF ANALYSING DEBATES AS POLITICS  141

me to find certain patterns of argumentation and show how the partici-


pants attempted to add more force to their individual speeches.
With the analysis of rhetorical genres and topoi I was able to find what
could be called ‘moves’ in a debate. These were shown in particular in the
politics of agenda-setting and the argumentation regarding parliamentary
versus popular sovereignty, which included both proponents and oppo-
nents of Brown’s proposal. In terms of political agenda-setting, the leader
of the opposition, David Cameron, used forensic rhetoric to show the
failures of the Brown government and made mockery of Brown’s pro-
posal. By using the topos of principle, Cameron also offered a redefinition
of what should constitute a desirable reform. According to him, what the
country needed was a general election to get rid of the government. He
claimed that Brown’s proposal was yet another failed attempt to save the
reputation of the government. By challenging Brown’s proposal and mak-
ing his own, he made his ‘move’ in the debate. While defending his side,
Brown also used the topos of principle to reiterate his argument about the
desirability of a new code of conduct for MPs. Brown protested his sincer-
ity and humility before the parliamentary audience, but Cameron, for his
part, was now directing his speech to the electorate.
The debate also showed argumentation on the issues of parliamentary
versus popular sovereignty. This turn in the debate was made possible by
Cameron’s successful changing of the agenda. He appealed to the British
people and, by doing so, was also portraying himself as someone who had
the authority to do so. He was suggesting that the political mandate that
is given to the government by the electorate would no longer be valid
unless Brown called for new elections. In contrast to restoring parliamen-
tary sovereignty, Nick Clegg took the opportunity to appeal to popular
sovereignty. He suggested a referendum ‘to give people a choice’ between
‘the bankrupt system’ and ‘serious proposals’ that would put ‘the people
in charge, not the politicians’. Clegg’s proposal was against the very idea
of a parliamentary sovereignty that operates through the political man-
date given to politicians through elections. Brown did not take a clear
stand and defend parliamentary sovereignty, which gave the Labour back-
benchers the opportunity to question his leadership. Brown’s position as
the Prime Minister was attacked from both sides of the floor. What made
him lose authority was that he failed to take a stand on the accusations
launched against him. When Cameron and the backbenchers started to
question the state of parliamentary sovereignty and address the electorate
outside of parliament, he kept on trying to convince the parliamentary
audience, but he had already lost his authority.
142  C. WIESNER ET AL.

4.2   Conceptual Controversies in Parliament—


‘Politics’ in the House of Commons
This section illustrates the central conceptual cluster of our book, namely,
controversies over ‘politics’ presented in Chapter 2. The first example (see
4.2.1) is an analysis of uses of concepts as historical and controversial enti-
ties, which are not necessarily problematicised explicitly in the debate, but
which still mark central dividing lines. Reinhart Koselleck (1996) called
such concepts ‘pivots’, around which the disputes turn. The second case
study (4.2.2) focuses on one of the four aspects of politics, namely the use
of ‘politicking’.
Our analytical scheme for the use of politics in specific debates is
sketched out in Chapter 1 as a typology of four aspects of politics. In this
section it is first developed further for the purpose of using it in the analy-
sis of the parliamentary debates.
The debates analysed here, as in the Section 4.1, are those of the British
parliament, as available in the Hansard debates from 1803 to 2005. The
first case study focuses on a thematic concept, the constitutional debates
of the post-war era under the Labour government of Clement Attlee, and
includes an analysis of the different polit-words, with the main focus on
the epithets for the adjective ‘political’, offering a sample of the spectrum
of the conceptual horizons of what is ‘political’ in that period and in that
type of debate. Whereas we use the concept of politicking in a purely
formal manner in the scheme of Chapter 1, the most common usage in
the British everyday language has been pejorative, and this holds also for
the debates among the MPs. The period studied in that section encom-
passes the period from 1945 to 2005, as documented in the Hansard. The
analyses are focused on the nuances and deviations from the pejorative
usage of politicking, on the willingness of MPs to adopt it to their usage,
and whereby politicking is shown to express something important for the
understanding of politics.

4.2.1  The Four Aspects of Politics in Parliamentary Debate


One way to explicate the disagreements in the understanding of politics
is to distinguish different aspects of how and in which respects some-
thing appears as or can be read as political. The English adjective ‘political’
relates to three nouns: ‘politics’, ‘policy’ and ‘polity’. This approximation
allows to discuss different types of studies. This can be linked, as we have
EXAMPLES OF ANALYSING DEBATES AS POLITICS  143

done in the first chapter, to a discussion of contingency, the omnipresent


possibility to act otherwise, as a necessary condition of politics as an activ-
ity. By making further distinctions within ‘politics’, ‘politicisation’, which
is contingency-creating, and ‘politicking’, which is contingency-applying,
we arrive at a simple typology of the types of questions within the activity
of politics (Palonen 2003) and which are the operational concepts we use
in this volume.
Policy debates concern the direction of politics, the resolutions of
meetings or legislation in a parliament, as well as the policies advanced,
supported, criticised or rejected by a government, a party or even an
individual politician. Policy refers to a programme, plan, strategy or
line that gives to the individual moves in a debate or measures in a
campaign a normative and future-oriented direction. It is commonly
assumed that it is better to have some policy rather than none at all. In
some contexts, however, an ‘opportunistic’ or ad hoc style of politicking
might be wiser than to commit to a definite policy that sharply restricts
the available alternatives.
Polity debates concern the rules and practices of proceeding within
an association, organisation or institution, in particular according to the
parliamentary procedures or the constitutions of the state or of supra-
national polities. Polity debates contain interpretations of or amendments
to the rules and whether actual proposals are in accordance with the rules.
Polities can be regarded as the historical results of past politicisations that
have ­established themselves as widely accepted and that place obstacles in
the way of new politicisations.
Politicisation relates to the marking of something as contingent or
controversial, as something where there is time and space to play with
possible alternative actions. Politicisation debates concern the activity
of presenting some contingent, controversial or contested dimension as
political. They can open up or put into question previous interpretations
of what is political, but they can also set new types of issues as political
on the debating agenda of the political agents. Claims of politicisation
specify new chances of politicking in some definite respect. Controversies
over parliamentary agenda-setting are good examples of debates over
what should be politicised due to a matter’s content and/or its impor-
tance or urgency.
Politicking refers to actions that use the time and space of the con-
tingency created by politicisations. Politicking debates concern the man-
ners, directions, qualifications and competences involved in using existing
144  C. WIESNER ET AL.

resources that have been already politicised. They may entail different
styles of politicking as well as different degrees of mastery of the styles,
comparable to the profiles and qualifications of agents in the political
game, which are always dependent on the situation. In this sense we can
see policy as a form of politics that relies on a line or programme, or one
that refers to one style of politicking among others, such as a politicking
that would limit ‘opportunism’ in the name of normative goals and judge-
ments whether the (allegedly) opportunistic line is realistic or not.
This section will suggest a number of exemplar analytical steps to
apply this distinction. The first examples refer to the actual uses of polit-­
vocabulary in debates among politicians. The polit-vocabulary items will
be interpreted in terms of the different understandings of the styles used
in setting the conceptual horizons of politics that are available for the
actors. The examples have been taken from the Hansard documentation
of debates in the British parliament.

4.2.2  ‘Political’ in British Constitutional Debates


from the Late 1940s
As an example of the ways to analyse the actual uses of polit-vocabulary
in debate, six plenary debates at Westminster on parliamentary reforms in
post-war Britain will be analysed. The Labour government led by Clement
Attlee arrived in power in the summer of 1945 with a major political pro-
gramme, mainly on social and economic issues. It also made proposals for
parliamentary and electoral reforms pertaining to the status and composi-
tion of the House of Lords as well as the elimination of plural voting by
abolishing the university constituencies and the City of London seat. In
the era that democratised suffrage and the parliament terms these were
regarded as British curiosities.
The six debates from 1947 to 1949 are as follows:

1. PARLIAMENT BILL, second reading, House of Commons, 10



November 1947
2. REPRESENTATION OF THE PEOPLE BILL, second reading,

House of Commons, 16 February 1948
3. CLAUSE I (Constituencies and electors), House of Commons, 16
March 1948
4. CLAUSE I (Constituencies and electors), an amendment, House of
Commons, 17 March 1948
EXAMPLES OF ANALYSING DEBATES AS POLITICS  145

5. REPRESENTATION OF THE PEOPLE BILL, third reading, House


of Commons, 23 June 1948
6. PARLIAMENT BILL, second reading, House of Commons, 31

October 1949

We shall not enter into the content of these debates in more detail,
nor discuss their antecedents (e.g. the Parliamentary papers) or go into
the question of what role they played in the confrontation between the
Labour government and the Conservative opposition, led by Winston
Churchill. Shall we discuss neither the role of the Liberals or other minor
parties, nor the divide, important for British politics, between frontbench-
ers and backbenchers, a divide that cut across party lines. The focus is
exclusively on the uses of the polit-vocabulary in these debates.
The general settings of the debate are interesting in that the content of
the Labour government’s motions shook some long-established features
of British parliament and politics in general. The university constituencies,
the privileges of the City of London and the character and existence of
the House of Lords appear to present-day non-British scholars as anach-
ronistic remnants, which prevent rather than support the original features
of the Westminster parliament from being fully realised. Nonetheless, it
is quite obvious that such reforms provoked fierce opposition not only
from the Conservative minority, but also from the representatives whose
seats were targeted for elimination, such as the MPs of the university con-
stituencies. In this situation, it is understandable that the vast majority
of the speakers in the debates were opponents of the reform—at least in
the forms that the Labour government’s motions presented them—and
the profile of the vocabulary accordingly became biased in favour of the
oppositional ‘discourse’.
One of the reasons to choose this set of examples is that it deals with
polity questions, while the vast majority of items on the parliamentary
agenda concern policy questions. Here the issues of parliament, represen-
tation, electoral systems, voting and citizenship and their relationship to
‘politics’ are at stake. Accordingly, we can expect a high number of uses
of the polit-vocabulary to be included to the debates, and the Hansard
word count confirms that this has indeed been the case. (When looking
for polit-words ‘policy’ has not been included, but in the British usage it
hardly serves as a distinct politics-related concept, unless one considers
the qualification of ‘political policy’, of which one expression was found
in the debates.)
146  C. WIESNER ET AL.

The first aim of the analysis consists of constructing a sort of ‘grammar


of the British polit-vocabulary’ of that period. At this stage the various
cognates of the words were simply enumerated and classified.
The entire corpus contains 263 polit-words (one ‘polite’ was excluded,
although it, too, is historically linked). Of these the adjective ‘political’
(including ‘non-political’, the adverb ‘politically’ and the compound
‘body politic’) were the clear majority of the cases, namely 219. The
British thematisation of the concept takes place in terms of adjectival
qualifications, in which respect it clearly differs from the German debates
(for which the noun Politik has been the main concept). Accordingly, the
British or Westminster polit-grammar can be built up around the cognates
and qualifications of ‘political’, complemented with those for ‘politics’ and
‘politician’.
It is no surprise that the most common expression in the debates is
‘political party’ or ‘political parties’, for which there are 38 mentions in
the six debates. There were 5 mentions of the compound noun ‘party
politics’. In the post-war era both in Britain and in continental Europe,
we can speak of a metonymical quasi-identification of ‘politics’ with ‘party
politics‘ and of ‘political’ with ‘partisan’, or perhaps slightly less strongly,
‘committed to a party’ (cf. Palonen 2006, 187, 189, 212).
In the language of the press such identification was probably even
stronger than in parliament, and the same holds for the usage among
party activists. In other words, membership and activity in a party as well
as the power of the party organisation in running parliamentary candi-
dates were almost without competition. The model of the socialist mass
parties, in which the party leadership decided the politics of the par-
liamentarians (see Kautsky 1911), had grown stronger also among the
non-socialist parties. Although this trend probably was weaker in Britain
than elsewhere, even there was the tendency to identify oneself with the
party rather than with strengths of the parliament as an institution, even
if parties were also criticised by those who opposed the planned Labour
reforms.
Conceptually, the party paradigm of politics can be given at least two
interpretations. The one emphasises the role of parties as quasi-natural
units of political organisation, working at all levels. The other stresses
rather the partisan commitment of the individuals: to participate in politics
requires committing oneself (in the existential sense of ‘daring’) to take
stand, even when all alternatives are unappealing. The first interpretation
contains a weaker commitment, which, however, will hardly be questioned
EXAMPLES OF ANALYSING DEBATES AS POLITICS  147

once it has been made, whereas the latter means a stronger commitment,
which is, however, liable to be revised. Most of the expressions in the
debate refer to the first, in the form of everyday partisanship, despised in
some measure by not only the intellectual, but also the traditional elites in
the Conservative and Liberal parties.
Much less obvious is the term that ranked number two in terms of
frequency, namely ‘political life’. This is an expression without a clear and
established content, maybe a British specialty. Again, it could be given at
least two interpretations. In the one sense, ‘political life’ is that part of
politics that goes beyond the formal institutions, activates citizens and
supports the institutions at the same time. Walter Bagehot writes in The
English Constitution: ‘A good Parliament … is full of political activity; it is
close to political life’ (Bagehot 1867, 17). Here ‘political life’ refers to the
activity as a substrate of parliamentary politics, one that extends beyond
parliament’s institutional and legal status.
Another perspective could be called again ‘existential’, referring,
firstly, to the ‘political aspect’ in the life of citizens, and secondly, per-
haps also to the varying intensity of this political aspect, which makes of
‘political life’ a journey of gains and losses, a movement between ups and
downs. Both aspects emphasise the contingency of life and politics, and
of their connections in particular. The first insists on ‘political life’ as an
aspect that keeps parliamentary politics ‘alive’ prevents its ossification
and stagnation. The second refers to the presence of the political aspect
in the lives of  the citizens of parliamentary and democratic polities, in
Weberian terms, the common substratum for the occasional and the pro-
fessional politician. The Weberian view includes the formal chances of
citizens to be become professional politicians as well as of the latter’s
possibility or danger of falling back into the status of merely occasional
politicians.
If we look at the references to ‘political life’, many of them are descrip-
tions of existing practices. Many illustrate the above interpretations, for
example, this simple statement: ‘It is one of the pleasant characteristics
of our political life in this country that differences of political opinion do
not sunder old friendships nor indeed prevent the forming of new ones’
(Derek Walker-Smith, 10 November 1947, 130). John Anderson, an MP
for the Scottish universities, formulates the point: ‘None of us, I suppose
would desire that these great constitutional matters affecting the political
life of this country for long periods of time should become the mere play-
things of party politics’ (16 February 1948, 885). John McKey emphasises
148  C. WIESNER ET AL.

the ‘existential’ aspect of entering ‘into politics’, but sees that this does not
require the privileged university constituencies:

If anyone wants to get into political life and he has been through a univer-
sity, has he not had an opportunity above many other people to develop his
capabilities and, as a result, will he not be in a better position to get into this
House in ordinary competition with the rest of his fellow men? (16 March
1948, 1944)

In both senses, the references to ‘political life’ are potentially opposed to


the idea of ‘politics’ as a separate sphere. ‘Politics’ is third on the frequency
list: 15 expressions for ‘in politics’ or ‘into politics’, in addition to one
‘outside politics‘. This view is, of course, closely connected to the para-
digm of the party as a quasi-natural unit of politics.
Nonetheless, among the adjectival expressions very few are explicitly
spatial. More generally, however, we can speak of expressions of political
order, which can be interpreted as attempts to reduce the range of the
contingent and controversial, and which historically is badly suited to par-
liamentary practices. Here perhaps ‘political system’ is the most obvious
candidate, along with the classical figures of ‘political body’ or ‘body poli-
tic’ and ‘political basis’. However, to claim that the US Supreme Court is a
‘political body’, to some extent comparable in its position to the House of
Lords, as the Labour MP Arthur Jones argues in the 1949 debate (column
112), does not in fact refer to a spatial character.
Compared with these examples are the figures of political time, with
different nuances or aspects of thematisation much more frequent in
the debates. This might partly be connected with the reform contexts of
these debates themselves. Kenneth Pickthorn, representing Cambridge
University, would like to have had more ‘medicine men’ in parliament, but
‘they had not the time for politic’ (13 March 1948, 1925). Among the
temporal terms we can detect past-oriented (‘political experience/history/
tradition’), future-oriented (‘political development/future/progress’) and
present-oriented (‘political moment/opportunism/situation’). In addition,
‘political crisis/emergency’ refers to a specific type of momentum together
with a note of warning. ‘Political fortunes’ refers explicitly to the changing
fates of a political career, and ‘political days’ to the changing importance of
politics in the life of the individual. All these expressions, even if not used
very frequently, indicate the presence of an inherent connection to time in
speaking about politics, available for different political purposes.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  149

Of the expressions relating to political play ‘political judgement’ fig-


ures notably, although this core concept was used only once. Some of the
criteria for this concept are explicitly normative, such as ‘political moral-
ity/knowledge/wisdom’. The dominance of the Conservative and other
opposition politicians in the debate becomes visible in the numerous pejo-
rative expressions, ‘political betrayal/claptrap/humbug’, or in their char-
acterisations of the reform, or in speaking of the ‘political offences’ of the
old Stuart kings (Stanley Reed, 17 March 1948, 2196). Thomas Moore, a
former officer, declares: ‘The Home Secretary knows as well as I do how
politically stupid many of our people are’ (31 October 1949, 71). ‘Political
agnosticism’ is an interesting term, borrowed from the nineteenth-­century
theorist Henry Maine, and referring to the lack of strong beliefs in ‘the
man of common sense’ (Cuthbert Headlam, 10 November 1947, 74).
Concepts such as ‘political equality’ and ‘political democracy’ also have
a normative character, but as well-established principles they can also be
used in a descriptive manner. The same holds true for the key precondi-
tions for parliamentary democracy, ‘political representation’ and ‘political
franchise’, of which more discussion could have been expected.
Besides these we can distinguish more formal criteria for political
judgement, referring less to the content than to the art of judging politics.
Formulae such as ‘political doctrine/ideas/philosophy/platform’ seem to
refer to views from the past or to those of other participants, rather than
to declarations of the speakers themselves. The appeal to ‘political theory’
sometimes contains an ironic tone and a demand to connect theory with
practice, as when John McKay states: ‘Another point made was that public
opinion should be assessed and enforced. That is right in political theory.
That is supposed to be the democratic principle, but it all depends on
the question of what public opinion is’ (31 October 1949, 100). Two
classical rhetorical topoi (on their uses prior to the 1640 parliaments, see
Peltonen 2013) are ‘political expediency’ and ‘political principle’, fre-
quently opposed to each other, but sometimes also combined. ‘Political
sense’, ‘political asset’ and ‘political consciousness’ lie on the borderline
between the formal and the conventional understandings of the concept.
Types of political agents are rather scarcely presented in these debates
with the polit-vocabulary. Occasionally such epithets are used as ‘eager/
radical/unsuccessful politicians’, with an obvious pejorative tone for the
two first of these. Besides the formal term ‘political agents’, ‘political
leader/students/supporters’ are also mentioned without further themati-
sation of the concepts. Henry Strauss, representing the Combined English
150  C. WIESNER ET AL.

Universities, accuses the reform proposal of ‘destroying a characteristic


invention of English political genius’ (23 June 1948, 1420). Referring
to a specifically political quality as a criterion can be seen in the formulae
‘political feelings’ and ‘political mind’, whereas Leah Manning refers to
her opposition to the removal of the privilege of the City of London in the
following terms: ‘I always think that in Debates like this I am some kind
of a political schizophrenic’ (17 March 1948, 2175), obviously a conflict
between nostalgia and a recognition of the need for the reform.
More prosaic terms for professional politicians are ‘political career’ and
‘political ascendancy’. Cambridge University representative Henry Harris
emphasises:

Universities are very important training grounds for politicians. […] They
train politicians and increase the interest of the average undergraduate in
politics generally, and in the politics of the moment. (16 March 1948, 1985)

This is, of course, historically important (see the Section 4.3 on the Union
Societies), but does it justify a plural voting for the university graduates,
when they otherwise seem to have better chances in getting elected than
others?
The linguistic border between political actors and political moves is
also relatively diffuse. However, the latter category contains a number of
expressions that either refer to the moves of individuals or their political
groups and so on, or to their different forms of confrontation with each
other. ‘Political activity’ and ‘political controversy’ are the basic forms,
which also illustrate an insight into the contingent and contested character
of politics.
The commitments of individual actors are formulated in such terms as
‘political colour/opinion/outlook/standing/views’, which characterise
their positions in the debate. They are complemented by the rhetorical
moves ‘political argument’ and ‘political speech’ in particular, but also
by ‘political grounds/reasons’, whereas ‘political advantage’ refers to get-
ting the upper hand in rhetoric. In arguing against reform the opposition
leader Winston Churchill accuses the head of the reform committee that
it ‘was too much for his political fibre to bear’ (10 November 1947, 381).
To speak of ‘political measures’ or ‘political policy’, a term used by the
Labour minister Herbert Morrison (31 October 1949, 157), might also
concern individual agents, but historically they rather presume the exis-
tence of a polity, within which these deictic terms are used.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  151

As forms of political controversy we can detect in the debates ‘political


battle/bargains/compromise’, which tend to dissolve the historical and
conceptual divide between parliamentary debates and diplomatic nego-
tiations. A condemnation of certain tactics in the redistribution of voter
constituencies is contained in ‘political gerrymandering’, borrowed from
the US political practice, which was originated by a governor, Gerry.
A small group of termini refer to the question of the political weight
of moves and measures. ‘Political importance’ is perhaps the basic fig-
ure here. It is surrounded by neighbouring concepts, such as ‘politi-
cal consequences/functions/influence’. Alan Lennox-Boyd quotes the
(Manchester) Guardian’s characterisation of the House of Lords’ reforms
as an exercise in ‘political irrelevance’ (31 October 1949, 148).
There remains one set of termini, which could be characterised as
‘political practices’. Some formulae within this section refer to the one-­
sidedly political perspective on the phenomena discussed: ‘political angle’
and ‘political point of view’ are the clearest perspectivist concepts in the
debates. Such perspectivism can also be read into the expressions ‘politi-
cal affairs/character/considerations/matters/problems’, which may seem
neutral, and at least ‘affairs’ and ‘matters’ could be almost used as descrip-
tions of politics as a separate sphere.
To this section we can also include references to ‘political power’ and
its specific constellations: ‘political arrangement/circumstances/complex-
ion/difficulties/forces/funds’. In the core topic of the selected debates
we can include the references to ‘political privilege’, which was something
the reform was to remove. Thomas Stamford claims: ‘Every proposal to
limit political privilege has been met by the resistance of the Conservative
Party’ (16 March 1948, 1980), and even the Conservative Quentin Hogg
does not want to uphold the ‘political privilege’ of the Upper House
members (10 November 1947, 82).
This illustration of the analysis of the polit-vocabulary in the six reform
debates of the Attlee government operates intentionally with some ele-
mentary criteria for classifying human activities. Above all, the discussion
illustrates that ‘political’ in twentieth-century English can be used to refer
to a great number of phenomena and their different aspects, without any
need for more systematic coordination. The terms used for the conceptual
aspects in this chapter—political order, political time, political judgement,
political agents, political moves, political practices, political weight—do not
aim at any deeper classification of the possible perspectives on politics.
They rather are meant to illustrate how, with some elementary criteria, it
152  C. WIESNER ET AL.

is possible to show how complex and multifaceted the understanding of


politics can be among the professionals in the field, here, the members of
the Westminster parliament.
You need not be a historian of the concept or a political theorist in order
to construct such typologies of action, while at the same time, one must
be aware that such constructions are matters of interpretation and political
imagination. In this rather ad hoc construction it is not possible to arrive
at any definitive typology, for we see their use is to some extent specific to
the context (Westminster), the theme (reform) and the time period (post-
war Britain). But for the scholars this result might also be an encourage-
ment: you can use existing typologies, if they suit your purposes, cases and
sources, but you can also alter them or construct new ones yourself.

4.2.3  ‘Politicking’ in the Westminster Debates


The previous section contained an analysis of the polit-vocabulary in a
definite context of parliamentary reform of the Labour government in
the post-war years (until 1949). In this section, an example of the use of
the vocabulary in debates for a longer time period (that covered by the
historic Hansard debates from 1945 until 2005) and independent of the
actual context of the debates will be given. This is more in line with a focus
on the historical changes, as studied in conceptual histories, but even here
the main interest is not so much in the historical dimension as in marked
extraordinary uses of an expression, which are more interesting than the
ordinary uses. The longer period is used here because the focus is just on
single expressions, and the uses of it are relatively rare in comparison with
the multiple expressions analysed in the previous section.
In English, ‘doing politics’ can be expressed with a single word, pol-
iticking. For historical reasons, this expression is usually understood as
a pejorative term, in line with the adjective politick to refer to cunning,
intrigue and so on. In the scholarly literature there have been attempts to
neutralise politicking to correspond to the everyday practice of politicians
(e.g. Gallie 1973).
This is nothing surprising. Attempts by British scholars to neutralise
‘politicking’ have hardly touched upon the vocabulary of parliamentari-
ans. The expression entered Westminster in the late 1960s, and used more
than 200 times between then and 2005 (http://hansard.millbanksystems.
com/search/politicking?sort=date).
EXAMPLES OF ANALYSING DEBATES AS POLITICS  153

Looking at the debates the overwhelming majority of expressions are


definitely pejorative. One crude example is: ‘It is time to accept that poli-
ticking is a cancer’ (Eduard Leadbitter, Commons, 27 October 1983). Or
John Mann’s declaration: ‘We should be dealing with politics, not poli-
ticking’ (Commons, 23 October 2001).
Occasionally we can find more nuanced views, for example, when
Harry Barnes claims: ‘Within politics, there is politicking’ (Commons, 27
January 1999). He recognises that politicking is an inherent part of poli-
tics that cannot be separated from the more ‘honourable’ parts. For the
analysis of concepts in debate, the recurrent and repetitive lamentations
about politicking are much less interesting than the fewer attempts to
revaluate or at least neutralise the concept.
The rhetorical procedure for altering the normative colour of a con-
cept (see Chapter 3) has since antiquity been called paradiastole. Its role
for conceptual changes in general has been revived by Quentin Skinner
(1996, Chapter 4; 1999, 2007). In the Hansard debates we can find at
least three paradiastolic revaluations of politicking:

As late as 1822 George IV, having forgiven the Scots, and wearing a kilt,
held his levee at Holyrood. And why not?—it was good politicking and he
liked dressing up. (Marquess of Aberdeen and Temair, Lords, 25 November
1975)
There will have to be some great and imaginative politicking by our
political leaders in Western Europe. (Michael Mates, Commons, 22 July
1981)
I have spent a lot of time politicking in this place and, if there is a good
idea, I do not mind from where it cometh … (Dale Campbell-Savours,
Commons, 27 April 1999)

We can detect here two different strategies of revaluation of the concept.


In the first two cases the possibility of good or imaginative form of poli-
ticking is recognised, which would not be possible if the concept were
regarded as pejorative per se. In the third case the MP does exactly the
contrary to Schmitt’s dictum, namely, adopts the term for his own activity,
proudly claiming to be politicking himself. For the study of conceptual
changes and innovations these few examples are much more interesting
than the routine disparagements. Indeed, the paradiastolic revaluations
are an old parliamentary device, emphasised already in William Gerard
Hamilton’s eighteenth-century maxims collected in Parliamentary
154  C. WIESNER ET AL.

Logick with the classical example: ‘Run a vice into a virtue; and vice versa’
(Hamilton 1808/1927, 6; see also the discussion of conceptual changes
in Palonen 2016).
The example relates to some ways in which the political concepts used in
parliamentary debates can be analysed. One approach uses debates in part
as a privileged type of source that is extensive and publicly documented.
Another part of the analyses involves actual debate situations, in which the
revision, modification, and the revaluation and devaluation of concepts
have been an inherent part of the debate, as Hamilton noted. A further
approach looks for explicit conceptual controversies in the debates, for
example, when the compensations and salaries of MPs are on the agenda.
Then the concept of representation and the acceptance of the profession-
alisation of politicians are part of the debate (see Palonen 2012a). Even
then, however, the debate is hardly a yes-or-no quarrel, but rather, various
indirect modes of the paradialistolic re-evaluation of concepts are used
(see also Skinner 1996).
In some contexts, especially when the government versus opposition
divide dominates the debate, it is easy for British MPs to employ the
common lament against politicking as a tool against the opposite side.
The few examples that are exceptions from the pejorative use might be
interpreted as a mark of distance, such as when parliamentarians want to
distance themselves from the bad reputation of politics and politicians.
They recognise that politicking also has formal and non-pejorative con-
ceptual resources and refuse to make a difference between ‘good’ politics
and ‘mere’ politicking. To defend politics and politicians against popular
attacks could be done to act contrary to Schmitt’s dictum and adopt the
term ‘politicking’ to describe one’s own activity.

4.2.4  ‘Politics’ in British Parliament


The two exemplary analyses above illustrate something of the uses of the
vocabulary and the conceptual thematisation of ‘politics’ in the debates
among Westminster MPs. One point to note is that there is no ‘ordinary
meaning’ for the concept, although some MPs in the post-war years seem
to have almost identified the expression ‘political’ with ‘partisan’. A closer
analysis illustrates that the use of the adjectives and the conceptual implica-
tions in them are much richer and more diverse than seen among the par-
liamentarians themselves. It is hardly thinkable that this was the case only
when the constitutional reforms were at stake in the House of Commons.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  155

When applying ‘politicking’ as an analytical tool for studying certain


types of acting politically, it is still frequent to hear from colleagues that it is
a pejorative term. The analysis of the use of the expression in the House of
Commons debates show that in many, if not most cases this holds true, but
the conceptual repertoire of the British MP is still much more complex and
nuanced, and they can, for example, in certain situations use it to describe
their own activity or even revaluate the normative colour of politicking.

4.3   Reading and Analysing Political Debate


in Debating Societies

Debating societies include a variety of different types of organisations.


