Вы находитесь на странице: 1из 90

Volume Six Number Three

ISSN 1476-413X

Portuguese Journal of Social Science | Volume Six Number Three

Portuguese Journal of Social Science
Volume 6 Number 3 – 2007 6.3
137–146 Open the social sciences: To whom and for what?
Michael Burawoy

147–154 ‘Open the social sciences: To whom and for what?’, by Michael Burawoy
José Madureira Pinto
155–169 ‘Fascist lackeys’? Dealing with the police’s past during Portugal’s transition
to democracy (1974–1980)