Some are more club-like and accept new members only by invitation,
while others are more public-oriented and organise open debates and
other events. In this case study, ‘debating societies’ refer to organisations,
usually founded by university students, that discuss current issues with the
aim of providing training, as their members often tend to pursue public
careers.
What makes debating societies politically interesting? Is it because they
tend to form political elites? It is important to note that the activities
of debating societies are not necessarily ‘political’ even if they are some-
how linked to parliaments or the political elite. The associations offer
fruitful material for political analysis in so far as they apply parliamentary
­procedure or other rules of debate that create controversy and adversarial
argumentation.
It will be argued here that the political aspect of debating societies
derives from their politics of debate. The politics of debate in debat-
ing societies is closely related to the interpretation of the rules by which
the debating is conducted. Especially in Britain, debating societies have
been known for their formal debates since the eighteenth century (cf.
Fawcett 1980). Historically, debating societies are arenas that have been
constituted for a variety of civic purposes. In Sweden, a debating society
at Uppsala University, for instance, was created to allow students to get
acquainted with then-current practices of eloquence (Burman 2012). The
oldest student debating society in the UK, the Historical Society of Trinity
College in Dublin, was established to provide a platform for those who
wanted to obtain practical knowledge of how to speak and act in public
fora (cf. Samuels 1923). Both still exist and operate, but with the aims and
requirements of our current standards of political life.
156  C. WIESNER ET AL.

Debating societies, just like parliaments, are deliberative assemblies


that adapt to their times. The major difference between debating soci-
eties and parliaments, however, is that societies are not representative,
national assemblies. They rarely have urgent issues to discuss that have
a bearing on the nation as a whole. The issues they vote on do not have
legislative relevance outside the societies themselves. But if we consider
the debating society as an extension of parliament in terms of proce-
dure, we can find distinct similarities in their political debates. Redlich
(1908, 215) argued that a ‘parliamentary body’ is essentially based on
debate, not legislation. This means that debating is the primary func-
tion of such a body. He further maintains that a parliamentary bill is a
series of motions. It then follows that the aims of political activity in
a ‘parliamentary body’ are twofold: firstly, to persuade what kinds of
motions are put onto the official agenda, and secondly, to debate those
motions that are successfully put onto the agenda. Some debating soci-
eties, such as the Oxford and Cambridge Unions, have the basic features
of such deliberative bodies: they operate with motions and resolutions.
According to Redlich, the aim of a deliberative assembly is to ensure
the  fair treatment of issues on the agenda with certain rules for the
benefit of the majority, but without forgetting the rights of the minor-
ity to express their opinions (ibid.). In short, a motion is treated as a
matter of debate before it is turned into a resolution by the majority of
the House.
There are multiple ways to conduct a political reading of these soci-
eties. It is possible to attend live debates, for example, in the Oxford
and Cambridge Unions or other academic debating societies, listen to
recorded debates available online, interview participants, or read ‘frozen’
debates from various sources such as minute books or autobiographies. As
discussed in Chapter 1, ‘live’ debates differ from ‘frozen’ ones only in the
practical sense: the ‘live’ debates can only be attended during the research-
er’s own lifetime. However, the analysis of debates is always a matter of
interpretation, and therefore, it does not matter whether the researcher is
present in the audience or reading the transcripts, as long as she is aware
of the limitations of the situation.
To give an example of a case study based on a researcher’s personal
experience, a detailed account will be given below of how to analyse ‘fro-
zen’ debates from the Oxford and Cambridge Unions’ nineteenth-century
archives. It will be based on a study of the Unions that focused on their
early practices of debate (Haapala, forthcoming 2017, esp. Chapter 5).
EXAMPLES OF ANALYSING DEBATES AS POLITICS  157

Concentrating on nineteenth-­century sources, of course, meant that the


sources for the study were quite limited. It was only possible to inter-
pret the debating practices through the surviving minute books and rule
compilations.
When the researcher (TH) first visited the archives of the Oxford and
Cambridge Unions (after receiving permission to do so), she was given
catalogues of all the materials that were available. The catalogue lists
included, among other things, the minute books, rule compilations, select
committee diaries and private correspondence. Her first task was to decide
which material was pertinent for the study. She chose to concentrate on
the minutes and the rule books as they would reveal the debate topics and
the debating rules. While going through them she found that the minute
books contained all the proceedings in chronological order, with hardly
any indexes or page numbers. Taking copies was not possible due to the
fragility of the records. However, she was allowed to take digital photo-
graphs, which helped a lot in the later stages of the study.
The minutes were handwritten and sometimes hard to follow. The pro-
ceedings provided a minimal amount of information about the debates:
the topic under discussion; when the debate took place; who had been for
and who against; and the result of the final vote. After the names of each
speaker, the records also mentioned the colleges they represented. The
proceedings of the private business meetings gave somewhat more infor-
mation regarding the contents of the debates. They showed, for instance,
debates on rule changes and on Union members’ initiatives.
The initial stage of the study was to acquire as much research material
as possible. When the researcher started reading through the proceedings,
the aim was simply to see if it was possible to get a sense of what the debat-
ing practices were like. But it soon became clear that without a detailed
plan of what to do she was not going to get to the stage of the analysis
proper. The amount of the proceedings was staggering, and though the
researcher kept on taking copies and photographs of the minutes, she was
not sure whether she would need even half of it.
She also needed to narrow down her research question from ‘What
is political about the Union debates?’ to ‘What kind of politics was con-
ducted in Union Society debates?’ At this point, she started to realise
that the proceedings would not provide answers unless she understood
how the debates were conducted. In order to analyse political action in
the minutes, she had to make sense of the rules that framed the debates.
And this led her to the discovery that, in fact, the rules in the Unions
158  C. WIESNER ET AL.

were constantly changed and amended. Why was this so, and who did
it benefit?
A closer examination of the minutes showed that rules were changed
after confusion or deliberate misuse of the rules. As rules provide legitimi-
sation for the actions performed in the Unions, the position of the Union
president caught her attention because it included acting as the head offi-
cer of the society, as well as being both the chairman in the debates and
the chief executive of the standing committee that ultimately decided all
the items of the agenda. The powers of the president were so extensive
that it is no surprise that the other Union members made several efforts to
try and diminish them.
The printed rule books became important for the investigation as they,
unlike the debates themselves (which the researcher was reading from
the minutes), revealed the ‘outcomes’ of the debates. During the mid-­
nineteenth century, the amended Union rules were printed almost every
year for the use of their members. It was clear that the rules were debated
in the Unions’ private business meetings, and the results of the debates
were seen in the rule changes. During the investigation, the researcher
became especially interested in the rules of debate, as well as in the pow-
ers the rules gave to various actors inside the societies. Even the slightest
change in the rules could mean a marked shift in the debating practices. So
she concentrated on finding all the changes in the rules. She was then able
to use the rules as a key to open up the political meanings of the Union
proceedings. For example, a change in the rule that gave the Union presi-
dent more power might catch her attention. To understand what ­triggered
the change she would then consult the minute books and take notes of the
debates on that particular rule. That seemed to be the most fruitful way
of finding out how the original rule had been challenged and what kind of
interpretations of it were presented.
The analysis of rule interpretation that was then conducted focused on
rhetorical strategies. As explained in Chapter 3, rhetoric has traditionally
been seen as the study of persuasion in public assemblies. To follow this
tradition, the study concentrated on the rhetorical aspects of the Oxford
and Cambridge Unions’ debates. The main intention was, first, to show
what kind of rhetorical skills the Union members learnt as they debated,
and second, the ways in which they challenged or interpreted the rules for
their own political benefit (Haapala, forthcoming 2017, esp. Chapter 5).
It was shown that the rhetorical skills were learnt through the conven-
tions and rules of debate, which had been adopted from Westminster’s
EXAMPLES OF ANALYSING DEBATES AS POLITICS  159

parliamentary procedure and applied in the Union debates. But the skills
learnt were used for the Union members’ own political purposes.
To give an example, the limitation of precedence-setting was the main
rhetorical strategy related to the interpretation of rules in the Oxford
Union. The rhetorical aspect, as discussed in Chapter 3, derives from
the idea that any form of language use that addresses an audience and
tries to make it to accept, reject or modify what is under debate can be
considered a form of rhetoric. In that sense, the strategy used in Oxford
Union involved persuasion for limiting the Union president’s powers of
rule interpretation.
One such incident took place in 1856 when treasurer of the Oxford
Union took the initiative to amend a rule that gave any member the per-
mission to propose adjournment and the chance to speak prior to others:
‘That in Rule LXIII,2 after the words “all other speakers” the following
to be added “But no Speaker on any motion for adjournment shall intro-
duce any matter not bearing solely on the question of adjournment”’
(OUS minute book vol. VIII, 2 December 1856). The rule was amended
accordingly. It is crucial to note that the debate preceding the rule change
had involved a controversy where the Union president had given a mem-
ber who had proposed adjournment unlimited permission to speak. The
treasurer had intervened by points of order, but to no effect. By proposing
an amendment to the rule the treasurer’s rhetorical aim was to persuade
the audience to limit the president’s vast powers in this respect.
As rules are often ambiguous to begin with (see Section 1.4), it is dif-
ficult to provide a non-controversial judgement in the first place. In that
regard, the politics of debate in Oxford related to the intentional cre-
ation of a controversy between precedents. At the same time, the increased
ambiguity of the rules also augmented the likelihood of creating more
occasions for debate. Since the rules of debate were so difficult to inter-
pret, it was more than likely that someone would question their applica-
tion at some point. As a rhetorical strategy, this was an excellent way of
assuring that no motion was left to a vote without first having undergone
extensive deliberation.
At Cambridge, the main rhetorical strategy used was reformulation of
the existing procedure. Ultimately, the intention was to make the rules as
explicit as possible in order to limit the president’s interpretative latitude.
The same aim can be read out of the fact that the Cambridge Union rules
were called ‘laws’. The idea behind this was that someone who was inter-
preted as acting against them could be accused of illegal conduct.
160  C. WIESNER ET AL.

In the Cambridge Union laws, it was stated that the president’s deci-
sions could be appealed through a requisition signed by at least one
hundred names: ‘If a requisition of One Hundred Members, with their
Names and Colleges affixed, be presented to the President, it shall be
incumbent on him to appoint an early day for a Committee of the whole
House, to inquire into the propriety of any decisions from the Chair’
(CUS laws October 1848, 16–17).3 This was used as a means to reclaim
from the standing committee the initiative to make procedural changes.
To offer an example, in 1868 a requisition was declared in a Union private
business meeting. It was claimed that the president had acted against the
rule that allowed Union members to question the standing committee
regarding the ‘interests or management of the society’. After a chairman
had been appointed, the members present proceeded to discuss the claim
that the Union president had failed to allow a member to ask a question
‘relative to the Interest of the Society’. According to the motion pro-
posed this had been contrary to the laws of the society, which had been
ratified only a day before the incident (Kemplay: CUS minute book vol.
19, 14 May 1868).
The minutes state that the debate on the requisition went on for two
hours before one member proposed an adjournment. After a short discus-
sion ‘the motion of adjournment was put to the House and lost’. The
‘House’ then voted: ‘When the numbers were for Mr Kemplay’s motion
134, against it 166. Majority against the motion 32. The motion was
declared lost’ (CUS minute book vol. 19, 14 May 1868).
The key point is to recognise that the critiques against the Union presi-
dent were meant to shift some of his powers, even if just for a moment,
to other members of the society. In other words, it is crucial to focus
on the political intention behind the actions against the president. The
intentions, however, must be contextualised. In the previous example, the
context of the debate was that the rules had been changed only recently,
and the Union president was, therefore, caught acting against them. This
allowed the other members to move a motion that gave them the chance
to challenge the president.
To conclude, the politics of debate that was practised in the Oxford and
Cambridge Unions derived from an interpretation of the rules. Some of
the rules had parliamentary references. However, it seems to be irrelevant
where those rules derived from as the actions themselves had political
intentions. In other words, the rules themselves are not political, except
for constituting the Unions as polities based on fair debate; however, the
EXAMPLES OF ANALYSING DEBATES AS POLITICS  161

use of the rules can be interpreted as political. The participants in Union


debates can try to apply the rules for their own advantage, just like in
any institution or organisation that deliberates and makes decisions based
on formal conventions. Ultimately, it is about the persuasive use of lan-
guage, or rhetoric, which enables political action to become legitimate.
The analysis conducted by the researcher shows just how skilfully Union
members were able to act politically when they were aware of the rules and
how they could be interpreted. Their persuasive skills were shown in the
controversies read from the minutes with a special focus on rule interpre-
tation. In other words, the researcher did not only read the minutes, but
also analysed them by paying close attention to the framework of the rules
in which the political action was conducted.

4.4   Parliamentary Rules of Procedure: The Case


of the European Parliament

4.4.1  Rules of Parliamentary Procedure as Documents


in Debates
All parliaments have their own rules of procedure, either in the sense
of written règlements or Geschäftsordnungen as in the Francophone or
Germanic parliaments, or in the combination of conventions, precedents
and written Standing Orders as in the British parliament. There also exists
a broad commentary literature, especially in the British and French parlia-
mentary cultures (discussed in Chapter 2), which is important to know, not
only for parliamentary members and officials, but also for ­parliamentary
scholars, journalists and citizens who follow the debates. The rules of pro-
cedure do not determine what the parliaments do, but they sketch the
horizons of possible and legitimate forms of political action in the parlia-
ment in question. The rules always need interpretation and are always sub-
ject to debate, especially when new events or political struggles arise for
which no regulations are available (see also Palonen and Wiesner 2016).
The following chapter analyses the current rules of the European
Parliament (EP) in relation to its own history, to the conflict between the
French and the British parliamentary cultures and to the power strug-
gle between the institutions of the European Union. It also takes up the
broader political controversies on parliamentary ideal types, on political
times and on the formation of the European types of political agenda and
on European politicians.
162  C. WIESNER ET AL.

4.4.2  Conceptions of Parliament
In both common and academic language, representative, legislative and
deliberative assemblies are equally called parliaments. The concepts refer,
however, to completely different ideal types of political assemblies, and it
would be heuristically important to present these ideal types in their ein-
seitige Steigerung (Weber 1904, 191) before applying them to the actual
parliamentary assemblies. The three ideal types of parliamentary assem-
blies are presented and some aspects contrasting them are summed up in
the following Table 4.3.
It is possible to apply this triad to the European Parliament’s
Rules of Procedure. While direct elections since 1979 have served to
strengthen the EP’s representative powers and treaties since Maastricht
have strengthened its legislative powers, it is the EP’s rules of debate
that, despite the effect of the Francophone procedural tradition, form
its deliberative core. The reading will focus on both the presence of
the rules and the political conflicts between them, with the one-sided
focus on the chances of strengthening their deliberative character. The
background to this is the book The Politics of Parliamentary Procedure
(Palonen 2014c).
Here selected aspects of the European Parliament’s power shares
(Machtanteile for Weber) will be discussed from the perspective of the
conceptual history of its rules of procedure—Rules of Procedure of the
European Parliament, 8th Parliamentary Term, July 2014. The topics of
discussion are the status and powers of the EP’s president, the parliamen-
tary agenda-setting, the rights of the members, the control of parliamen-
tary time as well as the intervention of the Commission and the Council
into the EP’s internal procedures.

Table 4.3  Three ideal Representative Legislative Deliberative


types of parliamentary
assemblies Elections Laws Debates
Majority Vote Procedure
Past Future Present
Instant Progress Journey
Manifesto Plan Amendment
Plenum Committee Plenum & Committee
Partisan Legislator Parliamentarian
EXAMPLES OF ANALYSING DEBATES AS POLITICS  163

4.4.3  A First Look at the EP as a Parliament


Before the procedural discussion, the character and powers of the EP as
formulated on its own website deserve attention. Its first point concerns
the EP’s powers and procedures:

The Parliament acts as a co-legislator, sharing with the Council the power to
adopt and amend legislative proposals and to decide on the EU budget. It
also supervises the work of the Commission and other EU bodies and coop-
erates with national parliaments of EU countries to get their input. (http://
www.europarl.europa.eu/aboutparliament/en)

Here the legislative aspect is presented first, and characterising the EP


as co-legislator refers to its status as the first parliamentary chamber of
the EU, while shared with the second chamber consisting of the Council
of Ministers. Indeed, this refers to the two principles of representa-
tion, namely, the EP as the directly elected chamber, and the Council of
Ministers, which is elected in a doubly indirect manner, as a chamber of
national governments composed of members selected on the basis of a cer-
tain ratio of votes to the inhabitants of the countries. No mention is made
of the European Council being another constellation of the second cham-
ber but without legislative powers and representing the member states
through their ‘prime ministers’, in the ideal-typical sense (see Palonen
and Wiesner 2016). The ‘supervision’ aspect is a euphemistic formulation
of the parliamentary responsibility of the Commission, corresponding to
­parliamentary representation as a means of control of government. And
the final paragraph alludes to the national parliaments, which together
form a kind of third representative and legislative chamber.
Under the title ‘About the European Parliament’ this description is
given:

The European Parliament is an important forum for political debate and


decision-making at the EU level. The Members of the European Parliament
are directly elected by voters in all Member States to represent people’s
interests with regard to EU law-making and to make sure other EU insti-
tutions are working democratically. (http://www.europarl.europa.eu/
aboutparliament/en)

Here both deliberative and the representative aspects of the EP are present.
The interconnection of debate and decision-making shows that the EP is
164  C. WIESNER ET AL.

no consultative assembly, but that its decisions mark, à la Westminster,


the last moves in a debate. The ‘people’s interests’ in lawmaking and the
EP as a guarantee of EU’s democratic character emphasise the representa-
tive aspect. If the two Councils and Commission were tempted to bypass
democratic forms, parliamentary deliberation and representation would
be the obstacles against such tendencies.
On the website the relationships between the three aspects of parlia-
mentary politics are presented as unproblematic. To analyse the rules of
procedure is the first step towards problematising the links between them.

4.4.4  EP’s Procedural History


The legislative view is strong in the US separation-of-powers system, in
which Congress has the monopoly on legislation, but only indirect means
to exercise control and oversight over the presidential government and
administration (see e.g. Kronlund 2013). In the European Union the pres-
idential aspect is concentrated in the European Council (see Palonen and
Wiesner 2016), while the legislative aspect is formed by the EP together
with the Council (of Ministers), the European Commission and, in certain
respects, the European Council as well.
The document analysed in this section, Rules of Procedure of the
European Parliament, 8th Parliamentary Term, was adopted at the
beginning of the present parliamentary term, in July 2014. There exists
an ­extensive ­historical and juridical commentary on the EP’s procedure,
namely Nicolas Clinchamps’s Parlement européen et droit parlementaire.
Essai sur la naissance du droit parlementaire de l’Union européenne (2006).
It is based on a thesis defended in September 2002, thus not entirely
up-to-­date but valuable in discussing the origins of the ‘parliamentary law’
of the EU. Also Clinchamps is keen to discuss in particular the powers of
the EP and of national parliaments. However, the interest here is to focus
on this from the viewpoint of the politics of the EP procedure.
For Clinchamps the key question is: Should the EP be considered as
a subject of international law or of constitutional law? He notes that its
parliamentary character has been increasingly recognised (Clinchamps
2006, 1–8). The EP is no longer a quasi-parliamentary assembly of an
international organisation, but it does consciously apply parliamentary law
(ibid., 20–29). Nonetheless, the parliamentary character is still incomplete
regarding, for example, the representativeness of the EP or the unequal
status between the Members of European Parliament (MEPs) of different
EXAMPLES OF ANALYSING DEBATES AS POLITICS  165

member countries (ibid., 39–141). Clinchamps further emphasises how


the EP has ‘borrowed’ its procedure from the member states (ibid., 144),
although adapting it to the EU’s character (ibid., 150). He divides the
parliament’s autonomy over its procedure into an administrative and a
political aspect, and the latter is subdivided into the organs of the presi-
dency and those of the internal formations, such as the committees and
party groups (ibid., 184).
The EP’s procedure must be seen both as a contingent result of power
struggles in the EU and as a regulator of the intra-parliamentary struggles.
The emphasis on controversies over the EP’s procedure and power shares
offers a perspective for discussing the possibilities and forms of politicisa-
tion of the parliament, as opposed to the quasi-administrative language of
the Commission and of the Councils, which tend to hide their thoroughly
political character.

4.4.5  French and British Procedural Styles


In order to understand a parliament’s procedural document it must be
situated in the genre of procedural commentaries. Thomas Erskine May’s
Treatise (first edition 1844) is the famous and regularly re-edited one,
now as an official part of the Westminster procedure. In France, Eugène
Pierre’s Traité (first edition 1893) has reached a status comparable to May.
Also a number of other parliamentary officials, members and others, in
Britain, France and elsewhere, have published procedural tracts from the
­seventeenth century to present and built up a textual genre, the knowl-
edge of which is indispensable for any study of procedural documents.
For Courtenay Ilbert, a Clerk of the House of Commons, ‘all modern
legislatures, with a very few exceptions, may be traced back to a single
prototype, the Parliament which sits at Westminster’ (Ilbert 1911, 9). The
Westminster procedure can be treated as the closest historical approxima-
tion to the parliamentary ideal type of politics (Palonen 2014c).
An opposite view insists on a radical difference between the Westminster
and the French style of parliamentary procedure. The locus classicus of the-
matising the differences between the French and British rules is Jeremy
Bentham’s Essay on Political Tactics. Bentham’s aim was to create a gen-
eral guide for all legislative and deliberative assemblies. He compared the
Westminster procedure with that of some French provincial assemblies
that were still active in the second half of the eighteenth century. He
found the French practices confusing, for example, as they did not strictly
166  C. WIESNER ET AL.

distinguish between debate and vote: opinion and avis have something of
both qualities (Bentham 1843, VI.5). Bentham realised that his ideal type
of legislative procedure is close to the actual practice of Westminster:

What was more, the very rules that suggested themselves as necessary to
every assembly turned out to be the very rules actually observed in both
assemblies of the British Legislature. What theory would have pitched upon
as a model of perfection, practice presented as having been successfully pur-
sued: never was the accord more perfect between reason and experience.
(Bentham 1999, 1)

The French parliamentary tradition emphasises the written rules (règle-


ments), the British gives the priority to the unwritten rules and precedents
(cf. Hatsell 1818, 4 vols., with Pierre 1887, 18–23). The Westminster par-
liament is dissensual in its dealing with the issues (for the years before the
English Revolution in the 1640s see Peltonen 2013), whereas in revolu-
tionary and in restoration France, only few people accepted the dissensual
Westminster style (see Roussellier 2005; Gunn 2009). Nonetheless, pro et
contra debate also shaped the internal procedure of the French Assembly
in the Third Republic, as Eugène Pierre emphasises (‘les orateurs parlent
alternativement pour et contre’ Pierre 1887, 99, referring to § 103 of the
Règlement).
The most visible difference between the British and the French parlia-
mentary cultures lies in their committees. In the Westminster t­radition,
motions and bills are introduced to the plenum for the first and sec-
ond reading, ‘committed’ only after that, with the ‘report stage’ from
the committee(s) adding an additional plenary before the third reading
(see Campion 1929, 176). At Westminster, plenum and committee do
not differ by membership or size, but by different rules of debate. The
‘Committee of the whole House’ includes all members, but the debate is
conducted along the committee rules, allowing for replies, whereas in the
plenum the members speak only once to the question (on the origins of
this committee see Campion 1958, 27).
The French assemblies submit motions and bills first to bureaux
and commissions before they are submitted to the plenum (on the pre-­
revolutionary origins of this procedure, see Pierre 1924, 747). A new
national assembly and senate first distributes its members by lot to the
bureaux, which do the ‘preparatory examination of projects’ and elect
a member as commissaire to edit and report the bureau’s opinion to the
EXAMPLES OF ANALYSING DEBATES AS POLITICS  167

plenum (see Pierre 1924, 748–751 and note 2 on p.  748 to the tract
of Valette from 1839). The membership of the bureaux was reassigned
by lot on a monthly basis during the Third Republic (ibid., 754). The
bureaux are, as British committees, non-specialist bodies in their member-
ship (Campion 1929, 206–207). There is no proper place in Westminster
for specialisation, but in France since 1848 the commissions, correspond-
ing to the committees, are divided according to the system of government
ministries (Pierre 1924, 775–779) and also choose their own rapporteur
(ibid., 786–787).
Gilbert Campion sees the French and US style as dividing ‘legislation
into distinct categories and the allocation of a separate category to each
Standing Committee’ as ‘entirely foreign to the House of Commons’
(Campion 1929, 206–207). He separates the necessary ‘division of labour’
between members from their specialisation and regards, similar to May,
Standing committees as ‘miniature Committees of the whole House, with
a shifting rather than permanent personnel’. This enables the committees
‘to judge the matters that come before them much as the House itself
would’ (ibid., 207). Later he makes his point more specific:

A standing committee in the House of Commons has no permanence or


individuality; its members are constantly changing; it is distinguished only
by a letter of the alphabet; and it does not receive one type of bills more
than another. It is a miniature committee of the whole House. (Campion
1953, 158–159)

Griffith and Ryle insist on the growing importance of the committees


in the present-day parliament, but neither their procedure nor composi-
tion has been changed. In the Standing committees ‘as in the chamber,
Members speak on their feet’. A better name for them would be ‘debating
committees’ (Griffith and Ryle 2003, 385).

4.4.6  EP’s Francophone Committees


The EP follows the French committee system. As a large and multilin-
gual parliament it has had its difficulties with plenary debates. Clinchamps
remarks on the committees’ decisive role: ‘c’est en commission qu’est
effectué l’essentiel du travail parlementaire‘ (Clinchamps 2006, 212). This
remark emphasises the political weight of the committees as well as their
place in parliamentary agenda-setting and the organisation of debates.
168  C. WIESNER ET AL.

The EP’s permanent committees are ‘en principe specialisées’ (Clinchamps


2006, 215). Clinchamps divides them into the few ‘ideological’ and the
many ‘technical’ committees (ibid., 217). The Rules of Procedure stress
the importance of the role of party groups in the committees:

Members of committees and committees of inquiry shall be elected after


nominations have been submitted by the political groups and the non-­
attached Members. The Conference of Presidents shall submit proposals
to Parliament. The composition of the committees shall, as far as possible,
reflect the composition of Parliament. (Rule 199, 1)

In Britain the committee procedure is characterised by their intensive


debate of the details. In these the exchange of replies between members
is the main aim, as opposed to the plenum with its rule that members
are allowed to speak only once. In the EP no such restrictions on ple-
nary speeches exist; the Committee of the whole House is unknown to it,
though it is practised in the US Congress.
Unlike Westminster, the Bundestag and the Nordic parliaments, the
EP does not have a chance to debate every motion in a first and sec-
ond reading before its commitment. The plenum receives only a report
from the committee, in which a judgement of the strengths and weak-
nesses of the proposal is already made and the subjects of debate are the
amendments.

1. Parliament shall examine the proposal for a legislative act on the basis of
the report drawn up by the committee responsible.
2. Parliament shall first vote on the amendments to the proposal with which
the report of the committee responsible is concerned, then on the proposal,
amended or otherwise, then on the amendments to the draft legislative
resolution, then on the draft legislative resolution as a whole, which shall
contain only a statement as to whether Parliament approves, rejects or pro-
poses amendments to the proposal for a legislative act and any procedural
requests. (Rule 59)

Individual members do have a right to propose amendments, but only


when they have 40 supporters (see Rule 150, discussed below). If the par-
liament alters the Commission’s proposal and the Commission does not
accept them, then ‘Only amendments tabled by the committee respon-
sible and seeking to reach a compromise with the Commission shall be
admissible at this stage’ (61.3).
EXAMPLES OF ANALYSING DEBATES AS POLITICS  169

The Rule 49 describes the legislative procedure of the committees:

1. The Chair of the committee to which a proposal for a legislative act is


referred shall propose to the committee the procedure to be followed.
2. Following a decision on the procedure to be followed, … the committee
shall appoint a rapporteur on the proposal for a legislative act from
among its members or permanent substitutes if it has not yet done so on
the basis of the Commission Work Programme.

The standing committees are not places of intensified debate; rather,


the chair and the rapporteur of the committees are the key figures. The
rapporteur both defends the committee’s work before the plenum and,
together with the chair, guarantees the optimal functioning of the com-
mittee itself (Clinchamps 2006, 235). Even more importantly, the rap-
porteur of the committee speaks on behalf of the Parliament before the
Commission and the Council (ibid., 236).
The committee reports are presented in the Rule 56:

1. The explanatory statement shall be the responsibility of the rapporteur


and shall not be put to the vote. It must, however, accord with the text
of the motion for a resolution as adopted and any amendments proposed
by the committee. If it fails to do so, the chair of the committee may
delete the explanatory statement.
2. The report shall state the result of the vote taken on the report as a whole.
In addition, if at least one-third of the members present so request when
the vote is taken, the report shall indicate how each member voted.
3. Where the committee’s opinion is not unanimous the report shall also give
a summary of the minority opinion. Minority opinions shall be expressed
when the vote on the text as a whole is taken and may, at the request of their
authors, be the subject of a written declaration not exceeding 200 words in
length, annexed to the explanatory statement.

The rapporteur summarises the committee debates and votes. In


Westminster committees the members can use their political imagination
and rhetorical competence to control the powers of experts and specialists
(see Weber 1918, esp. 235–237). As the committee memberships consist
of experts, MEPs outside the committee seldom have a real chance to
introduce new amendments in the plenum. The committee instead gathers
its members together with the Commission’s and member states’ officials
in the specialist fields and lobbyists on the basis of their common speciali-
sation. The chair and the rapporteur do, however, have some chances to
170  C. WIESNER ET AL.

break down specialist interests. Clinchamps’ worry about giving too much
power to committees sounds realistic (2006, 208).
The European Parliament has not changed its way of proceeding from
the time it was a merely consultative parliament and Britain not yet a
member. It seems never to have debated, whether it would not be better
to bring motions to the plenum first. Clinchamps thinks that the multi-­
local and multilingual character of the EP has further strengthened the
‘rationalisation’ of debates (Clinchamps 2006, 289–310).
The EP’s specialist-based committee system resembles the US Congress
as a legislature. In a strong representative assembly the committees would
be secondary to public plenary debates. At Westminster, parliament com-
mittees do have an importance equal to the plenum, but they are regu-
lated by different rules, which highlight on the political significance of
the amendments initiated by the members regardless of party affiliations.
With the chair and the rapporteur dominating the EP’s committees, the
possibilities of their rank-and-file members to achieve important amend-
ments are marginal.

4.4.7  Free Mandate versus Party Groups


A deliberative aspect of the EP lies in the exclusion of the imperative
mandate. The second rule proudly declares: ‘Members of the European
Parliament shall exercise their mandate independently. They shall not be
bound by any instructions and shall not receive a binding mandate’. This
rule is a necessary condition for the parliament as a deliberative assem-
bly: its members are not merely delegates, but representatives. It sets a
mighty obstacle on the imposition of mandates by parties or constituen-
cies, and is an insurance against the parliament becoming a ‘congress of
ambassadors’, as parodied by Edmund Burke in his Bristol speech of 1774.
Understanding the free mandate is necessary for understanding the EP as
a deliberating parliament and not as an interstate organisation.
The strong assertion of the free mandate is one of the reasons why EP
elections are not as decisive for the politics of the parliamentary term as
they are in many other parliaments. The government versus opposition
divide is absent, and the party groups are relatively heterogeneous. Even if
the Commission is responsible to EP (as has been the case at least since the
2014 election of the Juncker Commission), it is not formed as an inter-
party coalition. The politics of the EP remains much less predictable than
EXAMPLES OF ANALYSING DEBATES AS POLITICS  171

in countries in which either the electoral majority or a coalition agreement


fixes the agenda for the entire parliamentary term.
The parliamentarians’ free mandate corresponds to the ‘neo-Roman’
opposition between freedom and dependence (e.g. Skinner 2006). Other
parliamentary features of opposing freedom to dependence include free-
dom of speech and ‘parliamentary immunity’, or ‘freedom from arrest’ in
Westminster language (see e.g. Hexter, ed. 1992). Such a rule prevents gov-
ernments from manipulating the power relationships in an elected assembly
by imprisoning opponents. In the EP’s rules of procedure this principle is
recognised: ‘Parliamentary immunity is not a Member’s personal privilege
but a guarantee of the independence of Parliament as a whole and of its
Members’ (Rule 5.2). Today parliamentary immunity is important to pro-
tect the members against their own state, as formulated in Rule 7.

1. In cases where the privileges and immunities of a Member or former


Member are alleged to have been breached by the authorities of a Member
State, a request for a Parliament decision as to whether there has, in fact,
been a breach of those privileges and immunities may be made.
2. In particular, such a request for the defence of privileges and immunities
may be made if it is considered that the circumstances constitute an
administrative or other restriction on the free movement of Members
travelling to or from the place of meeting of Parliament or on an opinion
expressed or a vote cast in the performance of their duties.

The procedure for immunity can be found in Rule 9.1:

1. Any request addressed to the President by a competent authority of a


Member State that the immunity of a Member be waived, or by a
Member or a former Member that privileges and immunities be defended,
shall be announced in Parliament and referred to the committee
responsible.

The paragraph’s limitation of member states’ sovereignty is a political


implication of the parliamentary character of the EP.
The ‘Conference of the Presidents’ consists of the EP’s President and
of chairs of the parliamentary groups (Rule 26.1), and its agenda-set-
ting powers are greater than those of the ‘Conference of the Committee
Chairs’ (see Clinchamps 2006, 188–195). Party groups are formed on
the basis of negotiations between national party lists, as in the French
Third Republic, and they may change during the length of an EP’s
172  C. WIESNER ET AL.

term. The division into party groups is also a sign of the EP’s procedural
principles.
Members may form themselves into groups according to their politi-
cal affinities. Parliament need not normally evaluate the political affini-
ties of the members of a group. In forming a group under this rule,
the members concerned accept by definition that they have political
affinities. Only when the freedom to have political affinities is denied
by the members concerned is it necessary for parliament to evaluate
whether the group has been constituted in accordance with the rules
(Rule 32.1).
The EP requires a minimum of members to form a parliamentary
group. ‘A political group shall comprise Members elected in at least
one-quarter of the Member States. The minimum number of Members
required to form a political group shall be 25’ (32.2). The constitution
of a party group requires a statement be made to the President of the
EP (32.5) that they have a status within the organisation of the EP and
are provided a secretariat (33.1), whereas non-affiliated members lack the
group privileges. An attempt to form a parliamentary group of indepen-
dent ­members ­without a common political platform was rejected by the
EP (see Clinchamps 2006, 271–272).
The EP’s parties have great power in assigning their members to the
committees:

Members of committees and committees of inquiry shall be elected after


nominations have been submitted by the political groups and the non-­
attached Members. The Conference of Presidents shall submit proposals
to Parliament. The composition of the committees shall, as far as possible,
reflect the composition of Parliament. (Rule 199.1)

The same paragraph contains, however, an important qualification regard-


ing party changes:

When Members change their political group they shall retain, for the
remainder of their two-and-a-half year term of office, the seats they hold
in parliamentary committees. However, if a Member's change of political
group has the effect of disturbing the fair representation of political views in
a committee, new proposals for the composition of that committee are made
by the Conference of Presidents in accordance with the procedure laid down
in paragraph 1, second sentence, so that the individual rights of the Member
concerned are guaranteed. (Rule 199.1, italics removed)
EXAMPLES OF ANALYSING DEBATES AS POLITICS  173

This formula guarantees the free mandate of those members who change
their party affiliation—on the condition that the change does not disturb
the interparty configuration of the committee. While the freedom exists,
this warning against party changes during EP membership is issued in the
name of efficiency.
The priority of party affiliation also holds for putting questions to the
Council or Commission, when for independent members’ questions 40
members are required:

Questions may be put to the Council or the Commission by a committee,


a political group or at least 40 Members with a request that they be placed
on the agenda of Parliament … Such questions shall be submitted in writ-
ing to the President who shall immediately refer them to the Conference of
Presidents. (Rule 128)

No chance is thus given to individual MEPs to persuade their colleagues in


the plenum; instead questions must be put in advance, either by securing
the support of the party or committee or by individually recruiting from
the corridors the 40 supporters needed. This threshold seriously limits the
EP’s character as a deliberative assembly.
The absence of a government–opposition divide decreases the need for
whip control, but nevertheless the huge number of technical votes that
occur requires that members follow their party, if they otherwise have
no opinion. It is also worth emphasising that MEPs in their party groups
might learn to think in broader European terms and understand better the
provincialisms and anachronisms of their own countries.

4.4.8  The Intervention of the Commission and the Councils


The relative increase of powers of the EP by treaties from Maastricht to
Lisbon, as well as the de facto achievement to appoint the Commission
and hold it responsible to the Parliament is visible in the EP’s rules of
procedure. Still, the pre-parliamentary history of the EU shapes the Rules
of Procedure. Considering the Commission as the EU’s government, the
commissioners as cabinet ministers and the Council/European Council as
the second chamber of the EP, we can identify discrepancies between its
current procedures with those of strictly parliamentary regimes.
The treaties restrict parliament’s initiative to specific questions: ‘In
cases where the Treaties confer a right of initiative on Parliament, the
174  C. WIESNER ET AL.

c­ ommittee responsible may decide to draw up an own-initiative report’


(Rule 45). Of course, many governments have restricted not only parlia-
mentary sovereignty but also parliament’s procedural autonomy directly
or indirectly, as the Commission and the Council do in the EU.
The most visible deviation from the Westminster type of parliamen-
tary sovereignty is the binding of the EP to the ‘Commission’s Work
Programme’, as presented in the Rule 37.

1. Parliament shall work together with the Commission and the Council to
determine the legislative planning of the European Union.
Parliament and the Commission shall cooperate in preparing the
Commission Work Programme—which is the Commission’s contribu-
tion to the Union’s annual and multiannual programming—in accor-
dance with the timetable and arrangements agreed between the two
institutions and annexed to these Rules of Procedure.
2. In urgent and unforeseen circumstances, an institution may, on its own
initiative and in accordance with the procedures laid down in the Treaties,
propose adding a legislative measure to those proposed in the Commission
Work Programme.
3. The President shall forward the resolution adopted by Parliament to the
other institutions which participate in the European Union’s legislative
procedure and to the parliaments of the Member States.
The President shall ask the Council to express an opinion on the
Commission Work Programme and on Parliament’s resolution.

Most national governments declare their programme and frequently


also submit it for a debate and vote, in which the parliament might, in prin-
ciple, also reject the new prime minister. The jargon of ‘work programme’
and ‘legislative planning’, however, rather resembles the Soviet five-year
plans: the programme is not to be debated, amended and finally accepted
by the parliament. The EP has only a preparatory power with respect to
Commission programmes. The Commission’s proposals to the parliament
must be justified in terms of this work programme. The second point of
the Rule 37 leaves merely an exceptional status for the parliament’s own
legislative initiatives, admitting that ‘urgent and unforeseen circumstances’
may sometimes overstrain the planning capacity of the Commission.
In a parliamentary system the government initiates the main part of
the motions and bills to the parliament, but within the rounds of parlia-
mentary deliberation it has no formal rights to intervene. Still, the minis-
ters—even when they are not members of parliament—might express their
opinion in plenary sessions and the prime minister might, for example,
EXAMPLES OF ANALYSING DEBATES AS POLITICS  175

declare that if the parliament adopts a certain stand, the government will
consider it a question of no confidence. Informally, government support-
ers may present the government’s standpoint in the committees in which
ministers do not have a seat.
In the EP the formal autonomy of parliamentary deliberations is
violated. Even before a proposal arrives from committee to plenum,
­
the views of both the Commission and the Council must be taken into
account:

1. Before the committee responsible proceeds to the final vote on a pro-


posal for a legislative act, it shall ask the Commission to state its position
on all the amendments to the proposal adopted by the committee, and
the Council to comment.
2. If the Commission is not in a position to make such a statement or
declares that it is not prepared to accept all the amendments adopted by
the committee, then the committee may postpone the final vote.
3. If appropriate, the position of the Commission shall be included in the
report. (Rule 58)

A similar dualism follows throughout the entire legislative procedure.


The EP’s stands are ‘referred’ to the Council and the Commission,
these ‘inform’ the parliament on their view and so on. The parliament
does not pass a sovereign decision modified by negotiations between
the three ‘powers’; instead the Commission and the Council intervene
at each stage and pressure the parliament to agree upon a compromise,
which is, at the end of the day, regarded as acceptable. In other words,
the illusory jargon of ‘cooperation’ is a sign that the three institu-
tions do not follow their parliamentary procedure but instead make a
diplomatic-style adjustment to each other’s standpoints. Indeed, the
Commission and the Council intervene in the parliamentary debate
and subordinate this debate to the diplomatic logic of inter-institu-
tional negotiations.
The deliberative parliament is disturbed by the intervention of a
Commission, which acts both as a cabinet responsible to the parlia-
ment and as a super-bureaucracy taking care of the EU’s ‘general inter-
est’ over the partisan interests of the EP, the Council and the member
states. When parliament is seen as site of ‘particular interests’, it is lim-
ited to the representative aspect, and there is little understanding that
the ‘general European interest’ could be a result of a parliamentary-style
debate in the EP and its pro et contra deliberations. The Commission as a
176  C. WIESNER ET AL.

­ overnmental-­cum-administrative body is unable to engage in such multi-­


g
perspectivistic deliberations.
The Council does not act fully as the EU’s parliamentary Senate either,
as the member states with their veto powers intervene in its decision-­
making. In order to achieve any result at all, the parliamentary procedure
is subordinated to negotiations between the three institutions, as if they—
despite all talk on cooperation and coordination, or, rather, just because of
it—were powers foreign to each other.

4.4.9  The Parliamentary Regulation of Debate


The conduct and regulation of debate plays a secondary role in the EP’s
Rules of Procedure. The free speech of individual members does not
shape the debates, but instead a strict distribution of time for debating
is imposed, based on the principles known as clôture and guillotine. At
Westminster the Speaker of the House decides who speaks next with an eye
towards rotating between pro and con speakers, but the EP operates with
a list of speakers based on the party groups, as formulated in Rule 162.

1. The Conference of Presidents may propose to Parliament that speaking


time be allocated for a particular debate. Parliament shall decide on this
proposal without debate.
2. Members may not speak unless called upon to do so by the President.
Members shall speak from their places and shall address the President. If
speakers depart from the subject, the President shall call them to order.
3. The President may draw up, for the first part of a particular debate, a list
of speakers that includes one or more rounds of speakers from each polit-
ical group wishing to speak, in the order of their size, and one non-
attached Member.
4. Speaking time for this part of a debate shall be allocated in accordance
with the following criteria:

(a) a first fraction of speaking time shall be divided equally among all the
political groups;
(b) a further fraction shall be divided among the political groups in propor-
tion to the total number of their members; and
(c) the non-attached Members shall be allocated an overall speaking time
based on the fractions allocated to each political group under points (a)
and (b).

The Conference of Presidents distributes debating time by applying the


guillotine and speaking quotas to parties according to their parliamentary
strength. The fourth clause indicates first the priority of the parties over the
EXAMPLES OF ANALYSING DEBATES AS POLITICS  177

members, and secondly the preference given to the parliamentary strength


of the parties. The rest is then divided between ‘the non-attached Members’
by considering the implicit political groups among them. This party-based
speaking corresponds to the strong representative conception, for which
the plenary speeches are largely statements or pre-written standpoints. Only
on the sixth point is there a residual of free debate reserved for members.

The remaining part of the time for a debate shall not be specifically allocated
in advance. Instead, the President shall call on Members to speak, as a gen-
eral rule for no more than one minute. The President shall ensure—as far as
possible—that speakers holding different political views and from different
Member States are heard in turn. (162.6)

Here the EP suddenly appears as a deliberative assembly, although the


previous points have done their work at undermining such a vision. The
fair distribution of parliamentary time with open deliberation clashes with
the partisan power of numbers. Whereas Westminster excludes the reading
of speeches, nothing prevents an MEP from reading pre-written declara-
tions without any reference to the previous debate. Still, occasionally the
debates in the EP develop a parliamentary character that becomes more
important than the party quotas, but there is no guarantee or support for
that in the rules of procedure.
For the parliamentarians a degree of indirect dependence can also exist
towards interest groups and lobbyists. Rule 11.1 affirms the MEP’s inde-
pendence in opposition to this:

Parliament shall lay down rules governing the transparency of its Members’
financial interests in the form of a Code of Conduct which shall be adopted
by a majority of its component Members, in accordance with Rule 232 of
the Treaty on the Functioning of the European Union…Those rules shall
not in any way prejudice or restrict Members in the exercise of their office
or of any related political or other activity.

The Code of Conduct (Appendix 1 to the Rules of Procedure) has been


adopted according to the model of the ethics commissions. An alterna-
tive would be to consider conduct as unparliamentary. This tradition can,
however, be seen in the chapter on the regulation of debates, in Rule 165:

1. The President shall call to order any Member who disrupts the smooth
conduct of the proceedings or whose conduct fails to comply with the
relevant provisions.
178  C. WIESNER ET AL.

2. Should the offence be repeated, the President shall again call the Member
to order, and the fact shall be recorded in the minutes.
3. Should the disturbance continue, or if a further offence is committed,
the offender may be denied the right to speak and may be excluded from
the Chamber by the President for the remainder of the sitting. The
President may also resort to the latter measure immediately and without
a second call to order in cases of exceptional seriousness.
4. Should disturbances threaten to obstruct the business of the House, the
President shall close or suspend the sitting for a specific period to restore
order. If the President cannot make himself heard, he shall leave the
chair; this shall have the effect of suspending the sitting. The President
shall reconvene the sitting.

These violations of order are not called unparliamentary conduct but


‘disruption of the business’. The parliamentary tradition is followed in
Rule 166, marking the hierarchy of the penalties. Such were presented
at Westminster probably for the first time against the Irish members’
obstruction campaign in the 1870s and 1880s. The point ‘forfeiture of
entitlement to the daily subsistence allowance for a period of between
two and ten days’ (166.3.b) corresponds to Jeremy Bentham’s old idea of
pecuniary losses as a disincentive against absence without permission (see
Bentham 1843, VI.5).
Subsuming the MEPs’ activities under the ‘rules of conduct’ model
instead of extending the concept of unparliamentary conduct to include
financial sanctions illustrates the legislative paradigm. It shows a lack of
insight into the singularity of parliamentary politics and its requirements
for members.

4.4.10  The EP President as a Negotiator


Clinchamps notes that EP president’s political position lies between the
neutral position of the Westminster speaker and the partisan position of
speakers of the US House and Senate. Within the practice of a rotating
presidency the EP president is elected for half of the parliamentary term
from one party group and replaced by a president for another group for
the rest of the term (Clinchamps 2006, 200–207).
In contrast to Westminster, the EP president can intervene in debate
by leaving the chair to one of the 14 vice-presidents. ‘The President may
speak in a debate only to sum up or to call speakers to order. Should he
wish to take part in a debate, he shall vacate the chair and shall not reoc-
cupy it until the debate is over’ (Rule 22.3).
EXAMPLES OF ANALYSING DEBATES AS POLITICS  179

The EP president further represents the parliament in the EU events:


‘Parliament shall be represented in international relations, on ­ceremonial
occasions and in administrative, legal and financial matters by the
President, who may delegate these powers’ (Rule 22.4; see Clinchamps
2006, 208). The presidents of the Council and the parliament jointly
summon a ­conciliation committee to negotiate compromises between the
institutions (ibid., 208–209).

The President shall forward the resolution adopted by Parliament to the


other institutions which participate in the European Union's legislative pro-
cedure and to the parliaments of the Member States.
The President shall ask the Council to express an opinion on the
Commission Work Programme and on Parliament's resolution. (Rule 37)

The EP’s president thus acts politically by ‘forwarding’ and ‘asking’, serv-
ing as a ‘messenger’ both from the EP to the Council and Commission
and back. Her messages are intended to facilitate agreement upon compro-
mises. The president’s addresses justifying a stand by the EP serve as a first
instance of conciliation, before setting up a joint inter-institutional com-
mission to reconcile the decisions. The president’s role is also ­decisive if
the EP wants to delay its acceptance of the EU budget through  seek-
ing conciliation with European Council proposals (Clinchamps 2006,
210–211). The position resembles that of the head of the multi-
cameral institution, rather than that of a referee in an intra-parliamentary
debate.

The President shall participate in regular meetings between the Presidents


of the European Parliament, the Council and the Commission convened, on
the initiative of the Commission, under the budgetary procedures referred
to in Title II of Part Six of the Treaty on the Functioning of the European
Union. The President shall take all necessary steps to promote consultation
and reconciliation of the positions of the institutions in order to facilitate
the implementation of the procedures aforementioned. (Rule 89)

The EP president is a parliamentary diplomat negotiating between assem-


blies and other institutions. As a representative of the EP, its president is
considered by third states as something more than the head of an assem-
bly (Clinchamps 2006, 208). The EP president’s power towards third
states draws legitimacy from the fact that she is voted into office in ­general
180  C. WIESNER ET AL.

e­lections, unlike the Council and Commission presidents. Therefore,


the president is considered ‘aux yeux de l’opinion publique comme une
sorte de chef d’État européen’ (ibid.), as Clinchamps quotes Jean-Louis
Bourbon from 1997.

4.4.11  The Politics of Time in the EP


In Westminster-type parliaments political time is based on a parliamentary
journey of motions through different stages and types of debate. While
the final vote is a part of the journey, intermediate votes can also stop the
forward movement of a motion to the next stage or shelve it indefinitely
by adjourning it sine die. The use of time is a significant part of politics,
and amendments are its main medium. Due to thorough debates from
the different angles in each stage of debate, even important government-­
initiated motions and bills seldom go through without any changes. The
time of political calendar, with its daily, weekly, sessional and electoral
rhythms, is adapted to the intra-parliamentary modes of spending and sav-
ing time, not vice versa. The use of clôture (a motion to end the debate)
and guillotine (predetermination of the time used for a debate) have not,
contrary to expectations, changed the Westminster politics of time so radi-
cally (see Griffith and Ryle 2003).
In the time profile of the European Parliament the external criteria
enjoy priority over the parliamentary ones. EP’s parliamentary time is
divided between the plenary, the committee and the constituency weeks,
and in additional between the Brussels and Strasbourg weeks (Clinchamps
2006, 173–184). In other words, the plural spaces of activities restrict the
internal temporality, and the member countries as electoral districts seem
to reduce the intra-parliamentary focus of the MEPs.
As mentioned, what is missing from the EP is an opening plenary debate
where a commissioner or an MEP would introduce motions. Instead,
members face motions that have already been debated in committee and
formulated by the committee rapporteur. The plenary session corresponds
to the Westminster ‘report stage’, but the original motion and the debates
on amendments in the committee remain practically unknown to mem-
bers outside the committee. No extended debate on the specialist-bills
by parliamentary generalists is expected, and no time is reserved for that.
The speaker list, the time quotas and a party-based use of guillotine and
clôture prevent individual MEPs from subjecting the committee versions
to a thorough questioning.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  181

Westminster-style politics extends the parliamentary present to


encompass all stages of debate. It presupposes the priority of the intra-­
parliamentary time schedule, but such a schedule has little chance of com-
ing to the EP due to, for example, interventions of the Commission and
the Council. Finally the EP committees also require individual MEPs to
focus on specific issues, and they have no need to use the plenary or the
committee stage to augment their own political profiles. In other words,
no real career incentives exist for specialists to aspire to a European level
of politician.

4.4.12  
The MEP as a Politician
In many countries EP membership used to be a step backwards rather than
forwards in a politician’s career. This has changed with the increased pow-
ers of the parliament within the EU, although unlike Westminster-type
parliaments, the commissioners cannot retain their seats as MEPs. The
same is the case if an MP is elected minister in a member state.
Nonetheless, inside the EP we can discern possibilities for advancing
one’s political career. Among the frontbenchers we could include four
groups of MEPs: the presidents and vice-presidents of the EP, the chairs
and vice-chairs of parliamentary party groups, and the committee chairs
and rapporteurs in the committees. This would mean that some 50 to 100
MEPs (out of 751) could be counted as frontbenchers at a time. More
informally, the former presidents and party chairs, also former commis-
sioners or member state prime ministers within the EP, could be counted
as frontbench European politicians.
The members of the Conference of Presidents are the closest to the par-
liamentary type, combining deliberating and negotiating skills, while party
leaders correspond to the representative type. Committee chairs and in
particular rapporteurs are more like legislative specialists. Perhaps we can
also speak of a quasi-diplomatic type of negotiating MEP, who is involved
in the EP’s relationship with member state parliaments, a topic that has
gained greater importance within the EU.
By plenary speeches alone an MEP can hardly advance her political
career. Merely the fact that the length of the speeches is restricted to a few
minutes prevents the formation of a classical type of parliamentary orator
in the EP. Also for a debater used to replies and spontaneous interventions
from the floor, the EP seems to offer no proper opportunities to cultivate
such skills, either in the plenum or in the committee.
182  C. WIESNER ET AL.

The limits to parliamentary initiative as well as the intervention of


the Commission and the Council weaken the MEPs’ independence. The
strong prohibition against imperative mandates as well as the less strict
control by whips—except for voting—leave, however, the MEPs more
occasions for independent initiative and agency than to the rank-and-file
parliamentarians in the member states.
An orientation to the European political agenda instead of the national
one would be a necessity for any MEP, however critical of the EU or
the EP they might be. This would at least be the case with those who
would be interested and engaged in the matters on the EP’s agenda—
and not content merely with the status of an MEP, as many of the right-­
wing populists of today are. Especially MEPs with previous parliamentary
experience could be expected to master the agenda and the practices of
two types of parliaments as well as their inter-parliamentary relationships
to each other. Even backbenchers could be recognised as possessing an
EU-level of political competence that distinguishes them from the parlia-
mentarians of the member states.
A major reason for the MEP adopting a European view of politics lies in
the way the EP sets its agenda. The politics of the EU is the sole concern
of the motions and of the major decisions of the parliament. EU politics
also figure prominently in all EP relations with the Commission that are
debated and voted upon. To act efficiently, an MEP must understand that
there exists a distinct EU agenda and that it is to a great extent the MEPs
who are responsible for this form of European politics.
Two terms can be expected to be sufficient to teach MEPs that they
cannot remain parliamentary lobbyists merely for their own state or for
any special policy fields if they ever want to be more than backbenchers.
We can easily imagine even a scenario in which an MEP who declares to
be solely concerned with her home country could be ‘named’ by the presi-
dent for unparliamentary conduct. We can expect the members to learn to
think about politics in a European manner (see the interviews with Finnish
MEPs in Pekonen 2011). It is well known that there have been politicians
with strongly euro-sceptic profiles, but who nevertheless act just like the
others in the EP and its committees. Within the EP politicians can be
expected to learn about the particular ‘blind spots’ and provincial features
of their own countries and to think within a broader European perspective.
A parliamentarian with the experience of membership in European,
national, regional and local parliaments, on parliamentary boards of
EXAMPLES OF ANALYSING DEBATES AS POLITICS  183

international organisation (such as the Council of Europe or the United


Nations), in the Inter-Parliamentary Union and so on is a different type of
politician than one who has sat in a single parliament only. The former is
competent in the common parliamentary mode of doing politics as well as
in playing with the different powers and rules of the different assemblies.
The EP has enabled a new level of politician, the inter-parliamentary type,
to come into being.

4.4.13  Conclusions
Judgements about the character and quality of the European Parliament
depend on one’s conceptions of parliament in general. The oft-repeated
claim of the weakness of the EP is bound up with a legislative vision: the
EP does not have any real share in the many fields of legislation, and the
interventions of the Commission and the Council(s) tend to strengthen its
diplomatic rather than its parliamentary character.
But parliaments also do many other things than legislate—they pass
resolutions on a range of matters, offer occasions for proposing initiatives
and putting questions to ministers, debate and vote on the elections and
dismissals of governments, and so on. The EP has taken onto its agenda
questions to which it formally has no powers, but which it considers itself
justified to deal with by virtue of its representative and deliberative char-
acter (see e.g. Wiesner 2014c; Tiilikainen and Wiesner 2016). In both
of the latter respects it is superior to the Commission and to the two
Councils.
In order to combine the deliberative and representative visions of par-
liament, it is important to reinterpret the act of voting. Instead of being
regarded as an expression of the ‘will of the citizenry’, the election of EP’s
members could be understood as a contribution to the EP’s own debates.
Inverting Rousseau’s old topos this vision could extend on the election day
the EP’s membership to all voters (see Palonen 2010b). Such a reinterpreta-
tion could also provide chances to increase the turnout in the EP elections.
For the formation of a distinctly European parliamentarism, a de-­
nationalisation of the electoral districts would be a major reform to get
rid of the constituency-based quasi-mandate. The strong European-level
political agenda of the EP also marks the primacy of parliamentary-style
politics, that is, debates pro et contra both on issues on the agenda and on
agenda-setting itself.
184  C. WIESNER ET AL.

4.5   Doing Politics via Policy Documents


and Laws: The Case of EU Citizenship Rights

In the following, examples of doing politics via policy documents and law
texts will be studied. The examples relate to a case that is particularly per-
tinent for doing politics in a seemingly non-political manner, or a politics
that takes place behind closed doors and is related to government activi-
ties: European integration has for many decades been decisively linked
to lawmaking by non-public bodies. The examples studied relate to the
legal and political processes that shaped European citizenship laws and
also European citizenship practice. The analysis will focus on how these
rights have been shaped and fought over via policy documents and laws.
Before entering into the cases, it is important to discuss the context, that
is, the contemporary practices of citizenship in the EU member states, and
the various ways of lawmaking and policymaking in the EU.

4.5.1  The Conceptual Context: Citizenship in Western Nation-­


States and the EU
Citizenship is a key political concept in modern representative democra-
cies in several respects. Citizenship links a polity with the individuals; it
defines the electorate and it establishes the core rights that are essential for
participating in the social and political life of a country. But the concept
of citizenship has always been contested in theory and in political practice,
as well as according to the cultural contexts and different nation-states
(Koselleck 2006; Skinner 1993). Citizenship practices have also changed
over the centuries. If, in the Greek polis, citizenship practices were very
intense, one must remember that only a small part of the population
possessed full citizenship rights (Pocock 1998). Exclusions from citizen-
ship continued, but concerned different groups later on. In the history
of western nation-states, citizenship rights thus were at the core of the
democratisation processes; many battles were fought around them as well
as numerous heated political and parliamentary debates (Wiesner 2008).
As the way of organising citizenship in modern representative democra-
cies has for many centuries been via the nation-state, the practice of citizen-
ship has been closely related to the national political culture in question.
Despite national specificities, however, a few generalisations on national
citizenship can be made: In today’s national representative d ­ emocratic
EXAMPLES OF ANALYSING DEBATES AS POLITICS  185

systems, citizenship defines the democratic subject, the demos, or the sov-
ereign in a legal and political sense. There are four sub-dimensions that
shape nation-state citizenship in theory and practice: access, rights, duties
and the active content of citizenship (the latter can be termed ‘political
activity’). The four dimensions are covered by most national concepts of
citizenship, which define conditions of access (e.g. mostly nationality rules
and related laws), the legal consequences of citizenship in the sense of a
citizen’s rights and duties and how a citizen carries out her political activ-
ity. Lastly, national sovereignty has been closely linked to nation-state con-
ceptions of citizenship. Not only did the nation-state decide who was to be
admitted into the state’s sovereign territory and who was to be expulsed,
they also decided on the treatment and benefits accorded their nationals.
But nation-state-related concepts of citizenship have been challenged
by European integration and the changes it has brought about in the prac-
tices of citizenship. In today’s EU, all nationals of one of the member
states are also ‘Union Citizens’ and possess other EU-related rights, but
EU citizenship differs in many decisive respects from the established form
of nation-state citizenship. For example, there is no proper EU nationality,
nor EU citizen duties, and there is a relatively low level of citizen activity.
In sum, it is rights that play the key role with regard to EU citizenship.
These EU citizenship rights are put into practice within the EU’s mul-
tilevel system, and this fact impacts on the classical nexus between sover-
eignty and citizenship. It is no longer the nation-state alone that decides
upon who may enter and who may leave a territory. In the case of Union
Citizens, the EU member states have even lost this core competence
of sovereignty completely, as all Union Citizens have the right to Free
Movement within the EU (see below).
Another difference regards the political processes that were linked to
the institutionalisation of EU-related citizenship rights. They came about
in a very different shape than in the nation-states. Citizens or citizens-
to-­be did not actively fight for their rights, and there were no parliamen-
tary debates related to the creation of citizenship rights, at least until
rather recently (the European Parliament regularly debates on Union
Citizenship). EU-related citizenship rights were introduced and shaped in
a rather top-down manner by policy documents and laws drafted by the
EU Commission, voted upon by the Council (i.e. the EU member states
ministers), and interpreted by the Court of Justice of the European Union
(Wiesner 2007).
186  C. WIESNER ET AL.

4.5.2  The Institutional Context


The political struggles related to the creation, interpretation and imple-
mentation of EU citizenship rights are closely related to the EU’s insti-
tutional context and the ways of EU lawmaking and policymaking, and
they are to be interpreted against this background. There are, however,
different ways in which the laws in question have come about.
First, there are EU treaties. EU treaties are concluded between the
member states but have a constitution-like function: the treaties rule the
organisation of the EU as a polity. The treaties have changed over the
years. The first treaty was on the European Coal and Steel Community
(ECSC), concluded in 1951. It was followed by the Treaties establish-
ing the European Economic Community EEC (the Treaty of Rome)
discussed below. The first big reform treaty, the Single European Act
(SEA), was concluded in 1987. The SEA laid the basis for a series of
reform treaties in the 1990s and 2000s: the Maastricht Treaty in 1993,
the Amsterdam Treaty in 1999, the Nice Treaty in 2001 and the Lisbon
Treaty in 2009. The Lisbon Treaty contains most of the contents of the
draft Constitutional Treaty, which never was ratified after negative ref-
erenda in the Netherlands and France in 2005. In the cases prior to the
Lisbon Treaty, new treaties were voted upon by the European Council,
which consists of the EU’s heads of state and government. The European
Parliament did not have much of a say in treaty changes. Only in the
convention that drafted the Constitutional Treaty before 2005 (the
Constitutional Treaty being in most parts identical to the Lisbon Treaty)
were delegates from both the European Parliament and the national par-
liaments present for the first time.
Secondly, EU laws are based on the treaties. Formally, EU legislative
initiatives can only be issued by the European Commission. In practice,
both the Council, consisting of the Ministers of the member states, and the
European Parliament can exercise some influence on the Commission—
the Parliament can even cause it to propose laws. In the early days of
integration, EU laws were voted upon only by the Council, as it was the
sole legislative body. With the adoption of the Treaty of Maastricht, the
European Parliament obtained a few co-decision rights (meaning that
both Council and Parliament now had to vote upon legislation), which
were, however, only valid in a limited range of policy fields. Since the
Treaty of Lisbon, however, most laws are adopted via a co-decision by
EXAMPLES OF ANALYSING DEBATES AS POLITICS  187

both the Council and the Parliament. Apart from such laws, which are
called ‘directives’, the European Commission may issue directly binding
laws (‘regulations’).
Most of the legal acts discussed in the following did not come about
in co-decision procedures, except for one case. Hence, most of the
relevant laws were suggested by the European Commission and voted
upon by the Council. During most of the periods discussed here, meet-
ings of the Council were not even public (they are, however, public today
when they concern legislation). The last legal acts discussed in the follow-
ing were decided upon in 2004 in the co-decision of European Parliament
and Council.
To understand the political struggles related to EU laws, it also has to
be taken into account that they are put into practice by national laws, and
then executed by the member states’ governments as well as regional and
local institutions. Accordingly, there is some margin for different interpre-
tation of the laws as well as for political manoeuvring in their realisation.
The European Commission at this stage functions as the ‘guardian of the
treaties’, which means that it controls the correct implementation of the
law and can start infringement procedures against member states before
the Court of Justice of the European Union in case of noncompliance.
EU documents can be read as indicators of the interrelations and the
power struggles between the different EU institutions, as well as between
EU institutions and the member states, as will be outlined in the follow-
ing. Creating and implementing EU citizenship rights has been crucially
linked to these power struggles, and it also has been linked to a battle over
the interpretation and the implementation of the concept of citizenship.
This battle of EU ‘citizenship politics’ (Wiesner 2007, 23) has been car-
ried out by declarations, policy documents and laws.

4.5.3  
Citizenship Rights in the Treaty of Rome
The Treaty of Rome in 1957 not only founded the European Economic
Community, but also represented the starting point of the development
of EU-related citizenship rights, as it defined several rights concerning
the free circulation of persons and other rights concerning the economy
and the inner-EEC market (the European Economic Community exist-
ing until 1993 when it was renamed European Community, EC), among
188  C. WIESNER ET AL.

them the right to the free movement of employees and service providers
(Articles 48 and 59, Treaty of Rome, European Council 1957). Article
48 states:

1. Freedom of movement for workers shall be secured within the



Community by the end of the transitional period at the latest.
2. Such freedom of movement shall entail the abolition of any discrimina-
tion based on nationality between workers of the Member States as
regards employment, remuneration and other conditions of work and
employment.

Article 49 declares the task of proactively counteracting obstacles to


free movement in the future:

As soon as this Treaty enters into force, the Council shall, acting on a pro-
posal from the Commission and after consulting the Economic and Social
Committee, issue directives or make regulations setting out the measures
required to bring about, by progressive stages, freedom of movement for
workers, as defined in Article 48, in particular:
(b) by systematically and progressively abolishing those administrative pro-
cedures and practices and those qualifying periods in respect of eligibility for
available employment, whether resulting from national legislation or from agree-
ments previously concluded between Member States, the maintenance of which
would form an obstacle to liberalisation of the movement of workers; and
(c) by systematically and progressively abolishing all such qualifying periods
and other restrictions provided for either under national legislation or under
agreements previously concluded between Member States as imposed on
workers of other Member States conditions regarding the free choice of
employment other than those imposed on workers of the State concerned.

The right to equal payment for women and men is defined in Article
119 of the Treaty of Rome (European Council 1957):

Each Member State shall during the first stage ensure and subsequently
maintain the application of the principle that men and women should
receive equal pay for equal work.
For the purpose of this Article, “pay” means the ordinary basic minimum
wage or salary and any other consideration, whether in cash or in kind,
which the worker receives, directly or indirectly, in respect of his employ-
ment from his employer.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  189

Equal pay without discrimination based on sex means:


(a) that pay for the same work at piece rates shall be calculated on the basis
of the same unit of measurement and
(b) that pay for work at time rates shall be the same for the same job.

Several of these rights were necessary to realise regulations in the Treaty


or to guarantee minimum standards for the inner market, but these
­EEC-­related rights only concerned market participants: the right to free
movement, for example, was related only to market participation by work-
ing or by providing services in another member state, and it was part of
the four main freedoms (free circulation of goods, services, capital and
labour force) in the inner market. Hence these rights are not citizenship
rights in the proper sense, since they do not apply to citizens in their status
as citizens of a polity, but only in their status as market participants. The
rights did not apply outside the market and can hence be categorised as
creating a ‘market citizenship’.
Still, these rights, albeit limited in extent and applicability, created a
direct link between EU citizens (in their role as market participants) and
the EU, since they are directly applicable and EU citizens can claim them
before the Court of Justice of the European Union (CJEU). The CJEU
interpreted them in its judgements in such a way that, by and by, an EEC-
related catalogue of rights in the areas of work, healthcare and consumer
protection developed.
In the following years, the EU Commission and the CJEU developed
into agents of proactive non-discrimination policy by interpreting the
existing EU laws in the broadest possible way. A crucial case is here the
‘Defrenne’ case. In 1976, the Court of Justice of the European Union
(CJEU) judged that equal payment and equal treatment of women and
men had to be enacted in practice for individuals. A Belgian stewardess
named Gabrielle Defrenne thereby successfully claimed her right to equal
treatment with regard to retirement age and pension level. The Court
made a point of Article 119 being strictly applicable and having immedi-
ate consequences for the member states. It argued in its judgement that:

The principle that men and women should receive equal pay, which is laid
down by Article 119, may be relied on before the national courts. These
courts have a duty to ensure the protection of the rights which that provi-
sion vests in individuals, in particular in the case of those forms of discrimi-
nation which have their origin in legislative provisions or collective labour
190  C. WIESNER ET AL.

agreements, as well as where men and women receive unequal pay for equal
work which is carried out in the same establishment or service, whether pri-
vate or public. (European Court of Justice 1976, 481–482)

Several other rights that are related to market citizenship have been added
into the European treaties later on. The bigger part of these rights consist in
liberal rights like free market access and non-discrimination (see below). Since
then the EU has decisively contributed to developing new rights in the fields
of antidiscrimination and equal treatment. The rights that were thus created
were further reaching than most national citizenship rights, and what is more,
they had to be applied in the member states like all other EU legislation.

4.5.4  Creating and Interpreting Union Citizenship


The Treaty of Maastricht (1993) more than 35 years after the Treaty of
Rome introduced the ‘Citizenship of the Union’. It was the first time that
the concept ‘citizenship’ was used with regard to the EU and that hence a
key concept from nation-state representative democracy was applied in the
EU context. The expression ‘Citizenship of the Union’ has been explicitly
used since the Maastricht Treaty in the treaties as a headline (today to be
found in Articles 9 and 20, Treaty on the Functioning of the European
Union, TFEU, European Union 2010). The Maastricht Treaty in Articles
8 to 8e defines Union Citizenship rights. Article 8 states:

1. Citizenship of the Union is hereby established. Every person holding the


nationality of a Member State shall be a citizen of the Union.
2. Citizens of the Union shall enjoy the rights conferred by this Treaty and
shall be subject to the duties imposed thereby. (Council of the European
Communities and Commission of the European Communities 1992)

In the history of the concept of citizenship more generally, the key inno-
vation of EU citizenship is that political citizenship rights related to the
EU (and not only to member states) were created and applied. This is
obviously a major political innovation and a major political move, coming
about in a seemingly neutral way—that is, without any preceding parlia-
mentary or constitutional debate, without riots and fights for new rights,
and without much public interest, but simply in a new treaty ratified by
the heads of state and government of the EU-15 member states.
A closer look at the background of Union Citizenship rights shows
that these rights were not invented spontaneously in Maastricht, but
EXAMPLES OF ANALYSING DEBATES AS POLITICS  191

built upon ideas that had been developed over more than 20 years by the
Commission, the European Parliament and the Council and which were
finally adopted (Wiener 1998, 84). Thus the EU’s ‘innovating ideologists’
(Skinner 2002a, 148) had long developed their political ideas on creating
an EU citizenship, and then Maastricht was only a window of opportunity
for realising them. The background, again, relates to the strategic inter-
ests of the heads of state and government and the EU legitimacy crisis
of the 1990s. Weiler states that Spain’s Prime Minister Felipe Gonzalez,
in the very final stage of the Maastricht negotiations, suggested countering
the growing Euroscepticism in the population by ‘doing something about
citizenship’. The chapter on EU citizenship, based on the existing prepara-
tions, was then drafted in record time, says Weiler (Weiler 1996, 9).
Since the Maastricht Treaty came into force in 1993, all EU citizens,
in addition to their national citizenship, are citizens of the EU. One key
feature of EU citizenship is that it is derived from the possession of the
nationality of an EU member state. There is no proper EU nationality. EU
citizenship does not—and the treaties underlying state that it shall not—
endanger the respective member state nationalities, since the EU does not
interfere with their classical sovereign right of states to determine their
nationals:

Every national of a Member State shall be a citizen of the Union. Citizenship


of the Union shall be additional to and not replace national citizenship.
(Article 9, Treaty on the European Union, European Union 2010)

EU citizenship thus is a sort of hybrid compared to nation-state citizen-


ships, since it is based on the nationalities of the EU member states, but
creates EU-related citizenship rights, which intersect the former sovereign
rights of the states.
Despite this seemingly clear formulation in the treaties, the status of
EU citizenship became the object of an interpretative struggle between
the European Council and the Commission immediately after it had been
created. The Council stated in Birmingham in October 1992 that

Union Citizenship brings additional rights and additional protection for


our citizens, but in no respect takes the place of their respective nation-
ality … The nationality of a person will only be ruled by the respective
domestic laws of the member states. (European Union 2001, 2, translation
C. Wiesner)
192  C. WIESNER ET AL.

The Commission had another interpretation of the matter, as the fol-


lowing statement from an official Commission report on EU citizenship
underlines:

Citizenship of the Union is both a source of legitimation for the process


of European integration, by reinforcing the participation of citizens, and a
fundamental factor in the creation among citizens of a sense of belonging to
the European Union and of having a genuine European identity. (European
Commission 2001, 7)

The Commission furthermore highlights that EU citizenship concerns


both the EU and the member states:

Because of its origins and the rights and duties associated with it, citizenship
of the Union is sui generis and cannot be compared to national citizenship in
a member state. In this new type of multiple citizenship on different levels,
citizenship of the Union complements national citizenship and does not
replace it. (European Commission 2001, 7)

Union Citizenship rights today are defined in Articles 20–25 TFEU: the
right to free movement and residence all over the EU (but only for EU
citizens who can ensure their means for living), active and passive election
rights in  local and municipal as well as EP elections in their country of
residence, the right to diplomatic and consular protection outside the EU
by the embassies of all EU member states, the right to petition to the EP,
and the right to write in one of the 23 official EU languages to EU insti-
tutions and receive answers in the same language (Articles 20–25, TFEU,
European Union 2010).

4.5.5  Implementing Union Citizenship: The Example


of the Right to Free Movement
The institutional setting of Union Citizenship also was a novel one. After
it had been introduced in Maastricht, EU citizenship became part of the
ruling of the inner market, which in practical terms meant that all further
decisions in this field were subject to the co-decision of the Council and
the EP. As shall be sketched below, a number of such new decisions were
taken in the following years.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  193

The main reason behind those new legal acts is that putting EU citizen-
ship rights into practice has not occurred without problems. The analysis
reveals that a political struggle broke out on how the new rights were to
be interpreted and put into practice, as the new rights interfered with old
concepts, competencies and routines of the nation-states. So the imple-
mentation of Union Citizenship rights often led to political and concep-
tual struggles carried out via policy documents and laws, until finally a
new practice developed (see details in Wiesner 2007, 2014b). The related
struggles can be illustrated by looking at examples of the process in the
area of the free movement of persons.
The right to free movement proved a particularly conflicting issue when
it was put into practice. In 1997, 5.5 million EU citizens lived in member
states that were not their home countries. Political practice showed many
obstacles in the realisation of their right to free movement. The obstacles
were often related to unclear, incomplete or contradictory laws. Another
problem was that national, regional or local government institutions did
not facilitate the implementation, and in some cases they even hindered
it. When looking at the obstacles stated in an expert report issued in 1997
(European Commission 1997)4 it becomes obvious that the reactions of
the national authorities to the new rights can be interpreted as a kind of
political struggle around the interpretation and the implementation of the
new EU citizenship rights—and about the classical sovereign rights of a
nation-state.
Problems were related, for example, to the practice of residence per-
mits, the acceptance of foreign degrees, employment in the public sector,
and social and family-related rights.
As regards residence permits, to enter into another EU member state
EU citizens needed only a valid ID, but not for being resident there. A
residence permit was required if the EU citizen stayed for more than three
months. But there was no particular permit for stays between four months
and a year—theoretically EU citizens then needed to apply for a long-­
term residence permit. Many in this situation did not do so and hence
got into an irregular situation after three months. Students in most cases
could not apply for a residence permit that exceeded one year, so they
often had serial residence permits (European Commission 1997, 17–29).
Unemployed persons have a right to residence only when they do not claim
any social assistance in their country of residence (European Commission
194  C. WIESNER ET AL.

1997, 20)—a rule which is valid until today, even if the member states are
entitled to grant them social assistance if they decide it to be appropriate.
Another source of difficulties was local authorities. They often inter-
preted very restrictively the condition that EU citizens must prove their
means of livelihood. Students often had to hand in evidence, whereas the
EU laws only required a formal declaration. In cases where dealing with
these obstacles hindered the obtaining of a residence permit, it some-
times occurred that EU citizens were threatened with expulsion, or had
to endure difficulties in access to general services such as potable water,
electricity, heating or telecommunications.
To counter such problems and obstacles, the EU Commission first used
legal means. It led several infringement proceedings before the Court
of Justice of the European Union against member states. Most of these
proceedings changed the situation, but only after some years (European
Commission 2001, 10).
The Commission’s next move in the political struggle with the member
state authorities and governments was communication via policy docu-
ments. The Commission issued several documents and declarations that
claimed improvements had to be made regarding the situation of Union
Citizens living in EU member states and announced the changes to take
place. The Commission declared it wanted a uniform regulation of the
right to free movement and residence for all EU citizens and their family
members; it wanted to reduce the requirement for residence permits to
a minimum of cases, and it wanted a concretisation of the conditions of
expulsion (European Commission 2001).

The proposal for a Directive is a product of the legal and political envi-
ronment created by the introduction of citizenship of the Union. It takes
account of the results of the report of the high-level panel on the free move-
ment of persons, the Commission Communication on the follow-up to
the recommendations of the high-level panel, 12 the Second Report on
Citizenship, the European Parliament resolutions and the past rulings of the
Court of Justice. (European Commission 2001, 8)

The respective legislative initiatives were issued by the Commission until


2004. One decisive legal step was Directive 2004/38/EC (European
Union 2004), voted upon by the Council and the EP. As other EU direc-
tives in the area of EU citizenship rights (see below), it starts with a num-
ber of declarations aimed at clarifying aspects that have been the subject
EXAMPLES OF ANALYSING DEBATES AS POLITICS  195

of the political disputes before. It is interesting to note that the formal


authors of the directive are the European Parliament and the Council, that
is, not the Commission. But they clearly take on the role of agents who are
claiming to broaden and strengthen of EU-related citizenship rights in the
implementation stage:

(1) Citizenship of the Union confers on every citizen of the Union a pri-
mary and individual right to move and reside freely within the territory of
the Member States, subject to the limitations and conditions laid down in
the Treaty and to the measures adopted to give it effect.
(2) The free movement of persons constitutes one of the fundamental free-
doms of the internal market, which comprises an area without internal fron-
tiers, in which freedom is ensured in accordance with the provisions of the
Treaty.
(3) Union citizenship should be the fundamental status of nationals of
the Member States when they exercise their right of free movement and
residence.
(11) The fundamental and personal right of residence in another Member
State is conferred directly on Union citizens by the Treaty and is not depen-
dent upon their having fulfilled administrative procedures. (European
Union 2004)

The directive thus emphasises a number of keywords with regard to the


status of EU citizenship, naming it ‘a primary and individual right’ and a
‘fundamental status’, underlining that it is ‘conferred directly on Union
citizens’, rather than emphasising its dependence on the member states.
Among other things, the text of the law itself lays down common rules
of residence. It declares that EU citizens may stay in another member state
up to three months without a residence permit. After three months, mem-
ber states may require one, but then employees, self-employed persons,
students and their family members can require a permit on the basis of
their ID, working contract, certificate of training, proof of self-employed
status, proof of sufficient means for living and health insurance. The regu-
lation states that declaration is sufficient as a proof of livelihood, and no
fixed sum can be required. EU citizens can only be expulsed in cases of
severe threats to domestic security, with an exclusion for minors and per-
sons that have been resident for more than ten years in their respective
state. The regulation also requires member states to inform EU citizens of
their rights (European Union 2004).
196  C. WIESNER ET AL.

4.5.6  Shaping the Principle of Antidiscrimination


A field that is closely related to the right to free movement is antidiscrim-
ination. The principles of non-discrimination and equal treatment had
been first introduced in the Treaty of Rome, as explained above. But the
Treaty of Rome only had defined the right to equal payment for women
and men, and it had interdicted any discrimination based on nationality.
In the following years, however, the EU commission and the CJEU devel-
oped into agents of proactive non-discrimination policy by interpreting
the existing EU laws in the broadest possible way and thereby forcing
member states as well as employers to comply with the equal treatment
provisions of EU laws to the fullest extent possible. Policy documents,
laws and declarations again were their tools in this political struggle, as
explained above with regard to the Defrenne case (see also Wiesner 2007
for details).
After decades of active non-discrimination policies the Amsterdam
Treaty in 1999 considerably enlarged the legal basis and defined a political
goal of equal treatment and non-discrimination. In the version of the cur-
rent TFEU the respective article 19 now reads:

1. Without prejudice to the other provisions of the Treaties and within the
limits of the powers conferred by them upon the Union, the Council, act-
ing unanimously in accordance with a special legislative procedure and after
obtaining the consent of the European Parliament, may take appropriate
action to combat discrimination based on sex, racial or ethnic origin, reli-
gion or belief, disability, age or sexual orientation.
2. By way of derogation from paragraph 1, the European Parliament and
the Council, acting in accordance with the ordinary legislative procedure,
may adopt the basic principles of Union incentive measures, excluding
any harmonisation of the laws and regulations of the Member States, to
support action taken by the Member States in order to contribute to the
achievement of the objectives referred to in paragraph 1. (TFEU, Art. 19,
European Union 2010)

The Commission, on this newly founded legal basis, further defined the
principle of antidiscrimination, first in a communication. To enhance
antidiscrimination, the Commission suggested a complete policy pack-
age (European Commission 1999). It mainly consisted of a proposal of
two EU-directives that should put the article cited above into ­political
EXAMPLES OF ANALYSING DEBATES AS POLITICS  197

­ractice. Referring to the principles of freedom, democracy, human


p
rights, rule of law, as well as the four freedoms of the inner market, the
following statement underlines very clearly the Commission’s desire to
create a political practice that was wider-reaching than in the member
states—and it wanted to force the member states to change their legal
and political practices:

In more general terms, a number of Member States do not have com-


prehensive legislation (other than isolated provisions contained in various
labour or criminal codes) to combat discrimination on the grounds of racial
or ethnic origin, religion or belief, age and sexual orientation. (European
Commission 1999, 4)

The Council, acting alone (the law project not being subject to co-­
decision), followed the Commission’s initiative and voted two new
­antidiscrimination directives in 2000. To start with the second, Council
Directive 2000/78/EC of 27 November 2000 is more concrete. It inter-
dicts any discrimination on ethnic grounds and establishes equal treat-
ment in employment, education, social security and access to products
and services (European Council 2000b). The first directive, Council
Directive 2000/43/EC of 29 June 2000, is more far-reaching. It inter-
dicts any discrimination in employment and profession that is based on
race, ethnic origin, religion, ideology, age or sexual orientation (European
Council 2000a).
This general antidiscrimination directive also is a clear political state-
ment. The first part consists of a declaration listing 28 arguments for why
the Council and the EP adopted the directive. The list is very exhaustive
and includes not only several value-oriented statements and reasons in
favour of the EU’s non-discrimination policies, but also future perspec-
tives. Lastly it makes up a strong link between the EU and concepts and
ideas that are charged with positive connotations. Among the 28 argu-
ments there is a reference to the values of the European Union:

(2) In accordance with Article 6 of the Treaty on European Union, the


European Union is founded on the principles of liberty, democracy, respect
for human rights and fundamental freedoms, and the rule of law, principles
which are common to the Member States, and should respect fundamen-
tal rights as guaranteed by the European Convention for the protection of
198  C. WIESNER ET AL.

Human Rights and Fundamental Freedoms and as they result from the con-
stitutional traditions common to the Member States, as general principles
of Community Law.

There is also a reference to other international Treaties and the Human


Rights:

(3) The right to equality before the law and protection against discrimina-
tion for all persons constitutes a universal right recognised by the Universal
Declaration of Human Rights, the United Nations Convention on the
Elimination of all forms of Discrimination Against Women, the International
Convention on the Elimination of all forms of Racial Discrimination
and the United Nations Covenants on Civil and Political Rights and on
Economic, Social and Cultural Rights and by the European Convention for
the Protection of Human Rights and Fundamental Freedoms, to which all
Member States are signatories.

Later in the list of arguments, several future-oriented declarations


follow, which underline that other measures and policy initiatives are in
planning:

(12) To ensure the development of democratic and tolerant societies which


allow the participation of all persons irrespective of racial or ethnic origin,
specific action in the field of discrimination based on racial or ethnic ori-
gin should go beyond access to employed and self-employed activities and
cover areas such as education, social protection including social security and
healthcare, social advantages and access to and supply of goods and services.
(24) Protection against discrimination based on racial or ethnic origin would
itself be strengthened by the existence of a body or bodies in each Member
State, with competence to analyse the problems involved, to study possible
solutions and to provide concrete assistance for the victims.

The directive itself then defines a far-reaching ‘framework for combating


discrimination on the grounds of racial or ethnic origin, with a view to
putting into effect in the Member States the principle of equal treatment’
(Article 1).
Some member states were very reluctant in transferring these directives
into national legal practice. Germany was one of the last states to do so,
already under pressure of an infringement procedure before the Court of
Justice of the European Union (CJEU).
EXAMPLES OF ANALYSING DEBATES AS POLITICS  199

4.5.7  In Conclusion
The processes and conflicts related to the introduction, interpretation and
implementation of EU citizenship rights, as well as the interests of the
agents, are clearly readable in the studied documents. The analysis has
shown the EU Commission as the major driving force behind the creation
and also the implementation of the new EU-related citizenship rights. It
can be concluded that it is in the Commission’s interest that EU citizenship
rights be well-accepted. In relation to the institutional and conceptual con-
text sketched earlier, one reason is that, by enlarging EU-related citizenship
rights and breaking up the members state’s previous monopoly on them,
the EU was strengthened and hence also the Commission as its de facto
government. Moreover, strengthening EU citizenship rights in that sense
can be interpreted as symbolically creating subjects for the Commission.
The Commission has been seconded by the Court of Justice of the
European Union, which not only led infringement proceedings where
the new rights were not satisfactorily complied with, but also interpreted
the new rights as broadly as possible in its judgements.
The member states’ governments in their majority voted the new rights
in the European Council and the Council, seemingly ignorant of the fact
that these would curtail their sovereign powers. On the other hand, the
lower levels of the governments and state authorities were often reluctant
when it came to implementation.
The political struggles around the interpretation and implementation
of the new rights were hence apparently not carried out before drafting the
laws and voting on them. They seem to have been carried out mainly after
the laws had been decided upon and during the implementation phase.
This is partly related to the actors concerned by the implementation pro-
cesses, that is, the national executives, which are not included in drafting
the respective laws and accordingly can only intervene in the course of
their implementation.
Moreover, the agents that are the most concerned by the new rights,
the EU’s citizens, were not included in shaping the new rights. EU citi-
zens have just benefited from the rights, without having fought for them.
All in all, the processes sketched above are far from being unpolitical or
free of conflict. They may have taken place mainly behind closed doors,
and in texts and documents rather than in public debates, but they are
vivid, and they are ongoing. In the first half of 2016, the refugee ‘crisis’ is
one crucial example.
200  C. WIESNER ET AL.

4.6   What Is Said and What Is Not in Press


Debates: The Silencing Strategy in Germany
The following example is taken from research in the exemplar study that
was sketched in Chapter 3, the comparative analysis of the German and
the French press discourse on the draft Constitutional Treaty that pre-
ceded the ratification vote in the German Bundestag and the French EU
referendum in 2005. This example underscores that in a debate, it is not
important only what is said, but also what is not being said.
The point of ‘what is not being said’ is first based on the general
assumption of discourse analysis that every discourse is structured accord-
ing to certain rules that determine what can be said and what will be
ascribed meaning. Second, it follows the idea of Michel Foucault that in a
discourse, one should ask why some special utterance is made rather than
another one (see Chapter 3 for details).
Following along these lines, it is reasonable to assume that discourses
can also show rules that determine what cannot be said, and what cannot
be ascribed meaning.
The research design, sources, material selection and proceedings that
are the basis for the results presented below have already been sketched out
in Chapter 3. The quotations are German in their original; therefore, they
have been left, but an English translation by the researcher (CW) has been
added for the purpose of comprehension. The numbers and abbreviations
at the end of the German quotes refer to the journals they are retrieved
from and the dates they were published (SZ is Süddeutsche Zeitung, FAZ is
Frankfurter Allgemeine Zeitung, ND is Neues Deutschland and taz is taz;
thus SZ 120505_3 refers to the third article retrieved from Süddeutsche
Zeitung on 12 May 2005).
The main findings with regard to the German press discourse that pre-
ceded the German ratification vote in the Bundestag as well as the French
EU referendum in 2005 have also been outlined in Chapter 3. The dis-
course was European and open; the EU as well as EU neighbours were
addressed as important levels of reference; and the discourse was mainly
driven by politicians from the EU, the EU member states and Germany.
Besides those characteristics, the German discourse also was marked by
what was not said, or more precisely, by a rhetorical strategy that served to
silence EU criticism and EU critics in the discourse, which the researcher
(CW) accordingly has termed the silencing strategy.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  201

The silencing strategy relates to the deeply rooted consensus of most


mainstream parties, usually termed the ‘elite consensus’, in favour of EU
integration, a consensus that has marked German EU politics and, accord-
ingly, the very positive and supportive German conception of the EU (see
details in Wiesner 2014a, 300–307, 319–333). Most German political
parties represented in the Bundestag agree to a principled support of EU
integration, with the exception of the Left Party and its predecessor, the
PDS (the Party of Democratic Socialism), and quite recently the newly
founded AfD (Alternative for Germany), which is not represented in the
Bundestag, but in several of the Länder parliaments. This consensus of
the Social Democrat Party (SPD), the Greens, the Free Democratic Party
(FDP) and the Christian Democrat Parties, CSU (the Bavarian part of
German Christian Democrats, Christian Social Union) and CDU, in
favour of European integration (see Wiesner 2014a) illustrates that, in
contrast to France and many other EU member countries, European
integration and hence a post-nationalist orientation have become the
framework and a guiding principle of political action for a majority of
post-war German politicians. This positive German EU conception is not
only deeply rooted in German political culture, but also laid down in the
German Constitution, the Grundgesetz (Basic Law), where both the pre-
amble and Article 23 explicitly state that EU integration is a national goal.
With regard to the context of the 2005 press discourse it is important
to notice that Germany was governed by the so-called red-green coalition,
that is, a coalition that united the SPD and the Green Party. The FDP as
well as CDU and CSU were in the opposition.
The relation between the two Christian Democrat Parties is important
for what is sketched in the following: they are called ‘sister parties’—that
is, they are formally different parties, but traditionally form a joint group
in the Bundestag. It is also traditional that CSU is only represented in
Bavaria and in no other federal state, while the CDU is represented in all
other federal states except Bavaria. CSU in the German political culture
is considered a kind of peculiarly Bavarian party that underlines Bavaria’s
difference from the other federal states. CSU members as well as CSU
leaders usually are much more inclined to EU criticism than is the case in
the CDU.
In the 2005 run-up to the ratification vote, it was apparent in the
German discourse that the pro-EU majority of German politicians did not
202  C. WIESNER ET AL.

react to the EU or Constitutional Treaty criticism by distancing themselves


from it, or by counterarguments, but by downplaying the criticism and the
critics—in other words: by silencing both the criticism and the critics in the
discourse. Three patterns of silencing appeared. First, there were osten-
sibly diverging opinions in the bigger mainstream parties, all of which,
however, supported the elite consensus in principle, that is, mainstream
party leaders all argued in favour of the EU and the Treaty. Such EU crit-
ics as could be found were a considerable number in the CSU. Secondly,
in 2005, the PDS was the only party represented in the Bundestag that
as a party position openly took a critical position on the Constitutional
Treaty. And third, a prominent member of the Christian Social Union,
Peter Gauweiler, stood out from the other cases, as he is an important
individual political player who has put forward numerous constitutional
complaints against EU integration over the years.

4.6.1  Case 1: Marginalising EU Critics in the Mainstream


Parties
In 2005, CSU MPs put forward a relatively pronounced EU criticism. In
part this can be attributed to the fact that the party was in the opposition
at the time. This enabled the party and its leaders and MPs to be less strict
about showing support for EU integration.
Already in January 2005, when the MPs of the CSU Bundestag group
held their famous annual reunion in Wildbad Kreuth in the Bavarian Alps,
several MPs criticised the draft Constitutional Treaty and announced that
they would only vote in favour of ratification in the Bundestag if some
conditions were met in the ratification law, as for example more powers for
the Bundestag in EU matters. As an immediate reaction to this criticism,
both CSU Chair Edmund Stoiber and president of the Christian Social
Bundestag Group Michael Glos publicly neglected the criticism, sending
the signal that it was not to be taken seriously. At this point no debate on
the Constitutional Treaty had yet taken place within the CSU Bundestag
group, and it was not clear how many MPs actually would vote against
the Treaty, but still, both politicians emphasised there would be a positive
vote in the group:

CSU-Vorsitzender Edmund Stoiber rechnet mit einer breiten Mehrheit in


der CSU-Landesgruppe für die neue EU-Verfassung. Es würden letztlich
“nur sehr wenige” CSU-Abgeordnete im Bundestag mit Nein stimmen […]
EXAMPLES OF ANALYSING DEBATES AS POLITICS  203

Auch CSU-Landesgruppenchef Michael Glos rechnet mit einer ­einheitlichen


Haltung der CSU-Parlamentarier. Dies könne er allerdings “nicht für 100
Prozent garantieren”, schränkte Glos ein. (SZ 070105)
[CSU chairman Edmund Stoiber calculates on obtaining a broad major-
ity for the Constitutional Treaty in the CSU Bundestag group. In the end
only “very few” MPs would vote No, Stoiber said … CSU Bundestag
group Chair Michael Glos is counting on a united position by the CSU
MPs, even if he “cannot guarantee 100 % approval”, he then qualified.]

Despite such comments that downplayed the importance of inner-party


EU criticism, both Stoiber and Glos in the coming months adopted the
claims that had been put forward by the EU-critical CSU MPs. They
negotiated with the German red-green government until shortly before
the Bundestag ratification vote. The government finally accepted some of
the claims—but it did not publicly comment on them.
Despite the compromises the government had thus made, most of the
CSU EU critics announced they still wanted to vote No. In reaction, the
chairs of the CDU and CSU Bundestag group tried by several rhetorical
and political means to persuade them to join the Yes camp.
As a first step, the chairs tried to change the opinions of potentially
dissenting MPs by argument and by threat. In the group meeting that
preceded the Bundestag vote, a trial vote was held, with both Edmund
Stoiber (CSU chair) and Angela Merkel (CDU chair) speaking, making a
plea in favour of a united Europe and the Constitutional Treaty:

Auch die Kirche sehe die Verfassung als Fortschritt an, erklärte Stoiber den
EU-Skeptikern aus den eigenen Reihen—von denen manche den fehlen-
den Gottesbezug in der Verfassung beklagen. Im wahrsten Wortsinn sollten
die Kritiker nun, soll Stoiber gemahnt haben, nicht päpstlicher sein als der
Papst … Mit Pathos rief Bayerns Ministerpräsident in Erinnerung, wie
das kriegerische Europa von einst zu einer Friedensgemeinschaft zusam-
mengewachsen sei. CDU-Chefin Merkel erklärte, warum die Verfassung ein
großer Fortschritt sei. Ihr Vortrag sei imponierend gewesen, räumen selbst
EU-Skeptiker ein. (SZ 120505_3)
[Even the Catholic Church was regarding the Constitution as a progress,
Stoiber explained to the EU sceptics in his own ranks—some of them had
been complaining a reference to God was missing in the Treaty. Stoiber, as
was reported, added that the critics should not be more papal than the pope
himself […] Using a lot of pathos, the Bavarian Minister-President recalled
how an old warring Europe grew together into a new community of peace.
204  C. WIESNER ET AL.

CDU chair Merkel explained why the Constitution was a big step ahead.
Her presentation was impressive, even EU sceptics said.]

In addition, the critics’ arguments were taken up by CSU party and group
leaders. They explained the Constitutional Treaty, stressed the compro-
mises that had been made, and the CSU claims that had been accepted,
going so far as to sign an official declaration:

Bei der CSU hat die Führung der Landesgruppe EU-Skeptikern eine Brücke
gebaut. In einer Erklärung, die auch Landesgruppenchef Michael Glos
unterzeichnet hat, drücken die Skeptiker ihr Unbehagen aus—um dann
doch “trotz der schwerwiegenden Bedenken” zuzustimmen. Als wichtig-
stes Argument für die Zustimmung gelten die zusätzlichen Rechte, die dem
Bundestag im Gesetzgebungsverfahren eingeräumt werden. (SZ 120505_3)
[In the CSU the leaders of the Bundestag group have been build-
ing bridges for the EU sceptics. In a declaration that was also signed by
group chair Michael Glos, the critics express their discomfort—adding that
they nevertheless will, despite their criticism, approve of the Treaty. The
main argument for approval consists in the additional rights given to the
Bundestag in legislative procedures.]

In addition to this strategy of compromises and explanation, threats were


expressed:

So berichten Abgeordnete, dass ihnen von einflussreichen Kollegen jene


Folterinstrumente gezeigt wurden, die in der Demokratie zum Zwecke
der Mehrheitsbeschaffung entwickelt wurden—etwa die Drohung mit dem
Karriereknick. (SZ 120505_3)
[MPs report that influential colleagues have been brandishing the kinds
of instruments of torture developed in democracies that serve at creat-
ing majorities—for instance, the threat that one’s career might eventually
suffer.]

When it became apparent that despite all those efforts a minority in the
CSU group would vote No after all, this fact was downplayed, as in the
following quote by Angela Merkel:

“Die große Mehrheit wird zustimmen, und das ist auch gut so.“… Sie
werde ihrer Fraktion ein Ja zur EU-Verfassung empfehlen. Dass diese dann
geschlossen für die Ratifizierung stimmt, erwarte sie jedoch nicht. Es sei
seit langem bekannt, dass es Gegenstimmen geben werde. (taz 100505_3)
EXAMPLES OF ANALYSING DEBATES AS POLITICS  205

[“A vast majority will approve [of the Treaty], and this is good news” …
She said she would recommend a Yes vote to her group [the CDU group],
but did not expect a unanimous vote. It was known for a long time, she said,
that there would be some votes against.]

This strategy of downplaying the impact of the EU critics within the


group was combined with new offers to the potentially dissenting MPs, as
becomes apparent in the following quote by the first executive secretary of
the CDU group, Norbert Röttgen:

[Er] sagte, wenn am Ende beispielsweise ein Abgeordneter erkläre, er könne


dem Verfassungsvertrag wegen des fehlenden Gottesbezugs nicht zustim-
men, werde ihm niemand mit einem Fraktionszwang drohen. Röttgen
argumentierte überdies, es werde schon deswegen keine “Gruppe” von
Zustimmungsverweigerern bei der Verfassungs ratifizierung geben, weil die
Begründungen für das Nein sehr heterogen seien. (FAZ 110505)
[He said that if in the end one MP declared he or she could not vote Yes
because God was not mentioned in the Treaty, nobody would threaten him
or her with the whip. But there was no unified group of opponents, as the
reasons behind the potential No vote were so divergent.]

The 2005 government parties, the Social Democrats and the Greens, also
included some critics of the Constitutional Treaty, but there were much
less of them than in the ranks of CSU.  Still, the group leaders of the
Greens and the SPD tried to discipline these critics. In the SPD, the reac-
tion was to downplay the matter:

In der SPD haben zwei Abgeordnete angemeldet, sich bei der Abstimmung
enthalten zu wollen. Der parlamentarische Geschäftsführer der SPD-­
Fraktion, Wilhelm Schmidt, kündigte allerdings an, man wolle mit den
beiden noch einmal sprechen. (SZ 120505_3)
[Two SPD MPs have announced they would abstain from the vote.
The executive secretary of the SPD group, Wilhelm Schmidt, however
announced he would talk to both of them once again.]

In contrast, there was an intense debate in the Green group. The only
Treaty critic, Hans-Christian Ströbele, came under harsh criticism:

Bei den Grünen meldete in der Fraktionssitzung nur Hans-Christian


Ströbele Bedenken an—was ihm immerhin einen gut 20-minütigen
Meinungsaustausch mit Außenminister Joschka Fischer bescherte … Sollte
206  C. WIESNER ET AL.

Ströbele seine Zustimmung verweigern, so Fischer, müsse er sich auch


darüber bewusst sein, dass er sich mit den Europa-Skeptikern unter den
britischen Konservativen und auch den französischen Rechtsextremen
gemein mache. Mit dem Hinweis auf Letztere griff Fischer Zwischenrufe
der Parteivorsitzenden Claudia Roth auf, die wiederholt den Namen Le Pen
in den Raum warf … In Grünen-Kreisen wird erwartet, dass Ströbele der
Verfassung nun zustimmt. (SZ 120505_3)
[In the Green group meeting, only Hans-Christian Ströbele expressed
his concern—which resulted in a 20 minute exchange with Foreign Minister
Joschka Fischer … Should Ströbele not approve, Fischer said, he needed
to be aware that he was joining forces with Eurosceptics like the British
Conservatives and the French extreme right. In referring to the latter, Fischer
was reacting to interjections of Green Chair Claudia Roth, who repeatedly
mentioned the name Le Pen … Green insiders expect that Ströbele will now
approve of the Constitutional Treaty.]

4.6.2  Case 2: The PDS
While in the CSU as well as in the governing red and green parties Treaty
critics were in the minority, the situation was different for the PDS. It did
break with the German elite consensus in favour of integration as a party
strategy.
It has to be underlined that the quotations in this part are not taken
from one of the centre journals previously quoted, Süddeutsche Zeitung
or Frankfurter Allgemeine Zeitung, but from the party journal of East
Germany’s former Socialist Unitary Party (SED), Neues Deutschland
(New Germany), which is nowadays close to the Left Party and was close
to its predecessor PDS in 2005. The centre journals studied did not
report on PDS activites extensively. Their reluctance to publish extended
reports on PDS can also be regarded as part of the silencing strategy of the
German political centre: without the strategy being outspoken, explicit
EU-criticism de facto is not given an arena.
The PDS strategy was concluded at a party convention in January 2005:

Es gibt eine gemeinsame PDS-Haltung zu dem in Rom unterzeichneten


EU-Verfassungsvertrag. Diese Haltung wird ausschlaggebend bestimmt von
der Ablehnung zweier Grundelemente des Vertrags: a) Aufrüstungspflicht
und Militarisierung der Außenpolitik, b) Festlegung von Marktradikalismus
als obligatorischer Wirtschaftsordnung. Beides widerspricht sowohl
demokratisch-sozialistischen Überzeugungen über eine zukunftsfähige
Entwicklung Europas als auch dem Grundgesetz für die Bundesrepublik
EXAMPLES OF ANALYSING DEBATES AS POLITICS  207

Deutschland. Folgerung: NEIN zu diesem Vertrag in seiner gegenwärtig


vorliegenden Fassung. (ND 080105)
[There is a joint PDS position regarding the draft Constitutional Treaty
that was signed in Rome. This position is decisively influenced by a dedi-
cated rejection of two core elements of the Treaty: a) a duty to armament
and militarisation and, b) the fixation of market radicalism as the obligatory
economic order. Both of these goals contradict democratic socialist convic-
tions regarding the future sustainable development of Europe, as well as the
German Basic Law. In Consequence: NO to this treaty in its current version.]

The PDS did follow this strategy during the whole of the ratification process.
But, similarly to the other parties, PDS was not left without i­nternal con-
flicts. In addition, it experienced conflicts with regard to coalition partners.
The internal conflicts first showed when the PDS MEPs held a test
vote in January. The seven PDS MEPs did not unanimously vote No,
and in particular, Sylvia-Yvonne Kaufmann even voted Yes—but she quit
the PDS shortly afterwards because of her dissenting opinions on EU
matters.
In the Bundestag, matters went more easily, though the PDS had only
two MPs at that time. The two, Gesine Lötzsch and Petra Pau, announced
their No vote already during the first reading of the ratification law in May.
And indeed they did vote No, arguing as follows (quote once again from
Neues Deutschland):

Die PDS-Abgeordnete Gesine Lötzsch erläuterte die Ablehnung


der Verfassung durch die PDS mit der vom Bundestag verweigerten
Volksabstimmung wie mit inhaltlichen Gründen. Die Verfassung schreibe
den Weg für militärische Konfliktlösungen fest und setze auf freien Markt
statt soziale Marktwirtschaft. (ND 130505_6)
[PDS MP Gesine Lötzsch explained her rejection of the Constitution by
referring to the Bundestag refusal to hold a referendum on the matter, and
also for political reasons. The Constitution was establishing a military way
of conflict resolution and a free-market economy instead of a social-market
economy, she said.]

While the situation was comparatively easy in the Bundestag, where the
two PDS MPs were in the opposition, the PDS also was part of a coalition
government with the SPD in the federal states of Berlin and Mecklenburg-­
Vorpommern. This fact mattered with regard to the second part of the
ratification vote, which was to be cast in the Bundesrat, the parliamen-
tary chamber that represents the federal states. Traditionally, coalition
208  C. WIESNER ET AL.

­ overnments abstain from voting in the Bundesrat if they diverge on a


g
matter. However, in this case, the SPD-led governments of both Berlin
and Mecklenburg-Vorpommern planned to approve the Constitutional
treaty in the Bundesrat vote.
While the vote approached, the two governments reacted differently.
In Mecklenburg-Vorpommern the conflict was simmering in the run-up
to the Bundesrat vote. On the other hand, in Berlin the PDS-board and
a party convention opened the way for a compromise. In an official dec-
laration they underlined that the PDS was opposed to the Treaty in prin-
ciple, but added that this opposition should not threaten the coalition
­government. In consequence, PDS would not force its coalition partner
SPD to abide by the dictates of the coalition agreement, that is, abstain
from Bundesrat votes in the case of conflict.
This PDS position, however, was only adopted after an intensive
debate at a party convention. While Treaty opponents invited delegates to
vote for a resolution calling on the Berlin government to vote No in the
Bundesrat, the party board proposed an amendment:

Die PDS Berlin begrüßt, dass in Abstimmung mit dem Parteivorstand


Einigkeit in der Frage herrscht, dass ein “Nein zur Verfassung” kein “Nein
zur Koalition” bedeutet. Wir erwarten daher von den Abgeordneten der
PDS-Fraktion im Abgeordnetenhaus und den Senatoren mit PDS-Mandat,
einer Ratifizierung der Verfassung nicht zuzustimmen, dies jedoch nicht zur
Koalitionsfrage zu erheben. (ND 020505_3)
[Berlin chapter of PDS welcomes the fact that the party and party board
agree on the view that a No to the Constitutional Treaty does not mean
No to the coalition. Therefore we expect the Berlin PDS MPs as well as the
PDS government members not to approve of the Constitutional Treaty, but
also not to make this matter a question of keeping or leaving the coalition.]

In the end, the party board amendment obtained a narrow majority at the
convention: 52 delegates voted against, and 49 voted in favour.
In Mecklenburg-Vorpommern, on the contrary, the PDS did refuse
such a compromise. It threatened to end the coalition with the SPD
should SPD Minister-President Ringstorff approve of the Treaty in the
Bundesrat vote:

Man habe ein “gesteigertes Interesse” daran, die Koalition zu einem


erfolg­reichen Ende zu führen, sagte PDS-Landesvorsitzender Peter Ritter.
Allerdings brauche es dafür auch verlässliche Arbeitsgrundlagen wie das
EXAMPLES OF ANALYSING DEBATES AS POLITICS  209

Einhalten des Koalitionsvertrages. Da es seit Februar keine inhaltliche


Annäherung zum Thema EU gegeben habe, erwarte die PDS entspre­
chendes Abstimmungsverhalten von Ministerpräsident Ringstorff (SPD).
Sollte er dennoch mit “Ja” stimmen, … würde eine Entscheidung über das
Fortführen der Koalition zu treffen sein. (ND 270505)
[PDS is very interested in successfully preserving the coalition, PDS chair
Peter Ritter said. However, this required a solid base, including sticking to
the provisions in the coalition agreement. As the coalition partners have not
been able to agree on EU matters since February, the PDS expects Minister-­
President Ringstorff to vote as the coalition agreement says. Should he
nevertheless vote Yes […] a decision on whether to maintain the coalition
would have to be taken.]

In the end, Ringstorff gave in and abstained from the Bundesrat vote, argu-
ing that he wanted to save the coalition. Interestingly enough, this was the
first occasion for the centre journals to take up the issue of the PDS position
in the matter—maybe because of the intervention of Chancellor Schröder:

Bei einem Besuch in Mecklenburg-Vorpommern einen Tag vor der


Bundesratsentscheidung hatte Kanzler Schröder Ringstorff offenbar freie
Hand bei seiner Entscheidung gegeben. (FAZ 280505_14)
[Visiting Mecklenburg-Vorpommern one day before the Bundesrat vote,
Schröder apparently gave Ringstorff carte blanche.]

In sum, it can be said that the PDS had partial success in breaking the
German elite consensus for EU integration.
First, the oppositional strategy worked where the PDS was not in a gov-
ernmental role (as in the Bundestag), but not where it was part of a coali-
tion, for then it became difficult to keep up the oppositional position when
a coalition partner was in favour of the Treaty. The problem consisted
not so much in issuing public pronouncements of the ­oppositional posi-
tion (SPD coalition partners apparently grudgingly accepted this move);
rather, the main conflict related to differences in the strategic goals of the
PDS—while in the position of governing in a federal state, goals that it
did not want to put in jeopardy for the sake of an EU criticism, which was
considered less important, at least by party leaders, than participation in
the government.
Second, there were dissenting voices within the PDS as well. But Sylvia-­
Yvonne Kaufmann, the most prominent, was severely criticised for her
support of the Treaty.
210  C. WIESNER ET AL.

Third, and most decisively, the EU criticism did not have any discur-
sive success outside from the arena of Neues Deutschland. On the level
of the federal state, no politician officially reacted to the Mecklenburg-­
Vorpommerian abstention in the Bundesrat, and there had not been any
reactions to the conflict in the weeks before either. Minister-President
Ringstorff downplayed the impact of the vote:

Mecklenburg-Vorpommerns Ministerpräsident Harald Ringstorff (SPD)


[…] will der Enthaltung seines Landes keine große Bedeutung beimessen.
Es sei ein “Schönheitsfehler”, … sagte er. Wichtig sei, dass man eine Krise in
der Koalition abgewendet habe. (taz 280505)
[Mecklenburg-Vorpommerian Minister-President Harald Ringstorff
(SPD) does not want to credit the abstention with having a serious impact.
It was merely a blemish, he said. The important thing was that a coalition
crisis was averted.]

As has been said above, the centre journals rarely reported on PDS activi-
ties in the matter, and thus they gave only passing mention to the conflicts
when at all. In consequence, the active breach of the elite consensus was
silenced or downplayed in the German discourse as a whole, except for the
journal close to PDS.

4.6.3  
Case 3: Peter Gauweiler
Peter Gauweiler is a special case that cannot easily be compared with
the dissenters in the other mainstream parties. He is a prominent CSU
member, held different mandates and positions, and was a Bundestag MP
2002–2015. Gauweiler lodged different constitutional complaints against
EU integration, one of which was against the Constitutional Treaty’s rat-
ification in 2005. First he filed an urgent complaint, insisting that the
Constitutional Court stop the Bundestag from voting on the ratification,
and then, when the ratification law had been voted upon, he filed a con-
stitutional complaint against the law. Different from the other CSU dis-
senters, thus, and comparable to the PDS, Gauweiler openly broke with
the elite consensus.
But again, there was barely any reaction to his activities, the silenc-
ing strategy worked quite well: the government was keeping quiet, show-
ing no reaction whatsoever, while the Christian Democrat Party officials
underplayed the matter and also the importance of Peter Gauweiler. This
became apparent already in the reactions to his first complaint:
EXAMPLES OF ANALYSING DEBATES AS POLITICS  211

Elmar Brok, EU-Abgeordneter der CDU und im EU-Verfassungskonvent


der Vorsitzende aller europäischen Christdemokraten, nannte das Vorgehen
Gauweilers ‘Unsinn’ … “Das ist der Kampf eines in CDU und CSU iso-
lierten Anti-Europäers gegen eine Verfassung, die die Rolle des EU-Bürgers
und das Gewicht der nationalen Parlamente in Europa erheblich stärkt” …
Auch der Vizepräsident des europäischen Parlaments und stellvertretende
CSU-Vorsitzende Ingo Friedrich glaubt nicht an einen Erfolg der Klage
Gauweilers … CSU-Landesgruppenchef Michael Glos wertete Gauweilers
Vorstoß als ‘eigene Initiative’ des Abgeordneten. (SZ 220405_3)
[Elmar Brok, CDU MEP and chair of the Christian Democrats in the
Constitutional Convention, called Gauweiler’s strategy ‘nonsense’ … “This
is the battle of an anti-European who is isolated in both the CDU and the
CSU. He is fighting against a constitution that will considerably strengthen
the role of EU citizens and national parliaments in Europe” … EP vice
president and CSU vice-chair Ingo Friedrich does not believe in the suc-
cess of Gauweiler’s complaint … CSU-group chair Michael Glos judged
Gauweiler’s complaint to be an ‘individual initiative’ of the MP].

Gauweiler’s later actions against ratification were followed by similar reac-


tions. As an inner-party dissenter, Gauweiler did not receive speaking
time in the Bundestag debate on the Treaty. He used the only one option
­available: he made a personal declaration after the debate. Most interesting
were the press reactions to his complaint against the ratification law. All
of the journals studied reported on the complaint, but most gave it only
a brief sidebar rather than a longer article. The only exception was the
left-oriented taz, which published an interview with Gauweiler and SPD
dissenter Hermann Scheer.
Gauweiler’s second constitutional complaint, however, proved quite
influential. German President Horst Köhler put the law’s ratification
process on hold pending a Constitutional Court decision on the matter.
This meant that the German ratification of the Constitutional Treaty was
effectively suspended, which carried strong symbolic value in the heated
atmosphere after the French Non and the Dutch Nee in their respec-
tive referenda. But both Chancellor Gerhard Schröder and CSU Chair
Edmund Stoiber made only brief comments:

Schröder bekräftigte im Bundestag: ‘Deutschland hat ja gesagt und nicht


nein’. Man dürfe diese Entscheidung nicht neu interpretieren. Der CSU-­
Vorsitzende Stoiber, […] sagte, es sei ‘nur eine Frage der Zeit’, bis Köhler
den Verfassungsvertrag unterzeichnen werde. (FAZ 170605)
212  C. WIESNER ET AL.

[Schröder said in the Bundestag: ‘Germany has said yes, not no’. One
should not reinterpret the decision now. CSU chair Stoiber […] said, it was
‘only a matter of time until Köhler signed the ratification measure into law’.]

4.6.4  In Conclusion
The examples sketched have demonstrated how the German press dis-
course was marked by a silencing strategy that influenced what was said,
what was not said and how the utterances were received in the wider
discourse and whether they had an effect. The strategy proved quite effec-
tive as it either directly stopped EU-critical comments, or helped to mini-
mise or ignore completely that such comments were made at all in the
discourse.
The silencing strategy worked in three steps:

(1) EU criticism was silenced (there was no reaction to it), or it was mar-
ginalised (the major journals did not publish more than 10 lines).
(2) EU critics and EU criticism were downplayed (‘one cannot take this
seriously’, ‘this is not important’, ‘he/she does not have any
influence’).
(3) When all else failed, the party and group leaders used disciplinary
means and threats, be it against group or party members, or against
coalition partners.

The strategy should not be interpreted in moral terms, that is, the fact
that EU critics were silenced in the discourse does not mean the researcher
judges this as morally wrong. The silencing strategy should rather be
­considered as an effective use of a clear majority position, in both the par-
liament and the party system, to sway the political debate.
The examples given here underline what in Chapter 2 was discussed as
‘setting limits on debate’. This can also be part of political action, of the
game played around and within a debate, and the examples illustrate how
and where these limits were set in the German debate on ratification of the
Constitutional Treaty and how they were manifested in different parties.
While the majority position of the pro-EU politicians allowed them to
impose limits on the debate, or in this case, the discourse, it can be ques-
tioned whether the efforts were successful in terms of persuading citizens
and prevailing in the discourse.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  213

The silencing strategies’ effect on the press discourse itself was twofold.
First, for most of the time it was marked by only a few dissenting voices.
Second, there was barely any exchange of arguments in consequence, and
hence, also very few arguments supporting the Constitutional Treaty were
put forward.
This means that the fact of silencing the critics also weakened the argu-
ments in favour of the Constitutional Treaty: If we consider debate as
an exchange of arguments pro et contra, then it follows that if there is
no exchange and opposing views are silenced, there is no debate. This is
no wonder: how can a debate take place between just the supporters of a
matter?
On the other hand, the interested segments of the public will also
notice the lack of debate and arguments. Moreover, if no arguments are
put forward, how can one be convinced? All in all, in the German 2005
EU press discourse, the German elite consensus and the resulting elite
EU conception were neither actively defended or argued for, nor put to
debate—they only served as a framework for silencing potential EU or
Treaty critics.
This was also the case in the parliamentary debates on the ratification.
The Constitutional Treaty was ratified by a law that was subject to the
usual three readings in the Bundestag. The exchange of arguments that
is linked to the three parliamentary readings took place, but for a large
part in negotiations outside the debates. The contents of the law were
negotiated in the parliamentary committees and negotiation phases, and
especially the critics in the CSU obtained some changes that made it easier
for them to vote in favour of ratification in the end.
But in the parliamentary debates themselves, dissenters only had very
limited occasion to speak against the ratification law because of their party
whips. All Bundestag parties, except the PDS (which only had two MPs),
had adopted it as their group’s position to vote ‘yes’. In the German
­parliamentary culture, a vote that dissents from the whip in such a case is
subject to severe restrictions. Of course, the whip cannot forbid another
vote, but it can threaten, it can apply pressure (as happened in this case),
and it can silence criticism in parliament as well, as speaking time is allo-
cated to the parliamentary groups, and the group chairs may apportion
time to the speakers. In the parliamentary debate on the Constitutional
Treaty ratification, none of the major groups gave any group speaking
time to the internal dissenters.
214  C. WIESNER ET AL.

The only means to express an opinion in such a case is to make a per-


sonal declaration after the formal debate has been closed, just as Gauweiler
did. But this is an instrument that strongly exposes the dissenter, as it
makes his or her dissent from the group position public to the parliament
and also to other groups. There also is a high risk that it will weaken his or
her inner-group position, as group discipline is valued highly and public
dissent disapproved.
The case also shows how the parliamentary rules described in Chapter 2
can be used to manage and structure debates, in particular, rules on time
limits for debate. In the parliamentary sense, the ratification of a treaty
concerns a question of acclamation only, of voting yes or no. The major-
ity parties can justify their stand by emphasising that in this stage there is
no longer time for a thorough deliberation pro et contra, which is more
appropriate to the early rounds and fora of debate.
The results of the ratification vote reflected the huge majority of the
Treaty supporters, but probably also the success of the attempts to sway the
vote through and the silencing. The Constitutional Treaty ratification law
was voted upon in the Bundestag on 12 May 2005, two-and-a-half weeks
before the French referendum. Ninety-five per cent (568) of the depu-
ties voted Yes, two abstained, and only twenty-three voted against—many
of them are members of the Bavarian Christian Social Union (Deutscher
Bundestag 2010, 16386).
In sum, given German parliamentary culture with the consensus in
favour of integration and the clear majority finally obtained, it is not
surprising that Treaty critics were actively silenced in parliament. That
their criticism was not taken up in the press discourse is a bigger sur-
prise, because the press discourse ideal-typically should have been more
open to dissenting voices and more interested in an open exchange of
arguments. It is, therefore, an indicator of the elite consensus and the
influence of the established EU conception in setting limits to debate,
not only directly by the parties and leading politicians, but also indi-
rectly in what Habermas called the ‘public space’ in Germany. The
comparative perspective with France showed quite another reaction to
Gauweiler’s initiative: the fact that the German president had stopped
ratification was reported in long articles in all of the French quality
press journals.
Last but not least in importance, the silencing strategy has also been
tried in the German debate on the Euro crisis. However, the Euro cri-
sis debate not only is very intensive, but it also underlines the limits of
the silencing strategy: all in all, there were too many divergent voices in
EXAMPLES OF ANALYSING DEBATES AS POLITICS  215

the Bundestag parties to completely silence them. The FDP even held an
internal referendum on financial aid for debtor member states, and even-
tually, the euro crisis helped to give birth to AfD.

4.7   The Politics of a Subtext: Football


as a Polity

4.7.1  Text and Subtext
The concept of ‘text’ includes written texts, whether printed or online,
and documents signed or passed by persons or by institutions, associa-
tions, networks and so on. We can extend the concept to entities that are
analysable in a manner analogous to texts. In this volume we have not
dealt with pictures, statues or other works of art, though they can be read
as texts as well (see e.g. Skinner’s analysis of the Lorenzetti Buon governo
frescoes in Siena, included in Skinner 2002b). An interesting concept is
also the ‘city-text’, which includes maps, architecture and directories of
street names (Ferguson 1988). In Chapter 1 we referred to the political
analysis of street naming as a prominent aspect of the city-text.
One step further from the explicit to more implicit forms of politics,
in which the part of the interpreter in reading the political aspect of a
phenomenon becomes stronger than the original actors, can be taken in
the analysis of the ‘subtext’ of a phenomenon. This concept can be under-
stood in the sense of a symptomatic reading, which discards the intentions
of the agents and tries to detect some ‘deeper’ layers of the phenomena.
Psychoanalysis, structuralism and Althusserian Marxism are well-known
examples of such readings. Symptomatic reading, however, tends to lose
sight of politics as an activity.
Politics is intentional activity in the minimal sense of ‘to do something’,
or to strive for power (for its own sake or as a medium for other aims), as
the basic Weberian formula puts it (Weber 1919, 36). However, there are
multiple levels of intentionality, of which the agents themselves may have
different degrees of awareness (see Skinner 1988, 269–273), and actions
might have political implications—‘surplus meaning’ in the sense of Paul
Ricœur (1971)—that are not so visible to the actors, but that interpreters
with a distance to the subject matter can more easily claim to identify. Of
course, such claims are tentative and depend on the interpreter’s perspec-
tive and conceptual horizon. Nevertheless, a political reading of different
kinds of subtexts is frequently valuable for making previously unrecog-
nised political aspects of a topic visible.
216  C. WIESNER ET AL.

To illustrate the political reading of a topic, in which the agents them-


selves do not use the polit-vocabulary, or do so only marginally, in this
section football shall be analysed as a polity. We do not mean the politics-­
dependency or political conditions of the game, but the game itself as a
form of acting politically.

4.7.2  The Political Subtext of Football


Of course, if you ask players or coaches whether they are ‘doing politics’
on the pitch, most of them would deny it. Some players have had a par-
liamentary career after their football career, such as the Brazilian 1994
world champion Romario. The rivalry between clubs such as Celtic and
Rangers in Glasgow has not only a religious, but also a political dimension
(on other rivalries and on a globalisation perspective on football, see Foer
2004). Some FC Barcelona players have recently taken a stand in favour of
Catalan separatism. A number of political aspects can also be found in the
recent FIFA (Fédération Internationale de Football Assocation) presiden-
tial struggle and its background.
Our main point is that as action football is highly contingent and con-
troversial, it corresponds to our main criteria for political action. These
aspects make up the very sense of the game. If the results were all too
predictable on the basis of the quality of the players and the financial
and other resources enabling that quality, or pre-determined by means
of match-fixing or other forms of corruption, the interest in the game
would be lost. The fair play principle, which rests on the assumption of
contingency and controversiality, would in such cases be grossly violated.
Indeed, football associations have more recently introduced tighter rules,
including those of financial fair play, the violations of which have put lim-
its on the commercial transactions of the players, especially for the bigger
clubs, as well as control of club finances. The famous Bosman rule refers
to certain human rights of the players, which were considered superior to
the rights of national leagues to restrict the number of foreign (or non-
­EU) players, and this rule has set in motion a process of denationalising
the game since the 1990s.
Football has a specific contingency concerning the game and its con-
stitutive rules. The elementary rule forbidding playing the ball with
one’s hand, except for the goalkeeper, distinguishes football from rugby.
At the same time this very exclusion marks football’s special profile of
EXAMPLES OF ANALYSING DEBATES AS POLITICS  217

‘politicisation’ by opening up a specific range of possible moves as con-


troversial and by leaving time and space for specific strategies, tactics and
techniques of ‘politicking’ within this range.
Like all other game-type practices, football has certain actors—11
players per team, coaches, referees, club and association representatives,
assisting staff of varying kinds, player’s agents as well as the football jour-
nalists and writers, rank-and-file spectators and fans. It also has a lan-
guage that includes a) the rules of the game, shared by the actors, in
need of interpretation and subject to revision, b) a system of league and
cup style competitions at national and international levels, c) styles and
tactics for how to play the game and d) reports and commentaries on
the game. Football is, furthermore, played in a definite type of space, or
a pitch of certain length, breadth and internal divisions, partly regulated
by the rules, partly by the tactics and the actual uses, complemented
with spaces for referees, reporters and spectators. A football game has
its own time, of 90 minutes, divided into two halves and complemented
by specially regulated extra times, all which also serve as means of the
game. Finally, the name of the game refers to the medium by which it
is played, and the football corresponds to certain requirements of form,
weight and size.
In other words, we have a ‘polity’, for it shares, along with many other
games and other type of polities, a relationship with the five main elements
of a polity—actors, language, space, time and medium. The distinctive cri-
terion is the medium of ball, and although there are remarkable variations
in the ball’s composition and qualities, we leave it out of the following
analysis. The main focus of our analysis lies on the language of football,
especially the relationship of football’s constitutive rules and its stylistic-­
tactical regulations, as well as how the rules and regulations can change
due to the actors (i.e. the players, coaches and the FIFA), including the
regulations and tactics for using space and time.

4.7.3  The Politics of Football Tactics


The following analysis relies on one source, probably the most compre-
hensive overview of football tactics, namely Jonathan Wilson’s Inverting
the Pyramid: A history of football tactics (2008). It contains, first of all,
a draft for the conceptual history of the game, including the history
of the separation of football as a distinct game from rugby, American
218  C. WIESNER ET AL.

football and other related games, with the focus on nineteenth-century


Britain. Wilson tells us that the first attempts to create football rules were
held in Cambridge in 1848 (we can speculate that the Cambridge Union
Society may have played part for constructing the rules), and that only in
1863 in a meeting in London did football and rugby definitely separate
from each other.
Then followed the formation of the Football Association (of England),
and the coordination of English, Scottish and later international rules.
The Brits were the initiators of the game on the European continent and
in South America, which have remained the main metropoles of football
since. The FIFA was founded in 1904. The early rule changes concerned
above all the offside rule and the status of the goalkeeper. Otherwise the
rules have remained fairly stable from the early twentieth century, which
is a major reason for understanding football as one of the most universal
­political languages (the USA being until recently the main exception to
this universality, cf. Markovits 1987; Foer 2004).
Wilson’s point is that a radical change has taken place in football tac-
tics. In the first years football was—as rugby still is—a game in England
aimed at conducting the ball to the goal, passing was suspicious. The
Scots started the passing game already in the nineteenth century. In the
first game between the England and Scotland teams in 1872, England had
a pyramid formation of 7 forwards, 2 midfielders and 1 defender and 1
goalkeeper, whereas Scotland had 6 forwards, 2 midfielders, 2 defenders
and 1 goalkeeper. In the 1880s the setting of the game was established
in the pyramid form of 1 goalkeeper, 2 defenders, 3 midfielders (‘halves’)
and 5 forwards. This pyramid remained the official setting well until the
1960s and still plays a role in the naming of players, although it under-
went a de facto change in the 1920s to the ‘WM’ paradigm, that is the
front players are presented in a W format and the back players in a M for-
mat, with one central defender and the two ‘inner’ forwards behind the
central and flank forwards.
Wilson’s main thesis is that the terms of setting the pyramid were
inverted in the post-war period, and that the trend has led to a constant
decline in the number of forwards—the purest inversion of the pyramid
lies in the ‘Christmas tree’ 4-3-2-1 formation of field players. Wilson
also mentions the possibility of removing the single forward, and since
the publication of the book, the term false nine has been added to the
football lexicon.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  219

Wilson’s conclusion is that the team is losing the fixed division of


labour between the player types, and a broad palette of competences is
now required from every player. With these trends, the once strict divi-
sion of the pitch into parts within which each player acts has been relativ-
ised. This sets new and usually harder qualification requirements for each
player. In our terms, this can be regarded as a politicisation of the game,
in that it has removed areas traditionally ‘protected’ from the intervention
of certain players, and opened up new and unexpected chances every field
player to intervene. Even for the goalkeeper, Manuel Neuer’s style in the
2014 World Cup has created a new paradigm.
Within the trend of how players set on the field, Wilson tells us the
complex story of the shifting tactics, with different and frequently unin-
tended origins, suddenly applied by coaches to come through with an
unexpected win. This sort of politicking and the use of more or less fixed
policy within it will be described below in its main outlines. This topic
of political tactics is also a part of the language of football and includes
neologisms and morphological changes, which reach beyond issue of the
formal numerical setting of the players on the pitch.
Another level in Wilson’s narrative consists of various attempts to
overcome the traditional robust and results-oriented approach of English
football with either more artistic or more system-oriented styles of play-
ing. The revisions are presented with names. The Austrian Wunderteam,
coached by Hugo Meisl in the 1920s and 1930s, was the first cele-
brated example of what Wilson calls Danubian football, followed by the
Hungarian aranycsapat around Puskas and Hidegkuti, which dominated
the world from 1950 to 1956. Another river name of Wilson is Rio de la
Plata football, based on the ball possession tactics of the 1930 and 1950
world champion Uruguayan (la carrachárruasta) and Argentinian (la
nuestra) teams prior to and after WWII. The post-war artistic line was
practised by the Brazil world champions of 1958, 1962 and 1970, with
Pelé as the key player, sometimes called la diagonale. The Ajax and Dutch
national team of totaalvoetbal from the 1970s has inspired contempo-
rary ball-possession football, including the tiki-taka of FC Barcelona of
recent years, which have moved far from a game based on fixed positions
on the pitch.
The home of the defensive styles is Italy, la metodo of Vittorio Pozzo
(world champions in 1934 and 1938). Via the Swiss, verrou led to the
most notorious of football styles, the catenaccio, as practised by Inter
220  C. WIESNER ET AL.

Milan under Helenio Herrera in 1980s, against which rule changes were
introduced (see below).
Wilson sees the origins of pressing football in the Soviet Union, in the
works of such system-oriented trainers as Victor Maslov in the 1960s and
Valeri Lobanovski of Dynamo Kyiv from the 1970s onwards. It was fol-
lowed by, for example, the success of A.C. Milan under Arrigo Sacchi in
the early 1990s, and reactivated in the form of Gegenpressing by Jürgen
Klopp as the coach for Borussia Dortmund.
A political reading can point out that every new style in a sense politi-
cises the game by creating new spaces or times to play—until the adver-
saries are able to find the weaknesses of the new style. Even catenaccio,
with its aim of preventing the adversaries from playing their ‘own game’,
contains innovative aspects and gives the teams with weaker resources a
chance, although the more artistic styles are more unpredictable. National
traditions can become straitjackets, and one of Wilson’s points lies in the
inability of the English team to adopt innovations from the continental or
South American traditions.
Wilson’s narrative emphasises the increasing superiority of tactical sys-
tems—policy styles—over the brilliance of individual players. He regards
the celebrated Brazil team of 1970 as the last example of the success of
brilliant individuals. The ‘scientific’ approach of Lobanovski or Sacchi—or
today the Danish champions FC Midtjylland—and with it the authority
of coaches has achieved superior results, as compared to, for example, the
sometime Ajax Amsterdam practice in which the players debated on the
strategy. However, Wilson warns against the dogmatism of committing to
a definite style and finds politicising effects in changing it according to the
team resources, adversaries and even within a game—he emphasises on the
role of politicking over policy. When every player has versatile competen-
cies, probably the last word has not been said on the ‘romantic’ versus the
‘scientific’ style of play.
One aspect that Wilson has neglected is the politicising effect of the
temporalisation of the game via changing rules. The increase of allowed
player changed during a match changed from one (still in the 1960s) to
three, and the UEFA (Union des Associations Européennes de Football)
‘away goal’ rule (when the result of two-part cup competitions is a draw,
the number of away goals decide) has already changed the rhythm of the
game, allowing tactical changes within it. The first was probably intended
mainly to help players to recover injuries, the second to avoid too many
extra times, penalty shootouts and even replays.
EXAMPLES OF ANALYSING DEBATES AS POLITICS  221

As a reaction to the lack of goals through the introduction of catenaccio-­


like styles, in the 1990s, back passes to the goalie by foot were prohibited;
more recently the extra time has been announced by a fourth official.
Sepp Herberger’s—the coach of the West German 1954 world champion
team—declaration of the 90-minute duration of matches (‘Das Spiel dau-
ert 90 Minuten’, in the film Das Wunder von Bern attributed to a clean-
ing lady) has become relativised, and the politics of the rhythm inside the
game appears more important than ever. An important part of the politics
of time in football is the winter break. Northern countries playing from
spring to autumn tend to have a disadvantage, but also the absence of
a winter break has been considered a major reason why British national
teams have lost their standing on the world and European football stage.
The creation of the Champions League has by itself also changed the
annual football calendars of the leading teams since the early 1990s.
In any case, these aspects of the temporalisation of the game have altered
football as a (political) action to a considerable degree. Until the 1960s
teams at least used to play every match with the same regular players,
if possible, without many reserve options. The changes in the rules, the
higher pace of the matches as well as the increased range of tactical variants
have led to a larger staff of players, and to some extent a regular rotation
among them has become necessary. When changes of players cause greater
breaks in a match than other interruptions, they might also be routinely
used to break down the momentum of the adversary. The coaches can,
in addition, adapt their tactics to each specific adversary and respond to
falling behind by changing players and tactical formations. Currently, the
two-part cup competitions with UEFA rules might dramatise the need to
make more changes of players within the game.
One unintended—although probably predictable—consequence of the
extended possibilities of changing players within the match as well as of
the Bosman rule has been the increasing player mobility with a tendency
to raise their ‘price level’, both for the clubs and the players themselves.
This has led to a further concentration of the football powers that be
among the big and financially strong clubs, including changes in organ-
isational structure, from members’ associations into objects of financial
investment.
For football romantics, all this means a decline in the game. We could
expect further measures to prevent declines in competition and contin-
gency, especially against the clubs’ (such as Bayern München’s) attempts
to monopolise the championships in the national leagues. In terms of
222  C. WIESNER ET AL.

financial fair play, much remains to be done, although the appropriate


measures are by no means self-evident. In terms of football aesthetics and
the richness of tactical variation, we cannot speak of a decline, and the
same holds for football’s denationalisation.
To some extent, a growing political awareness can be expected also
from professional football players themselves. Lack of political sensitiv-
ity led to the end of, for example, Nicolas Anelka’s professional football
career. More political skills in ‘reading the game’—an expression that was
not yet used in the 1970s—are required also from individual players.

Notes
1. The Council (of the European Union) is to be distinguished from the
European Council which consists of the Heads of State and Government.
The Council consists of the relevant ministers of the member states and
for the first decades of integration was the only institution that could
decide on EU laws. The Council has also sometimes been termed the
‘Council of Ministers’ in everyday use.
2. “LXIII. Amendments to such Motions [referring to motions that have
been announced beforehand in writing], or Riders to such Motions or
Amendments, may be moved without notice in the course of the
Debate; and at any time during the Meeting any Member may bring
forward a Motion which has reference solely to the Conduct or
Adjournment of the then Meeting, and the Mover of such Motion shall
have precedence of all other speakers” (OUS rules 1856, 40–41).
3. The terms ‘requisition’ and ‘Committee of the whole House’ were
commonly used in the House of Commons as well. However, their
practical usage was different from the parliamentary procedure.
4. The EU commission established the expert group to analyse the appli-
cation of EU citizenship rights only after the respective EU laws had
been valid for four years.
Concluding Notes: Reappraising Politics
and Debate

Why should a researcher concern herself with the activity of politics? In


this book we hope to have given some answers to this question for stu-
dents and scholars alike in the humanities and social sciences. We also
hope to have offered compelling reasons to reappraise politics as a contin-
gent and controversial activity, as it is our impression that this perspective
is sometimes underappreciated or left out completely in current research.
Students and scholars have much to gain by studying politics as an activ-
ity. This perspective opens up several dimensions that are critical for under-
standing politics as action, as opposed to a perspective that focuses primarily
on the outcome of politics, as seen for instance in studies of votes and legis-
lative processes. If the researcher does not include the dimensions linked to
the activity-character of politics, her results will be based only on a part of
the full panorama. As an example, in EU studies frequently the European
Parliament’s powers against the Commission and the Council are evalu-
ated by counting the voting results of legislative processes. While this is a
possible and perfectly legitimate approach, it does not tell the researcher
how the battle of forces between the institutions and their members went
about, what strategies the actors used, how they applied them, and why
they won, lost or adjourned their motions. Moreover, concentrating on the
mere outcomes of decision-making does not tell the researcher much about
the backgrounds, moves and rhetoric of the politics involved. As politics is
mainly carried out by linguistic acts (speeches, documents, letters, legislative

© The Author(s) 2017 223


C. Wiesner et al., Debates, Rhetoric and Political Action,
DOI 10.1057/978-1-137-57057-4
224  Concluding Notes: Reappraising Politics and Debate

texts, etc.), this dimension is crucial to its understanding. Scholars should


therefore be sensitive to the processes and actions, strategies, intentions
and speech acts that underlie the outcome of politics, when they analyse
speeches and writings of persons involved in the studied contexts of politics.
We have endeavoured to show that, if a person wants to study politics as
an activity, she should focus on the moves and strategies of the actors involved
and gain some knowledge about the rules and the language of politics; in
other words, she needs political literacy. In addition, she should have or
develop concrete approaches and practices that allow for studying the actions,
moves and speech acts of the agents, their claims and their backgrounds.
Besides our general aim of highlighting why politics should be studied
from the point of view of its activity dimensions, we also wanted to offer
some useful hints and comments in this respect. While we strongly believe
that each research design must be individual and custom-tailored to the
research question and the researcher’s own interests, we still think there
are some approaches and practices that could be helpful and valuable for
a broader public. Moreover, we wanted to provide concrete examples of
how politics as an activity can be studied.
We also hope to have encouraged students and scholars to reconsider
debate as a political activity. Debates, as has been underlined, can take dif-
ferent forms and take place in different arenas, even if parliamentary debate
is considered its classical paradigm. They may be either virtual (‘frozen’)
or live, and they may occur within a single sitting or be extended over long
time periods.
With our theoretical and introductory thoughts and with our exem-
plar cases, we have aimed to show that the study of the rhetorical aspects
of politics, as found in various types of debate, should not be limited
to scholars of rhetoric or discourse analysis alone. On the contrary, the
case studies are important for understanding aspects of political life itself,
including the outcomes of elections or referenda, where quite often the
vote is preceded by some form of debate or an exchange of arguments
(see, in particular, Chapters 3 and 4).
As mentioned, there are different interpretations in the work of classi-
cal and modern political thinkers about the concept of politics and its role
in human life. The different interpretations can open our eyes today to a
wider range of interpretations as to what is to be considered political.
While Aristotle’s characterisation of human beings as zoon politikon is
usually translated as political animal, a closer approximation would be to
say that human beings are bound to the polis ‘by nature’ (Politics, 1253a).
Concluding Notes: Reappraising Politics and Debate  225

However, the ancient Greek polis itself was not necessarily viewed as ‘natu-
ral’, for it was a counter-formation to the old despotic regime (see e.g.
Finley 1983; Hansen 1998). In contrast, modern defenders of politics,
such as Max Weber in his essay on the city (1922, 727–814) and Hannah
Arendt (1993) denied that human politics had a ‘natural’ character and,
instead, emphasised its voluntary or human character.
A claimed rejection of or disinterest in politics should also, therefore,
not be seen as contrary to ‘human nature’, for it might itself be of politi-
cal significance, especially in participatory regimes with universal suffrage
and parliamentary government. In today’s representative democracies, the
low regard for ‘politics’ and ‘politicians’ is a common practice and seems
to be on the rise. Many opinion polls show a low degree of ‘interest in
politics’. In contrast to other periods, such as the 1960s and 1970s, the
reputation of ‘politics’ seems once again to have hit a nadir (for similari-
ties between the current climate and the period around 1900, see Palonen
2012b). Celebrities in arts and sports, for example, frequently state a dis-
interest in politics or may even take pride in declaring ‘I know nothing
about politics’. Even scholars can be heard to declare ‘I never vote’. While
in totalitarian regimes such declarations may have a touch of protest (see
Konrad 1985), in legitimate democratic polities with the right to express
controversial opinions, they are rather signs of resignation. Political action,
even in the minimal sense of voting or expressing one’s opinion in a political
conflict of the day, is then abandoned and one’s own political fate is left for
others to decide.
But such ‘anti-political’ declarations also refer to the narrow concep-
tion of politics as a sphere (as sketched in Chapter 1), that is, as refer-
ring only to the deeds or misdeeds of professional politicians. However,
when everything—even the denial of being involved in ‘politics’—can be
regarded as political, it does not make sense just to declare that something
has political aspects or to blame others to be apolitical, unpolitical or anti-
political. To understand such uses, we have to move the discussion to the
level of different conceptions of politics. Or they might be a question of
political literacy: when someone claims that there is nothing political in a
phenomenon, the claim challenges others to invent modes of identifying
political aspects in it.
Our book has a normative perspective (Wertbeziehung for Weber
1917a) in so far as it welcomes, so to say, ‘more politics in all our lives’,
and as it discusses various possibilities for confronting political (i.e. con-
tingent and controversial) situations. For Weber, ‘all of us’ who may vote,
226  Concluding Notes: Reappraising Politics and Debate

speak in political meetings or write about political issues to newspapers are


‘occasional politicians’ (Weber 1919, 41).
Today, the possibilities for acting as an occasional politician have mul-
tiplied due to the numerous internet fora available for expressing opin-
ions. On the other hand, these fora create very limited political spaces,
sometimes only reserved for a small public interested in a special question.
However, as we have illustrated in this volume, the range of what is seen as
contingent and controversial is also expanding. Traditions and c­ onventions
have been losing their force. Not only is the vote no longer ‘inherited’
from the family or the home village (in the way that Wright 2012, for
example, writes about his ‘inherited’ commitment to the Labour Party),
but the choices of life and lifestyle have become politicised in our sense
of the concept. The choices of what one eats, what clothing one wears,
how and where one travels, for example, have become highly ­contingent,
controversial and therefore thoroughly political.
We would like to offer a political mode of thinking that emphasises the
actors and the activity dimension of politics, as well as its contingent and
controversial character. This approach presents an alternative to the seem-
ingly ‘unpolitical’ language of markets, businesses, capital and so on that
have been metaphorically extended to everyday language, media discourse
and even the evaluation of research. Using terms like systems, structures,
processes and functions forms a kind of common sense within administra-
tive jargon which they have adopted from the social sciences as they were
a half-century ago. If politics appears at all in such languages, it is in the
binary codes of government versus opposition and right wing versus left
wing (see in particular Luhmann 2000).
In contrast, we have attempted to draft a concept of ‘political literacy’,
defined as the ability and willingness to deal with the political in seemingly
everyday situations, and to discuss how such literacy may be used for read-
ing and analysing debates and documents. We have also illustrated some of
the different ways of reading and interpreting politics and debates, despite
some common standards and practices that mark all qualitative and inter-
pretative analyses. As the examples in Section 4.1 have shown, one and
the same debate can be read with quite different foci, depending on the
research interest and the research question. In order to carry out research,
a researcher thus needs to develop a particular and individual research
design, going continually back and forth between the sources, the research
questions and intermediate findings. The sources (to ­reiterate the point
once more) do not speak for themselves: they must be ­interpreted by the
Concluding Notes: Reappraising Politics and Debate  227

researcher. But the sources do carry a ‘veto power’: they must be taken
seriously, and the researcher must be prepared to have her predefined cat-
egories overturned by them.
In conclusion, therefore, we hope to have indicated a number of possi-
ble paths and approaches to a broad understanding of the study of politics
as an activity, and of debates and documents as an important part of it, in
a way that transcends a narrow specialisation into subfields and subdisci-
plines. Politics has many facets which must be valued in their entirety and
in their full complexity.
Metaphorically, politics can be seen as a game or as play, as it is a sym-
bolic form of organising life structured by distinct rules. Politics is also a
language-based activity, and the speech acts performed possess a political
character. Moreover, to see politics as a human action means to emphasise
its intrinsic relation to contingency, just as a politician always faces a hori-
zon of possibilities, not knowing in advance what the consequences and
outcomes of her actions may be, as they always depend also on the other
actors involved and their moves. Politics relates to human beings and how
we interact.
References

Archival Sources
Cambridge Union Society Laws.1827–1870. Laws of the Union Society. Cambridge:
Cambridge Union Society.
Cambridge Union Society Minute Books. 1823–1874. Minute Book [vols. 1–14].
Cambridge: Cambridge Union Society.
Oxford Union Society Minute Books. 1831–1869. Minute Book [vols. I-X].
Oxford: Oxford Union Society.
Oxford Union Society Rules.1837–1867. Rules and Regulations. Oxford: Oxford
Union Society.

Databases, Websites and Treaties


Council of the European Communities; Commission of the European Communities.
1992. Treaty on European Union. http://europa.eu/eu-law/decision-mak-
ing/treaties/pdf/treaty_on_european_union_en.pdf. Accessed 29 Feb 2016.
Deutscher Bundestag. 2010. Plenarprotokoll 16/157. http://dip21.bundestag.
de/dip21/btp/16/16157.pdf. Accessed 1 Apr 2010.
European Commission. 1997. Bericht der hochrangigen Arbeitsgruppe zu Fragen
der Freizügigkeit unter dem Vorsitz von Frau Simone Veil. http://www.uni-­
mannheim.de/edz/pdf/1997/hlpde.pdf. Accessed 12 Feb 2013.
———. 1999. Communication from the Commission to the Council, the
European Parliament, the Economic and Social Committee and the Committee
of the Regions on Certain Community Measures to Combat Discrimination.

© The Author(s) 2017 229


C. Wiesner et al., Debates, Rhetoric and Political Action,
DOI 10.1057/978-1-137-57057-4
230  References

http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=COM:1999:0564:
FIN:EN:PDF. Accessed 12 Feb 2013.
———. 2001. Third Report from the Commission on Citizenship of the Union,
COM (2001) 506 Final. http://eur-lex.europa.eu/LexUriServ/LexUriServ.
do?uri=COM:2001:0506:FIN:EN:PDF. Accessed 12 Feb 2013.
European Council. 1957. The Treaty of Rome. http://ec.europa.eu/economy_
finance/emu_history/documents/treaties/rometreaty2.pdf. Accessed 11 Feb
2013.
———. 2000a. COUNCIL DIRECTIVE 2000/43/EC of 29 June 2000
Implementing the Principle of Equal Treatment Between Persons Irrespective
of Racial or Ethnic Origin. http://eur-lex.europa.eu/LexUriServ/LexUriServ.
do?uri=OJ:L:2000:180:0022:0026:EN:PDF. Accessed 19 Nov 2013.
———. 2000b. COUNCIL DIRECTIVE 2000/78/EC of 27 November 2000
Establishing a General Framework for Equal Treatment in Employment and
Occupation. http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:
2000:303:0016:0022:EN:PDF. Accessed 19 Nov 2013.
European Court of Justice. 1976. JUDGMENT OF 8. 4. 1976—CASE 43/75.
http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:6197
5CJ0043&from=FR. Accessed 9 Oct 2015.
European Parliament. 2016.  About Parliament. http://www.europarl.europa.
eu/aboutparliament/en. Accessed 15 May 2016.
European Parliament. 2014.  Rules of Procedure of the European Parliament, 8th
Parliamentary Term. http://tinyurl.com/mcrcpgo. Accessed 15 July 2014.
European Union. 2001. Der Vertrag von Amsterdam: Gebrauchsanweisung
Unionsbürgerschaft. http://europa.eu.int/scadplus/leg/de/lvb/a12000.
htm. Accessed 25 Aug 2005.
———. 2004. DIRECTIVE 2004/38/EC OF THE EUROPEAN PARLIAMENT
AND OF THE COUNCIL of 29 April 2004. http://eur-lex.europa.eu/
LexUriServ/LexUriServ.do?uri=OJ:L:2004:158:0077:0123:en:PDF. Accessed
12 Feb 2013.
———. 2010. Consolidated Treaties. Charter of Fundamental Rights. http://
bookshop.europa.eu/is-­bin/INTERSHOP.enfinity/WFS/EU-Bookshop-
Site/en_GB/-/EUR/ViewPublication-Start?PublicationKey=QC3209190.
Accessed 5 Feb 2013.
Hansard Debates. http://hansard.millbanksystems.com. Accessed 15 May 2016.
House of Commons, Parliamentary Papers (ProQuest), Subscription University of
Jyväskylä Library. https://jyu.finna.fi/Record/metalib_jyu.FIN30598.
Accessed 09.05.2016.
Report from the Select Committee on the House of Commons (Procedure),
together with the proceedings of the committee, minutes of evidence and
appendices 1914-070470-p1to345.pdf. House of Commons, Parliamentary
Papers (ProQuest), Subscription University of Jyväskylä Library, https://
jy,finna.fi/Record.metalib_jyu.fi.FIN30598. Accessed 09.05.2016.
References  231

Literature
Adamson, Sylvia, Gavin Alexander, and Karen Ettenhuber, eds. 2007. Renaissance
Figures of Speech. Cambridge: Cambridge University Press.
Ankersmit, F.R. 1996. Aesthetic Politics. Stanford, CA: Stanford University Press.
———. 2001. Historical Representation. Stanford, CA: Stanford University Press.
———. 2002. Political Representation. Stanford, CA: Stanford University Press.
Arendt, Hannah. 1958 [2013]. The Human Condition. Chicago: University of
Chicago Press.
———. 1960. Vita activa. München: Piper.
Aristotle. 2000 [350 B.C.]. Rhetoric. Translated by W. Rhys Roberts. South Bend,
US: Infomotions Inc. Accessed 26 August 2016. ProQuest ebrary.
———. 1993. Was ist Politik? ed. Ursula Ludz. München: Piper.
———. 2000. Politics. South Bend, US: Infomotions, Inc. Accessed August 30,
2016. ProQuest ebrary.
———. 2000. Rhetoric. Translated by W.  Rhys Roberts. South Bend, US:
Infomotions Inc. Accessed 26 August 2016. ProQuest ebrary.
Atkins, Judi, Alan Finlayson, James Martin, and Nick Turnbull, eds. 2014. Rhetoric
in British Politics and Society, Series: Rhetoric, Politics and Society. Basingstoke:
Palgrave Macmillan.
Austin, J.L. 1962 [1990]. How to Do Things with Words. Edited by J.O. Urmson
and Marina Sbisà. Oxford: Oxford University Press.
Bagehot, Walter. 1867 [2001]. The English Constitution. Edited by Paul Smith.
Series: Cambridge Texts in the History of Political Thought. Cambridge:
Cambridge University Press.
———. 1872a [2001]. Introduction to the Second Edition. In Bagehot, Walter. The
English Constitution. Edited by Paul Smith. Series: Cambridge Texts in the History
of Political Thought. Cambridge: Cambridge University Press, 193–229.
———. 1872b [1956]. Physics and Politics. Boston, MA: Beacon Press.
Bagehot, Walter. 1956 [1872]. Physics and Politics. Boston, MA: Beacon Press.
Baker, Paul, Costas Gabrielatos, Majid Khosravinik, Michal Krzyzanowski, Tony
Mcenery and Ruth Wodak. 2013. A Useful Methodological Synergy? Combining
Critical Discourse Analysis and Corpus Linguistics to Examine Discourses of
Refugees and Asylum Seekers in the UK Press. In Critical Discourse Analysis.
Vol. 2: Methodologies, ed. Ruth Wodak, 1–34. London [u.a.]: SAGE.
Beetham, David. 2006. Parliament and Democracy in the Twenty-First Century.
A Guide to Good Practice. Geneva: Inter-Parliamentary Union. http://tinyurl.
com/oa5wpz2
Bellamy, Richard.  2007. Political Constitutionalism. Cambridge: Cambridge
University Press.
Bentham, Jeremy. 1843. An Essay on Political Tactics. In Collected Works of
Jeremy Bentham, vol. 2. Edinburgh: William Tait. http://oll.libertyfund.org/
title/1921/113915
232  References

———. 1999. Political Tactics. Edited by Michael James/Cyprian Blamires and/


Catherine Pease-Watkin. Oxford: Clarendon Press.
Brecht, Bertolt. 1967. Gesammelte Werke, Bd. 15. Wikipedia [LD1]http://de.
wikipedia.org/wiki/Verfremdungseffekt#Brechts_Idee_als_Gegenpart_zum_
aristotelischen_Dramenbegriff. Accessed 5 May 2016.
Brewer, John Marks. 1916. Oral English. Directions and Exercises for Planning and
Delivering Common Kinds of Talks, Together with Guidance for Debating and
Parliamentary Practice. Boston, MA: Ginn. http://tinyurl.com/6nycqbj
Bryce, James. 1888/1914 [1995]. The American Commonwealth. Indianapolis:
Liberty Fund.
Burke, Edmund. 1774 [1999]. Speech to the Electors of Bristol. In Selected Works
of Edmund Burke, vol. 4, 5–14. Indianapolis: Liberty Fund. http://oll.liberty-
fund.org/title/659/20392.Burke,Kenneth
Burke, Kenneth. 1945 [1969]. A Grammar of Motives. Berkeley: University of
California Press.
———. 1950 [1969]. A Rhetoric of Motives. Berkeley: University of California
Press.
Burkhardt, Armin. 2004. Zwischen Monolog und Dialog. Zur Theorie, Typologie
und Geschichte des Zwischenrufs im deutschen Parlamentarismus. PhD disserta-
tion, Tübingen, Niemayer.
Burman, Lars. 2012. Eloquent Students: Rhetorical Practices at the Uppsala Student
Nations 1663–2010. Uppsala: Acta Universitatis Upsaliensis.
Campion, G.F.M. 1929. An Introduction to the Procedure of the House of Commons.
London: Allen & Co.
Campion, Gilbert. 1953. Parliamentary Procedure: Old and New. In Parliament.
A Survey, ed. C.  Campion, L.S.  Amery, and D.  Brogan, 141–167. London:
Allen and Unwin.
———. 1958. An Introduction to the Procedure of the House of Commons, 3rd edn.
London: Macmillan.
Carlyle, Thomas. 1850 [1872]. Latter-­Day Pamphlets. London: Chapman and
Hall.
Charteris-Black, Jonathan. 2014. Analysing Political Speeches: Rhetoric, Discourse
and Metaphor. Basingstoke: Palgrave Macmillan.
Clausewitz, Carl von. 1832 [1980]. Vom Kriege. Frankfurt/M: Ullstein.
Clinchamps, Nicolas. 2006. Parlement Européen et droit parlementaire. Essai sur la
naissance du droit parlementaire de l’Union Européenne. Paris: L.G.D.J.
Colclough, David. 2005. Freedom of Speech in Early Stuart England. Cambridge:
Cambridge University Press.
Collingwood, R.G. 1946 [1994]. The Idea of History. Edited by Jan van der
Dussen. Revised edition with Lecture 1926–1938. Oxford: Oxford University
Press.
Craig, Robert.  1913. A History of Oratory in Parliament, 1213–1913. London:
Heath, Cranton and Oaseley.
References  233

Crewe, Emma. 2015. The House of Commons. An Anthropology of the MPs at Work.


London: Bloomsbury.
Crick, Bernard. 1962 [2000]. In Defence of Politics. London: Continuum.
Curzon, George Nathaniel.  1913. Modern Parliamentary Eloquence. London:
Macmillan.
De Mille, James. 1878. Elements of Rhetoric. New  York: Harper & Brothers.
http://tinyurl.com/mhps972
Dijk, Teun A. van. 2001. Critical Discourse Analysis. In The Handbook of Discourse
Analysis, ed. D.  Tannen, D.  Schiffrin, and H.  Hamilton, 352–371. Malden,
MA [u.a.]: Blackwell.
Dunn, John. 1968. The Identity of the History of Ideas. Philosophy 43: 85–104.
Fairclough, Norman.  2004. Analysing Discourse: Textual Analysis for Social
Research. London: Taylor & Francis.
Fairclough, Norman, and Ruth Wodak.  1997. Critical Discourse Analysis. In
Discourse as Social Interaction, ed. Teun A. van Dijk, 258–284. London [u.a.]:
SAGE.
Fawcett, Trevor. 1980. 18th-Century Debating Societies. Journal for Eighteenth-
Century Studies 3(3): 216–229.
Ferguson, Priscilla Pankhurst. 1988. Reading City Streets. The French Review 61:
386–397.
Finlayson, Alan.  2014. Proving, Pleasing and Persuading? Rhetoric in
Contemporary British Politics. The Political Quarterly 85(4): 428–436.
Finley, Moses I.  1983. Politics in the Ancient World. Cambridge: Cambridge
University Press.
Floehr, Ralf, and Klaus Schmidt, eds.  1984. “Unglaublich, Herr Präsident!”
Ordnungsrufe/ Herbert Wehner. Frankfurt/M.: Zweitausendeins.
Flynn, Paul. 2012. How to be an MP. London: Biteback.
Foer, Franklin. 2004. How Football Explains the World. London: Arrow Books.
Foucault, Michel. 1969. L’archéologie du savoir. Bibliothèque des sciences humaines.
Paris: Gallimard.
———.  1971. L’ordre du discours: Leçon inaugurale au Collège de France pronon-
cée le 2 décembre 1970. Paris: Gallimard.
Gagarin, Michael, and Paul Woodruff, eds.  1995. Early Greek Political Thought
from Homer to the Sophists. Cambridge: Cambridge University Press.
Gallie, W.B. 1973. An Ambiguity in the Idea of Politics and its Practical
Implications. Political Studies 21: 439–452.
Garrigues, Jean, ed. 2007. L’histoire du parlement. Paris: PUF.
Gladstone, W.E. 1838 [1953]. Public Speaking. The Quarterly Journal of Speech
39: 266–272.
Gauger, Hildegard. 1952. Die Kunst der politischen Rede in England. Tübingen:
Niemayer.
Gladstone, W.E. 1838 [1953]. Public Speaking. The Quarterly Journal of Speech
39: 266–272.
234  References

Glaser, Barney G., and Anselm L. Strauss. 1967. The Discovery of Grounded Theory:
Strategies for Qualitative Research. Observations. Chicago, IL: Aldine.
Gläser, Jochen, and Grit Laudel.  2004. Experteninterviews und Qualitative
Inhaltanalyse. Wiesbaden: VS.
Griffith, J.A.G., and Michael Ryle. 2003. Parliament. Functions, Practices and
Procedures. 2nd edition by Robert Blackburn and Andrew Kinnon. London:
Sweet & Maxwell.
Gunn, J.A.W. 2009. When the French Tried to Be British. Party, Opposition and the
Quest for Civil Disagreement 1814–1848. Montreal: McGill-­Queens University
Press.
Haapala, Taru. 2012. “That in the Opinion of this House”: The Parliamentary
Culture of Debate in the Nineteenth-Century Cambridge and Oxford Union
Societies. PhD dissertation, University of Jyväskylä.
———.  2015. The Issue of the Secret Ballot in the Cambridge and Oxford Union
Societies, c.1830–72: An Extension of the Nineteenth-­Century Parliamentary
Culture of Debate. Parliaments, Estates & Representation 35(1): 66–83.
———. forthcoming 2017. Political Rhetoric in the Oxford and Cambridge Unions,
1830–1870, Series: Studies in Modern History. Basingstoke: Palgrave Macmillan.
Hamilton, William Gerard. 1927 [1808]. Parliamentary Logic. Edited by
Courtney S. Kenny. Cambridge: Heffers.
Hansen, Mogens Herman. 1998. Polis and the City-State. Copenhagen: Munksgaard.
Hatsell, John. 1818. Precedents of Proceedings in the House of Commons; with
Observations, Vols. 1–4. London: L.  Hansard and Sons. Vol. 1, Privilege of
Parliament: http://tinyurl.com/os3pkuz; Vol. 2, Relating to Members,
Speaker etc.: http://tinyurl.com/7p7br4h; Vol. 3, Relating to Lords and
Supply: http://tinyurl.com/ncazhkg; Vol. 4, Relating to Conference and
Impeachment: http://tinyurl.com/6muph6o.
Hexter, J.H., ed.  1992. Parliament and Liberty. Princeton, NJ: Princeton
University Press.
Ihalainen, Pasi, and Kari Palonen.2009. Parliamentary Sources in the Comparative
Study of Conceptual History: Methodological Aspects and Illustration of a
Research Proposal. Parliaments, Estates & Representation 29: 17–34.
Ilbert, Courtenay. 1901. Legislative Methods and Forms. Oxford: Clarendon Press.
http://tinyurl.com/oe7x8p6
———. 1911. Parliament, Its History, Constitution and Practice. London:
Williams and Norgate. http://tinyurl.com/nm5gvvk
Ilie, Cornelia.  2001. Unparliamentary Language. Insults as Cognitive Forms of
Confrontation. In Language and Ideology. Vol. 2, eds. R. Driven, R. Frank, and
C. Ilie, 235–263. Amsterdam: John Benjamins.
———. 2004. Insulting as (un)parliamentary Practice in the British and Swedish
Parliaments. In Cross-­Cultural Perspectives on Parliamentary Discourse, Series:
Discourse Approaches to Politics, Society and Culture, ed. Paul Bayley, 45–86.
Amsterdam: John Benjamins.
References  235

———, ed. 2010. European Parliaments under Scrutiny. Amsterdam: John


Benjamins.
Johnston, James. 1927. Westminster Voices. London: Hodder and Stoughton.
Johnstone, Barbara.  2008. Discourse Analysis, 2nd edn. Malden, MA [u.a.]:
Blackwell.
Kautsky, Karl. 1911. Parlamentarismus und Demokratie. Berlin: Dietz.
Kelle, Udo, and Susann Kluge. 1999. Vom Einzelfall zum Typus. Opladen: Leske
+ Budrich.
Kivistö, Hanna-Mari. 2013. ‘Dubliners’ in the European Union—A Perspective
on the Politics of Asylum-Seeking. In The Distant Present, eds. Tuula Vaarakallio
and Taru Haapala, 106–128. Jyväskylä: University of Jyväskylä, SoPhi.
Konrad, Gyorgy. 1985. Antipolitik. Mitteleuropäische Meditationen. Frankfurt/M:
Suhrkamp.
Kopperschmidt, Josef, and Helmut Schanze, eds.  1994. Nietzsche oder “Die
Sprache ist Rhetorik”. München: Fink.
Koselleck, Reinhart. 1972. Einleitung. In Geschichtliche Grundbegriffe,
edited by Otto Brunner, Werner Conze and Reinhart Koselleck.vol. I.
xiii–xxvii.
———.  1979. Vergangene Zukunft. Zur Theorie historischer Zeiten. Frankfurt/M:
Suhrkamp.
———. 1982 [2010]. Archivalien—Quellen—Geschichten. In Vom Sinn und
Unsinn der Geschichte, ed. Carsten Dutt, 68–79. Berlin: Suhrkamp.
———. 1983. Begriffsgeschichtliche Probleme der Verfassungsgeschichts-
schrei­bung. Der Staat-Sonderheft 6: 7–21.
———. 1996. A Response to the Comments on the Geschichtliche Grundbegriffe.
In The Meaning of Historical Terms and Concepts. Occasional Paper No 15, eds.
Hartmut Lehmann and Melvin Richter, 59–70. Washington, DC: The German
Historical Institute.
———. 2000. Zeitschichten. Frankfurt/M: Suhrkamp.
———. 2006. Begriffsgeschichten. Edited by Carsten Dutt. Frankfurt/M:
Suhrkamp.
———. 2010.Vom Sinn und Unsinn der Geschichte, Epilog by Carsten Dutt. Ed.
Berlin: Suhrkamp.
———. 2011. Introduction (Einleitung) to the Geschichtliche Grundbegriffe.
Translated by Michaela Richter. Contributions to the History of Concepts 7, 7–25.
Koselleck, Reinhart, Ulrike Spree, and Willibald Steinmetz. 2006. Drei bürgerli-
che Welten. Zur vergleichenden Semantik der bürgerlichen Gesellschaft in
Deutschland, Frankreich und England. In Begriffsgeschichten: Studien zur
Semantik und Pragmatik der p­ olitischen und sozialen Sprache, edited by Reinhart
Koselleck, 402–463. Frankfurt am Main: Suhrkamp.
Kronlund, Anna. 2013. Parliamentary Oversight of the Exceptional Situations in a
Presidential System. Debating the Reassertion of the Constitutional Powers of the
US Congress. PhD dissertation, University of Jyväskylä.
236  References

Kurunmäki, Jussi. 2015. How Women’s Suffrage Was Devalued: The Burden of


Analytical Categories and the Conceptual History of Democracy. In
Parliamentarism and Democratic Theory, eds. Kari Palonen, and José María
Rosales, 31–52. Leverkusen: Budrich.
Laclau, Ernesto, and Chantal Mouffe.1985. Hegemony and Socialist Strategy:
Towards a Radical Democratic Politics. London: Verso.
Lausberg, Heinrich. 1962. Elemente der literarischen Rhetorik. München: Huber.
Lowe, Robert. 1867. Speeches and Letters on Reform. London: Bush.
Luhmann, Niklas. 2000. Die Politik der Gesellschaft. Frankfurt/M.: Suhrkamp.
Mack, Peter.  2002. Elizabethan Rhetoric: Theory and Practice. Cambridge:
Cambridge University Press.
Marchart, Oliver.  2007. Post-­ Foundational Political Thought. Edinburgh:
University of Edinburgh Press.
Markovits, Andrei S.  1987. Why There Is No Soccer in the United States?
Variationen zu Werner Sombarts große Frage. Leviathan 30: 486–525.
Martin, James.  2014. Politics and Rhetoric: A critical introduction. London:
Routledge.
May, Thomas Erskine. 1844. A Treatise Upon the Law, Privileges, Proceedings and
Usage of Parliament. London: Charles Knight. http://tinyurl.com/puv2sjd
———. 1883. A Treatise on the Law, Privileges, Proceedings and Usage of Parliament.
9th edition. London: Butterworths. http://tinyurl.com/oklmhdv
Mayring, Philipp.  2008. Qualitative Inhaltsanalyse. Grundlagen und Techniken,
10th edn. Weinheim [u.a.]: Beltz.
Meier, Christian.  1980. Die Entstehung des Politischen bei den Griechen.
Frankfurt/M: Suhrkamp.
Mergel, Thomas.  2002. Parlamentarische Kultur in der Weimarer Republik.
Düsseldorf: Droste.
Merleau-Ponty, Maurice. 1947 [1980]. Humanisme et terreur. Paris: Gallimard.
Mill, John Stuart.  1861 [1991]. Considerations on Representative Government.
Buffalo: Prometheus Books.
Mouffe, Chantal. 2005. On the Political. London: Verso.
———.  2013. Agonistics. Thinking the World Politically. London: Verso.
Nelson, John S. 1998. Tropes of Politics. Madison: University of Wisconsin Press.
Nelson, John S., Donald McCloskey, and Alan Megill.  1987. Rhetoric of the
Human Sciences. Madison: University of Wisconsin Press.
Nietzsche, Friedrich. 1887 [1981]. Zur Genealogie der Moral. In Werke II, ed.
Karl Schlechta, 761–900. Frankfurt/M: Ullstein.
———.  1995. Werke. Kritische Gesamtausgabe 2/4. Berlin: de Gruyter.
Oakeshott, Michael. 1975 [1991]. On Human Conduct. Oxford: Clarendon
Press.
Palgrave, Reginald. 1878. The Chairman’s Handbook. Suggestions and Rules for the
Conduct of the Chairmen in Public and Other Meetings, Drawn from the
References  237

Procedure and Practice of the Parliament. London: Knight & Co. http://
tinyurl.com/p2wloby
Palonen, Kari. 1993. Reading Street Names Politically. In Reading the Political,
eds. Kari Palonen, and Tuija Parvikko, 103–121. Helsinki: The Finnish Political
Science Association.
Palonen, Kari. 2002. Eine Lobrede für Politiker, Ein Kommentar zu Max Webers
‘Politik als Beruf’. Opladen: Leske+Budrich.
———.  2003. Four Times of Politics: Policy, Polity, Politicking and Politicization.
Alternatives: Global, Local, Political 28(2): 171–186.
———.   2006. The Struggle with Time. A Conceptual History of ‘Politics’ as an
Activity. Hamburg: LIT Verlag.
———.  2008. The Politics of Limited Times: The Rhetoric of Temporal Judgment in
Parliamentary Democracies. Baden-Baden: Nomos.
———.  2010a. ‘Objektivität’ als faires Spiel. Wissenschaft als Politik bei Max Weber.
Baden-Baden: Nomos.
———.  2010b. The Parliamentarisation of Elections. Redescriptions 14:133–156.
———.   2012a. Rhetorik des Unbeliebten. Lobreden auf Politik im Zeitalter der
Demokratie. Baden-­Baden: Nomos.
———.   2012b. Towards a History of Parliamentary Concepts. Parliaments,
Estates & Representation 32: 123–138.
———.  2013. Taking Distance as a Condition of Political Analysis. In The Distant
Present, eds. Tuula Vaarakallio, and Taru Haapala, 12–27. Jyväskylä: University
of Jyväskylä, SoPhi.
———.  2014a. Fair Play and Scarce Time: Aspects of the 1882 Procedural Reform
Debate in the British Parliament. In The Politics of Dissensus: Parliament in
Debate, eds. Kari Palonen, José María Rosales, and Tapani Turkka, 327–348.
Santander: Cantabria University Press/McGraw Hill.
———.  2014b. Politics and Conceptual Histories. Rhetorical and temporal perspec-
tives. Baden-Baden: Nomos.
———.   2014c. The Politics of Parliamentary Procedure. The Formation of the
Westminster Procedure as a Parliamentary Ideal Type. Leverkusen: Budrich.
———.   2016. From Oratory to Debate. Parliamentarisation of Deliberative
Rhetoric in Westminster. Baden-­Baden: Nomos.
Palonen, Kari, and Claudia Wiesner.  2016. Second Chamber, ‘Congress of
Ambassadors’ or Federal Presidency. Parliamentary and Non-parliamentary
Aspects in the European Council’s Rules of Procedure. Parliaments, Estates &
Representation 36: 71–89.
Palonen, Kari, José María Rosales, and Tapani Turkka, eds. 2014. The Politics of
Dissensus: Parliament in Debate. Santander: Cantabria University Press/
McGraw Hill.
Parvikko, Tuija. 2008. Arendt, Eichmann and the Politics of the Past. Helsinki: The
Finnish Political Science Association.
238  References

Pekonen, Kyösti. 2011. Members of the European Parliament as Delegates and as


Parliamentarians. In Parliament and Europe, eds. Claudia Wiesner, Tapani
Turkka, and Kari Palonen, 59–76. Baden-­Baden: Nomos.
Pekonen, Onni. 2014. Debating ‘the ABCs of Parliamentary Life’: The Learning of
Parliamentary Rules and Practices in the Late Nineteenth-­Century Finnish Diet
and the Early Eduskunta. PhD dissertation, University of Jyväskylä.
Peltonen, Markku.  2013. Rhetoric, Politics and Popularity in Pre-­Revolutionary
England. Cambridge: Cambridge University Press.
Perelman, Chaïm. 1977. L’empire rhétorique. Paris: Vrin.
———.  1987. Rhétoriques. Bruxelles: L’éditions de l’Université libre de Bruxelles.
———.   1989. Éthique et droit. Bruxelles: L’éditions de l’Université libre de
Bruxelles.
Perelman, Chaïm, and Lucie Olbrechts-­ Tyteca. 1958 [1983]. Traité de
l’argumentation. La nouvelle rhétorique. Bruxelles: L’éditions de l‘Université
libre de Bruxelles.
Pierre, Eugène. 1887. De la procédure parlementaire. Étude sur le mécanisme inté-
rieur du pouvoir législatif. Paris: Maison Quantin.
———. 1893 [1924]. Traité de droit politique électoral et parlementaire. Paris:
Librairies-Imprimeries Réunies.
Pocock, J.G.A. 1971 [1989]. Politics, Language and Time. Chicago, IL: University
of Chicago Press.
———1998. The Ideal of Citizenship Since Classical Times. In The Citizenship
Debates: A Reader, ed. Gershon Shafir, 31–41. Minneapolis: University of
Minnesota Press.
Popper, Karl R. 1934 [1971]. Logik der Forschung. Tübingen: Mohr.
Poudra, Jules, and Eugène Pierre. 1878. Traité de droit politique, électoral et par-
lementaire. Versailles: Cerf. http://tinyurl.com/nvkt9r9
Proksch, Sven-Oliver, and Jonathan B. Slapin. 2014. The Politics of Parliamentary
Debate. Parties, Rebels and Representation. Cambridge: Cambridge University
Press.
Rawls, John. 1993. Political Liberalism. Princeton, NJ: Princeton University Press.
Redlich, Josef. 1905. Recht und Technik des Englischen Parlamentarismus. Leipzig:
Duncker & Humblot.
———. 1908. The Procedure of the House of Commons: A Study of Its History and
Present Form. Transl. from German by A. Ernest Steinthal, with an introduc-
tion by Sir Courtenay Ilbert. Vol. 2. London: Constable.
Ricœur, Paul. 1971. The Model of the Text. Meaningful Action Considered as a
Text. Social Research 38: 529–562.
Robert‚ Henry M. 1876 [1915]. Robert’s Rules of Order (1915 edition). Chicago,
IL: Scott, Foresman. http://www.bartleby.com/176
Rohe, Karl. 1978 [1994]. Politik. Begriffe und Wirklichkeiten. Stuttgart:
Kohlhammer.
References  239

Roussellier, Nicolas. 2005. The Political Transfer of English Parliamentary Rules in


the French Assemblies (1789–1848). European Review of History 12: 239–248.
Samuels, Arthur P.I. 1923. The Early Life Correspondence and Writings of the Rt.
Hon. Edmund Burke. Cambridge: Cambridge University Press.
Sartre, Jean-Paul. 1960 [1985]. Critique de la raison dialectique. Paris: Gallimard.
Sartre, Jean-Paul. 1966 [1972] Plaidoyer pour les intellectuels. In Jean-Paul Sartre,
Situations VIII, 375-455. Paris: Gallimard.
Schmitt, Carl. 1922 [1979]. Politische Theologie. Berlin: Duncker & Humblot.
———. 1923/1926 [1979]. Die geistesgeschichtliche Lage des heutigen
Parlamentarismus. Berlin: Duncker & Humblot.
———. 1928 [1970]. Verfassungslehre. Berlin: Duncker & Humblot.
Schmitt, Carl. 1932 [1979]. Der Begriff des Politischen. Berlin: Duncker &
Humblot.
Scobell, Henry.  1656. Memorials of the Method and Manner of Proceedings in
Parliament in Passing Bills. London: Hills and Fields. EEBO Editions.
Simons, Herbert W.  1990. The Rhetorical Turn. Chicago, IL: University of
Chicago Press.
Skinner, Quentin.  1969. Meaning and Understanding in the History of Ideas.
History and Theory 8: 3–53.
———. 1970. Conventions and the Understanding of Speech Acts. Philosophical
Quarterly 20: 118–138.
———. 1978. The Foundations of Modern Political Thought I–II. Cambridge:
Cambridge University Press.
———. 1988. A Reply to My Critics. In Quentin Skinner and His Critics, ed.
James Tully, 231–288. Cambridge: Polity.
———. 1993. Two Concepts of Citizenship. Tijdschrift voor Filosofie 55: 403–419.
———. 1996. Rhetoric in the Philosophy of Hobbes. Cambridge: Cambridge
University Press.
———. 1998. Liberty Before Liberalism. Cambridge: Cambridge University Press.
———. 1999. Rhetoric and Conceptual Change. Finnish Yearbook of Political
Thought 3: 60–73.
———. 2002a. Visions of Politics. Vol. 1: Regarding Method. Cambridge: Cambridge
University Press.
———. 2002b. Visions of Politics. Vol. 2: Renaissance Virtues. Cambridge:
Cambridge University Press.
———. 2006. Rethinking Political Liberty. History Workshop Journal 61: 156–170.
———. 2007. Paradiastole. In Renaissance Figures of Speech, eds. Sylvia Adamson,
Gavin Alexander, and Karin Ettenhuber, 147–163. Cambridge: Cambridge
University Press.
———. 2008. Quentin Skinner Interviewed by Alan Macfarline 10th January
2008. https://www.repository.cam.ac.uk/bitstream/handle/1810/197060/
skinner.txt?sequence=1. Accessed 7 May 2016.
240  References

———. 2014. Forensic Shakespeare. Oxford: Oxford University Press.


Smith, Thomas. 1583. De republica anglorum. London: Midleton. http://tinyurl.
com/6scv7zs
Steiner, Jürg, André Bächtiger, Markus Spörndli, and Marco R. Steenbergen. 2005.
Deliberative Politics in Action. Analyzing Parliamentary Discourse. Cambridge:
Cambridge University Press.
Steinmetz, Willibald, Ingrid Gilcher-­Holtey, and Hans-Georg Haupt, eds. 2013.
Writing Political History Today. Frankfurt/M: Campus.
Taylor, Harriet.  1851. Enfranchisement of Women. Westminster and Foreign
Quarterly Review 54: 289–311.
Tiilikainen, Teija, and Claudia Wiesner. 2016. Towards a Political Theory of EU
Parliamentarism. In Parliament and Parliamentarism. A Comparative History
of a European Concept, eds. Pasi Ihalainen, Cornelia Ilie, and Kari Palonen,
292–310. Oxford: Berghahn.
Tomkins, Adam. 2005. Our Republican Constitution. Oxford: Hart.
Turkka, Tapani. 2007. The Origins of Parliamentarism. A Study of Sandys’ Motion.
Baden-Baden: Nomos.
Urbinati, Nadia.  2006. Representative Democracy. Concept and Genealogy.
Chicago, IL: University of Chicago Press.
———. 2014. Democracy Disfigured. Cambridge, MA: Columbia University Press.
Vaarakallio, Tuula. 2013. The Anti-­parliamentarism of the French Front National
Party. In The Distant Present, eds. Tuula Vaarakallio, and Taru Haapala, 28–47.
Jyväskylä: University of Jyväskylä, SoPhi.
———. 2015. The Borderline Between Parliamentary and Extra-­Parliamentary
Rhetoric: The Case Of The Populist (True) Finns Party. In Challenges to
Parliamentary Politics: Rhetoric, Representation and Reform, eds. Suvi
Soininen, and Tuula Vaarakallio, 99–124. Baden-­Baden: Nomos.
Vaarakallio, Tuula, and Taru Haapala. 2013a. Introduction. In The Distant Present,
eds. Tuula Vaarakallio, and Taru Haapala, 4–11. Jyväskylä: University of
Jyväskylä, SoPhi.
———, ed. 2013b. The Distant Present. Jyväskylä: University of Jyväskylä, SoPhi.
http://urn.fi/URN:ISBN:978-951-39-5102-3.
Vieira, Ryan. 2015. Time and Politics. Parliament and the Culture of Modernity in
Britain and the British World. Oxford: Oxford University Press.
Weber, Max. 1904 [1973]. Die ‘Objektivität’ sozialwissenschaftlicher und sozial-
politischer Erkenntnis. In Gesammelte Aufsätze zur Wissenschaftslehre, ed.
Johannes Winckelmann, 146–214. Tübingen: Mohr.
———. 1917a [1973]. Der Sinn der ‘Wertfreiheit’ in sozialen und ökonomischen
Wissenschaften. In Gesammelte Aufsätze zur Wissenschaftslehre, ed. Johannes
Winckelmann, 489-540. Tübingen: Mohr.
———. 1917b [1988]. Wahlrecht und Demokratie in Deutschland. In Max-
Weber-Studienausgabe I/15, ed. Wolfgang J. Mommsen, 155-189. Tübingen:
Mohr.
References  241

———. 1918 [1988]. Parlament und Regierung im neugeordneten Deutschland.


In Max-Weber-­Studienausgabe I/15, ed. Wolfgang J.  Mommsen, 202–302.
Tübingen: Mohr.
———. 1919 [1994]. Politik als Beruf. In Max-Weber-­Studienausgabe 1/17, ­eds.
Wolfgang J. Mommsen, and Wolfgang Schluchter, 35–88. Tübingen: Mohr.
———. 1922 [1980]. Wirtschaft und Gesellschaft. Johannes Winckelmann.
Tübingen: Mohr.
———. 2012. In Collected Methodological Writings, eds. Hans Henrik Bruun, and
Sam Whimster. London: Routledge.
Weiler, Josef H. 1996. The Selling of Europe. http://centers.law.nyu.edu/jean-
monnet/archive/papers/96/9603.html. Accessed 12 Feb 2013.
White, Hayden.  1973. Metahistory. Baltimore, MD: Johns Hopkins University
Press.
Wiener, Antje. 1998. European Citizenship Practice. Boulder, CO: Westview Press.
Wiesner, Claudia. 2007. Bürgerschaft und Demokratie in der EU. Münster: LIT.
———. 2008. Women’s Partial Citizenship. In The Ashgate Research Companion
to the Politics of Democratization in Europe: Concepts and Histories, eds. Kari
Palonen, Tuija Pulkkinen, and José María Rosales, 235–250. Farnham: Ashgate.
———. 2014a. Demokratisierung der EU durch nationale Europadiskurse. Baden-
Baden: Nomos.
———. 2014b. The Europeanisation of Citizenship: Conceptual Innovations,
Legal Changes, and Development of New Institutional Practices. Contributions
to the History of Concepts 9(1): 88–104.
———. 2014c. The European Parliament as Special Parliament and Political
Actor. In The Politics of Dissensus. Parliament in Debate, eds. Kari Palonen, José
María Rosales, and Tapani Turkka, 101–126. Santander: University of Cantabria
Press/McGrawHill.
Wilson, Jonathan.  2008. Inverting the Pyramid. A history of football tactics.
London: Orion.
Wittgenstein, Ludwig. 1953 [1971]. Philosophische Untersuchungen. Suhrkamp:
Frankfurt/M.
Wodak, Ruth, ed.  2001. Methods of Critical Discourse Analysis. London [u.a.]:
SAGE.
———, ed. 2008. Qualitative Discourse Analysis in the Social Sciences. Basingstoke
[u.a.]: Palgrave Macmillan.
Wood, Linda A., and Rolf O. Kroger. 2000. Doing Discourse Analysis. Methods for
Studying Action in Talk and Text. Thousand Oaks [u.a.]: SAGE.
Wright, Tony. 2012. Doing Politics. London: Biteback.
Zantovsky, Michael. 2014. Havel. A Life. New York: Grove.
Index

A argument, argumentation, ii, 1–3, 6,


adjournment, 8, 44, 52, 53, 55, 9, 13, 14, 19, 21–3, 26, 34, 40,
56, 66, 126, 159, 160, 180, 223 48, 49, 52, 53, 57, 60, 63, 65,
adversity, adversary, 9, 12–14, 16, 69–77, 81–4, 91, 93, 97, 98,
18–20, 22, 23, 35, 36, 47, 72, 101, 102, 107, 111, 115, 120,
73, 155, 220–2 122, 124–6, 131–41, 148, 150,
agenda, agenda–setting, v, 8, 11, 155, 156, 190, 197, 198, 202–5,
14–17, 19, 21, 23, 25, 213, 215, 224
30–3, 35–8, 41, 42, 44–7, Aristotle, vi, 14, 76, 224
51–4, 56, 57, 66, 73, 74, 78, 79, Assemblée nationale, 42, 110, 166
103, 112–4, 116, 119, 132, assembly, 14–16, 18, 23, 29–31, 33–5,
134–7, 139, 141, 143, 145, 154, 37–42, 45–9, 52, 56, 62, 70, 74,
156, 158, 161, 162, 167, 171–3, 156, 158, 162, 164–6, 170, 171,
182, 183 173, 177, 180
amendment, 13, 32, 44, 45, 49, 51–3, deliberative, 15, 16, 36, 41, 52, 62,
56, 66, 118, 134, 143, 144, 158, 156, 162, 165, 170, 173, 177
159, 162, 163, 168–70, 174, legislative, 41, 42, 48, 120, 138,
175, 180, 181, 208, 209, 222n2 162, 165, 169, 179, 183
Ankersmit, Frank, 38, 62, 64 representative, 42, 156, 162, 170
applause, acclamation, 36–40, 72, association, 19, 20, 34, 48, 49, 143,
112, 122, 214 155, 215, 217, 222
Arendt, Hannah, vii, 28, 57, 225 Attlee, Clement, 112, 142, 144, 151

Note: Page numbers followed by ‘n’ refer to foot notes.

© The Author(s) 2017 243


C. Wiesner et al., Debates, Rhetoric and Political Action,
DOI 10.1057/978-1-137-57057-4
244  INDEX

audience, 9, 12, 13, 16, 18–20, 22, 27, chance, 7, 9, 27, 45, 46, 64, 100, 115,
28, 36, 47, 67, 72–5, 122, 131, 135, 138, 143, 147, 150, 159,
132, 134, 137, 141, 156, 159 160, 162, 168, 170, 173, 181,
Austin, J.L., 4, 26, 71, 79 184, 219, 220
Churchill, Winston, 145, 150
citizenship, 80, 86, 110, 111, 145,
B 184–200, 222n4
backbench, backbenchers, 8, 37, 38, Clegg, Nick, 38, 112, 114, 117–9,
42, 45–7, 49, 52, 53, 55, 56, 66, 128, 129, 137, 138, 140, 141
112–5, 119–21, 123, 124, 129, clerk (of the House of Commons),
130, 138–41, 145, 182, 183 34, 45, 165
Bagehot, Walter, 15, 40, 46, 48, 56, clôture, closure, 8, 15, 33, 50, 53, 56,
120, 147 66, 176, 181
Ballot, secret voting, 39, 206 code, coding, 43, 82, 83, 94–7, 124,
Bentham, Jeremy, 17, 34, 43, 165, 125, 197, 226
166, 178, 179 committee (parliamentary), 13, 14,
bill (parliamentary), 32, 44, 46, 51, 19, 20, 27, 31, 42–4, 46, 47,
79, 112, 117, 119, 126, 127, 49–53, 56, 113–5, 117, 119–21,
129, 135, 144, 145, 156, 166, 126, 128, 150, 157, 158, 160,
175, 180, 181 162, 163, 166–75, 179, 181–3,
Brown, Gordon, 38, 111–20, 125–41 188, 213, 222n3
budget, 31, 45, 66, 74, 76, 163, 179 communism, communists, 28, 34, 40,
Bundesrat, 92, 208–10 76, 79, 105
Bundestag, 92, 98, 124, 168, 200–4, computer-supported analysis, 106–7
207, 208, 210–15 concepts, conceptions, v, vi, 2–6, 8, 10,
bureaucracy, administration, 5, 19, 26, 14, 15, 21–4, 26, 27, 29, 35, 37,
31, 46, 48, 49, 53, 55, 75, 85, 38, 41, 43, 47, 51, 60–3, 68, 72,
113, 164, 165, 171, 176, 179, 74–82, 99, 103, 104, 111–21,
189, 195, 226 124–6, 130, 132, 142–55, 162,
Burke, Edmund, 39, 170 177, 179, 183, 185, 186, 188,
Burke, Kenneth, 71, 75, 76 190, 191, 193, 198, 201,
213–5, 217, 224–6
conceptual analysis, conceptual history,
C 2, 27, 60, 67–70, 72, 77–80, 110,
cabinet, 46, 48, 116, 120, 136, 137, 142–55, 162, 185–6, 199, 217
174, 176 Congress (United States), 41, 45, 55,
Cambridge Union, vii, 9, 45, 75, 110, 164, 168, 170
156–61, 218 consensus, 15, 40, 63, 64, 100, 101,
Cameron, David, 38, 112, 114, 113, 137, 201, 202, 206, 210,
116–9, 127–9, 134–7, 139–41 211, 213, 215
Campion, Gilbert, 34, 37, 42–4, 51–3, constituency, 32, 39, 49, 113, 116,
55, 81, 166, 167 123, 138, 144, 145, 148, 155,
Carlyle, Thomas, 38, 54 170, 181, 183
INDEX  245

constitution, 19, 38, 57, 65, 76, 84–7, debating societies, 20, 48, 110,
91, 92, 100, 104, 105, 111–22, 155–61
124–7, 130, 131, 135, 136, decision–making, 5, 20, 26, 43, 46, 53,
138–9, 147, 164, 186, 187, 191, 54, 80, 91, 122, 127, 131, 137,
198, 200–5, 207–09, 211–4 160, 161, 163, 164, 176, 223
Constitutional Court, 43, 210, 211 deliberation, 14, 15, 21, 31, 39, 45,
contingency, vi, 7, 9, 10, 12, 25, 29, 49, 54, 117, 118, 130, 159, 164,
67, 111, 143, 147, 148, 150, 175–7, 214
165, 216, 217, 222, 225–7 deliberative rhetoric,
controversy, v, vi, 3, 7, 9, 14, 15, 17, See rhetoric deliberative
20, 21, 24–6, 29, 32, 35, 41, 43, De Mille, James, 9, 15, 30, 73
44, 48, 54, 63, 67, 77, 78, 87, democracy, v, 27, 30–2, 40, 49, 78,
111, 142, 143, 148, 150, 151, 80, 82, 113, 118, 124, 125,
154–6, 159–61, 165, 216, 217, 127–30, 134–8, 147, 149, 163–5,
225–7 190, 197, 198, 205, 225
conversation, 15, 29 democratisation, 31, 32, 49, 78, 80,
85, 144, 184
diplomacy, 14, 28, 29, 36, 37, 39, 40,
D 72, 151, 175, 176, 180, 182,
debater, 3, 4, 12, 13, 17, 18, 24, 27, 183, 193
29, 30, 34, 36, 47, 49, 64, 67, discourse, 6, 7, 14, 15, 18, 60, 61,
72, 73, 124, 181, 182 68–70, 84, 86–93, 95, 98–102,
debate types 104–6, 111, 122, 145, 200–2,
academic, 1, 14, 19, 21–3, 34, 35, 210, 212, 213, 215, 226
57, 60, 79, 86, 156 discourse analysis, theories, 2, 59,
on agenda and of agenda, 103, 183 60, 67–70, 77, 98, 102, 104,
documents as debates, 14, 19–21, 200, 224
24, 26, 35, 44, 57, 59–107, 161 discussion, 8, 9, 14–16, 20, 31, 40,
frozen, vi, 13, 14, 20, 156, 226 41, 43–5, 48, 51, 67, 74, 79,
live, 13, 14, 16, 18–21, 24, 88, 91, 92, 105, 106, 109,
156, 224 112, 114, 116, 119, 121,
multistage and-layer debates, 50–4 126, 130, 134–6, 139, 143,
oral, 23, 45 149, 151, 154, 157, 160, 162,
parliamentary, v, 2, 13–7, 19, 23, 163, 225
25–57, 60, 67, 73–5, 78, 79, 87, dispute, 1, 3, 4, 14–16, 22–4, 30, 34,
98, 110–55, 167, 176–9, 180, 35, 40, 63, 73, 77, 78, 82, 114,
185, 186, 191, 213, 214, 224 142, 195
press, 200–15 dissensus, dissent, 14–16, 56, 166,
public, 14–6, 18, 20, 21, 27, 47, 203, 205, 207, 210–15
49, 70, 71, 86, 88, 91, 155, distance, distanciation, 13, 18, 23, 24,
156, 200, 214 60–4, 80, 81, 101, 114, 154,
virtual, 1, 18–19 202, 215
written, 19, 23, 24, 27 distribution of time, 8, 55, 176, 177
246  INDEX

division (vote by), 2, 23, 50, 172 F


document analysis, 81–96, 161–200 fair play, fairness, 10, 16, 30, 36, 41–3,
documents, 1, 2, 14, 18–21, 24, 26, 46, 55, 63, 81, 156, 160, 172,
35, 44, 57, 59–107, 110, 111, 177, 216, 222
161, 165, 184–200, 215, 224, figures. See rhetorical, figures
226, 227 forensic rhetoric. See rhetoric,
forensic
Foucault, Michel, 60, 68, 69, 200
E freedom, liberty, 41, 76, 78, 80,
election, 8, 12, 22, 28, 29, 33, 37, 38, 127, 171, 173, 188, 189, 195,
44, 47, 54, 78, 105, 110, 197, 198
112–19, 121, 128, 129, 135, free and fair elections, 41,
136, 141, 162, 170, 171, 180, 78, 198
184, 193, 224 free mandate, 16, 40, 78,
electorate, 47, 126, 134, 137, 170–3, 182
141, 184 free speech, 36, 53, 76, 78, 176
eloquence, oratory, 9, 31, 38, 48, 70,
73, 74, 155, 182
epideictic rhetoric. See rhetoric, G
epideictic Gladstone, William, 31, 43, 54
equality, 149, 198 guillotine (parliamentary), 8, 50, 53,
European Commission, 111, 162–5, 56, 176, 177, 181
168–71, 173–84, 186–8, 190–5,
197, 222n4, 224
European Council, 29, 39, 65, 92, H
163, 164, 174, 180, 187–9, 192, Habermas, Jürgen, 40, 215
197, 199, 222n Hamilton, William Gerard, 8, 75, 78,
European Parliament, 10, 11, 17, 42, 153, 154
46, 47, 50, 110, 161–84, 186, Hatsell, John, 42, 43, 52, 53, 166
187, 191, 195–7, 223 House of Commons, lower House, vii,
European Union EU, 4, 11, 12, 26, viii, 8, 32, 37, 42, 43, 46, 75, 80,
29, 40, 45, 84–107, 110, 111, 98, 110–4, 117, 119–40, 142–55,
138, 161–205, 207, 209–13, 165, 167, 222n3
215, 217, 222, 223 House of Lords, upper House, 49, 51,
citizenship, 86, 110, 111, 184–200, 113, 126, 132, 133, 135, 144,
222n4 145, 148, 151, 152
Council of the European Union
(of ministers), 29, 111, 162–4,
169, 173–6, 179–83, 186– I
9, 191, 193, 195–9, 222n, 223 ideal type, 12, 13, 18, 21–3, 25, 27,
EU elections, 22, 110, 162, 170, 29–41, 64, 76, 84, 94, 103,
180, 184, 193 161–3, 165, 166, 215
INDEX  247

interests, interest groups, 1, 5, 6, 20, 133, 139, 156, 160, 162, 171,
29, 37, 39, 40, 42, 59, 60, 82, 178, 199, 201–3, 205, 206, 209,
97, 104, 106, 111, 123, 126, 212–4
131, 134, 150, 160, 163, 164, Martin, James, 7, 67, 72
170, 176, 178, 191, 199. See material selection(s), 59, 88,
research, interest 90–3, 200
interjection(s), 32, 33, 74, 206 May, Thomas Erskine, 32, 33, 43, 45,
interpretation, 1, 11, 12, 17, 18, 21, 52, 53, 165, 167
22, 30, 34, 40, 43, 44, 47, 55–57, member of parliament. See
60, 63, 64, 68, 74, 76, 80–3, 96, parliamentarian /MP
110, 119, 132, 143, 146, 147, of Bundestag, 92, 98, 105, 124, 164,
152, 155, 158–61, 184, 185, 187, 168, 200–5, 207, 208, 210–5
188, 194, 199, 217, 224 of European Parliament, (MEP),
interpretative 164, 171, 173, 177, 178, 181
approach, analysis, 59, 67, 81–4, 90, of House of Commons/Lords, 32,
93, 96, 98, 107, 125, 226 43, 81, 113, 120, 128, 138,
strategies, 59, 70, 82, 93, 97, 109 147, 148, 153, 155
techniques, tools, 22, 60, 109 metaphor, vii, 6, 10, 65, 71, 75, 76,
interruption(s), 33, 38, 52, 66 226, 227
irony, 71, 75, 76, 81, 149 Mill, John Stuart, 32, 38, 52
item (on the agenda), 14, 19, 23, 31, 33, minister, 5, 28, 29, 40, 45, 47, 111,
41, 45, 51, 52, 56, 73, 119, 134 119, 135, 150, 151, 163, 164,
174, 175, 181, 184, 186, 187,
204, 206, 209–11, 222n1
K minority, 13, 31, 35, 50, 145, 156,
Koselleck, Reinhart, vi, 8, 54, 61, 62, 159, 205, 206
65, 77, 78, 142, 184 motion (parliamentary), 8, 9, 14–6,
18, 23, 30, 32–4, 37, 44–6, 49,
51–4, 66, 79, 112, 113, 118,
L 121, 138, 145, 156, 159, 160,
Laclau, Ernesto, 68, 69, 77 166, 168–70, 175, 180, 181,
legislation, 9, 31, 32, 45, 46, 54, 65, 183, 222n2, 223
79, 112, 114, 120, 126, 131, Mouffe, Chantal, 7, 68, 69, 77
133, 143, 156, 163, 164, 167–9,
174, 175, 182, 183, 187, 189,
190, 195–7, 204, 223 N
naming, 5, 11, 24, 36, 50, 76, 195,
215, 218
M negotiation, 14, 28, 36, 37, 39, 40,
majority (parliamentary), 2, 13, 15, 42, 72, 151, 172, 175, 176, 179,
22, 23, 31, 35, 47, 50, 52, 56, 180, 182, 191, 203, 213
57, 66, 80, 114, 116–8, 126, Nietzsche, Friedrich, 21, 22, 30, 63, 72
248  INDEX

O 79, 87, 98, 110–44, 151, 154,


objectivity, 4, 21, 63, 73, 81 176, 180, 185, 186, 213, 224
obstruction, 43, 50, 54–6, 178 eloquence, oratory (see eloquence,
opinion, 7, 13, 16, 20, 30, 34, 38, 53, 56, oratory)
86, 100, 111, 123, 145, 149, 150, freedom, 36, 41, 78, 80, 171, 172
156, 166, 168, 169, 171, 173, 174, government, 30, 40, 45, 46, 48–50,
178, 180–2, 202, 207, 214, 225 54, 55, 111–41, 163, 175, 176,
opposition (parliamentary), v, 3, 8, 13, 184, 199, 225
16, 24, 37, 38, 40, 45–7, 50, 55, initiative, 45–7, 120, 174, 175,
66, 76, 111, 112, 114, 116, 118, 180, 182, 185, 195, 197, 199,
119, 122–5, 127, 128, 134, 136, 211, 214
137, 140, 141, 145, 146, 149, procedure (see procedure, rules of
150, 154, 171, 173, 178, 201, procedure)
202, 207–9, 226 questions, 32, 33, 37, 43–6, 52, 54,
order (parliamentary), 8, 15, 16, 31, 66, 75, 111, 123, 124, 129,
33, 34, 38, 42, 43, 50, 66, 148, 140, 143, 145, 172–4, 184
151, 159, 176, 178, 179 records, protocols, 19, 33, 83, 87,
Oxford Union, vii, 9, 45, 75, 110, 88, 176
156–60 reform, 31, 32, 38, 48, 79, 144–52
rhetoric, 31–3, 42, 47, 70, 73–5,
110, 131–41, 150, 153, 169
P sovereignty, 132, 134, 137–41, 174,
pamphlet, 14, 19, 21, 26, 57, 88 175
paradiastole, 22, 72, 76, 110, 153 time, 8, 31, 34, 37, 41, 51–7, 73,
parliamentarian /MP, 6, 8, 16, 30, 32, 114–8, 146, 162, 176, 177,
33, 38–40, 43, 46, 47, 49, 50, 180, 181, 211, 214
53–5, 77, 80, 113, 120, 121, party
125, 128, 136, 146–8, 152, 154, conference, 16, 18, 29, 47, 208
155, 162, 171, 178, 181–3 discipline, whips, 37, 121, 214
parliamentarism, parliamentarisation, leader, leadership, 12, 118, 119,
47, 49, 75, 119, 184 123, 146, 182, 202, 209, 212
parliamentary manifestos, programmes, 18, 19, 35,
agenda, 8, 14–7, 30–2, 35, 37, 42, 46, 57, 65, 82, 87
44–7, 51–4, 73, 74, 76, 103, Perelman, Chaim, 71, 75, 76
112–4, 116, 119, 132, 134–6, performance (rhetorical), 10, 11, 16,
139, 141, 143, 145, 154, 162, 17, 19, 23, 72, 74, 171
163, 167, 168, 171, 173, perspective, perspectivism, vi, 2, 4–6,
182, 183 15, 16, 20–5, 30, 31, 35, 38, 41,
calendar, 11, 54–6, 180 51, 54, 57, 60–4, 69, 70, 73,
committee (see committee 77, 78, 82, 85, 87, 92, 105,
parliamentary) 110, 111, 115, 131, 147, 151,
debate, 2, 13, 15, 16, 19, 23, 25, 162, 165, 183, 198, 215, 216,
27, 29–45, 54, 67, 73–5, 78, 223, 225
INDEX  249

persuasion, 3, 13, 14, 18, 22, 28, 39, politicisation, 9–12, 15, 26, 27, 29,
47, 70, 72–5, 131, 132, 137, 65, 78, 143, 144, 165, 217, 219,
140, 156, 158, 159, 161, 173, 220, 226
203, 212 politicking, 4, 9–12, 29, 65, 73, 110,
platform, 47, 149, 155, 172 143, 144, 152–5, 217, 219, 220
plenum, plenary, 8, 14, 27, 31, 32, 41, politics
45, 50–3, 56, 112, 144, 162, as activity, v, vi, 1, 2, 5–15, 23, 143,
166–70, 173, 175, 177, 181, 182 216, 223–7
policy, 1, 2, 7, 9–11, 14, 29, 35, 49, language of politics, 3–4, 12, 13,
65, 112, 142–5, 150, 184, 186, 23, 62, 67–73, 110, 132,
219, 220 144–6, 149, 151, 152, 154,
policy papers, documents, 18–20, 161, 216, 218, 224–7
87, 110, 111, 184–200 party politics, 146–8
polis, vi, 185, 224, 225 as sphere, 5, 6, 10, 22, 110,
political 148, 151
action, activity, v, vi, 1–3, 5, 6, 8, polity, 1, 9–12, 29, 65, 112, 124, 142,
10, 12, 13, 20, 24, 25, 28, 143, 145, 146, 150, 151, 185,
59, 60, 62, 64, 67, 69, 70, 81, 186, 189, 215–22, 224
96, 132, 147, 150, 156, 157, popular sovereignty, 132, 137, 138,
161, 185, 201, 213, 216, 221, 140–2
224, 225 president
actors, agents, v, 1–5, 22, 23, 62, of parliament, 17, 34, 37, 42, 44,
65, 110, 143, 149–51 49, 50, 92, 100, 125, 162,
controversy, 48, 150, 151, 161 164, 168, 171–80, 182, 183,
judgement, 149, 151 212, 215
literacy, 2, 4, 25–9, 57, 106, 109, of state, 92, 192, 212, 215
110, 226–8 presidentialism, 37, 48, 100, 164
science, theory, 9, 10, 16, 18, previous knowledge, 90, 95, 96
62, 67, 71, 72, 113, 120, prime minister, 29, 43, 74, 111, 112,
149, 152 114, 116–25, 127, 129–32,
struggle, 8, 26, 79, 84, 161, 187, 134–41, 163, 174, 175, 182, 191
188, 193, 194, 196, 199 procedure, rules of procedure, vii, 8,
thinking politically, 6, 30, 35, 225 15–7, 19, 30, 33, 34, 37–9, 41–6,
time, 41, 148, 151, 161, 180 53, 54, 63, 66, 67, 70, 73, 80,
politician, v, 5, 8, 9, 17, 21, 23, 81, 87, 103, 110, 111, 113, 118,
26–30, 35, 48, 49, 63, 65, 75, 119, 123, 132, 133, 143, 155,
147–8, 101, 115, 116, 137, 138, 156, 159–84, 187, 196, 197,
141–4, 146–50, 152, 154, 161, 204, 222n3
181–3, 201–3, 210, 213, 215, pro et contra, for and against, 2, 13,
226, 227 14, 16–8, 29–31, 39, 47, 49,
occasional, 5, 27, 28, 35, 147, 226 50, 62, 70, 166, 176, 184,
professional, v, 6, 28, 55, 75, 78, 213, 214
147, 150, 152, 154, 157, 225 public meeting, 15, 18, 27, 47, 71, 88
250  INDEX

Q epideictic, 14, 36, 37, 40, 42, 47,


qualitative 48, 72, 131, 133, 140
analysis, 84, 88 forensic, 14, 26, 42, 43, 71, 72, 131,
research, 82, 102 132, 135, 140, 141
quantitative analysis, quantitative text rhetorical
analysis, 81 figures, 60, 72–7, 103, 114, 148,
151
genres, 14, 15, 41, 42, 72, 131, 132
R strategies, 1, 71–3, 121, 153, 158,
rapporteur, 47, 167, 169, 170, 181, 182 159, 223, 224
readings (parliamentary), 44, 51, 55, topoi, 26, 57, 60, 66, 72, 74, 75,
110, 112, 125, 144, 145, 166, 97, 103, 115, 117, 132–4, 137,
168, 207, 213, 214 140, 141, 149, 184
reality, 21–3, 43, 73, 86, 94 tropes, 60, 71–7, 83
Redlich, Josef, 17, 32, 43, 55, 81, 156 ridicule, 6, 79, 80, 117, 136
referee, 17, 34, 49, 50, 180, 217 rights, 49, 76, 78, 110, 111, 113,
reply (parliamentary), 18, 32, 50, 53, 135, 156, 162, 173, 175,
74, 112, 117, 123, 128, 129, 131, 184–200, 204, 217, 222n4
134, 136, 138, 166, 168, 182 Rousseau, Jean-Jacques, 38, 183
representation (political), 16, 38, 39, 42,
50, 62, 78, 80, 110, 116, 118, 123,
133, 137, 139, 145, 149, 154, 156, S
162–4, 170, 173, 176–80, 182, sampling, 83, 88, 89, 91–3, 95, 96, 102
184, 185, 190, 201, 208, 225 Sartre, Jean-Paul, 5, 40
research scarcity (of time), 8, 54–6
as activity, 2, 20, 59, 60, 62, 64, 67, Schmitt, Carl, 4, 6, 38, 54, 73, 153, 154
70, 132, 223–7 Scobell, Henry, 32, 53
design, 65, 85, 87, 90, 95, 200, Skinner, Quentin, 3, 17, 21, 22, 26,
224, 226 35, 60, 61, 63, 67, 68, 71, 72,
interest, 2, 19, 59, 70–2, 84–8, 103, 76, 78, 79, 153, 154, 171, 184,
109, 152, 158, 164, 224 191, 215
materials, 87–90, 157 socialism, socialists, 39, 49, 81, 104,
methodology, 59, 68, 69, 82 146, 201, 207
practices, 2, 59–107 speaker (of parliament), 8, 16–9, 32,
question, 52, 59, 82–90, 95, 102, 103, 33, 36, 38, 43, 44, 47, 49, 50,
109, 122, 131, 157, 224, 226 52, 54, 66, 80, 124, 159, 177
responsibility (to the parliament), 29, speech acts
48, 104, 124, 126, 128, 136, illocutionary, 4, 26, 60, 79
137, 140, 141, 163, 169, 171, locutionary, 4, 26, 79
174, 176 speech competitions, 13, 29, 36, 71
rhetoric standing orders, 42, 161
deliberative, 14, 15, 31, 39, 42, 72, strategy, 18, 35, 52, 73, 93–6, 105,
131–5, 140 135, 143, 159, 160, 200–15, 220
INDEX  251

street names, 5, 24, 215 urgency, 43, 53, 54, 112–5, 128, 132,
Suffrage, franchise, 32, 33, 78–80, 136, 143, 156, 174, 175, 210
144, 149, 225

V
T vote of confidence, 37, 45, 46,
tactics, 35, 45, 47, 111, 151, 165, 133–4, 175
217–219, 221 voter(s), 18, 27, 32, 62, 151, 163,
talk, 12, 15, 23, 29, 54, 55, 62, 135, 184
139, 176, 205 vote, voting (as a procedure) 2, 13,
textbook, 3, 21, 23, 26, 63 17, 18, 22, 30, 31, 33–6, 38–40,
theatre, 16, 27, 61 44–50, 53, 56, 78, 80, 85, 91,
theoretical relevance, 83, 88, 91 98, 118, 133, 144–6, 156–60,
theoretical sampling, 88, 93, 95, 96, 102 162, 163, 166, 169–76, 180–5,
topos, 66, 74, 115, 117, 118, 132– 187, 223–5
134, 137, 141, 183
trope, 60, 71–7, 83
W
Weber, Max, v, vii, 21–3, 27–30,
U 32, 35, 39, 49, 55, 63, 64, 73,
unparliamentary language/ conduct, 81, 147, 162, 163, 170, 215, 225,
16, 33, 42–4, 66, 80, 178, 226
179, 183 Wodak, Ruth, 68, 69

Vous aimerez peut-être aussi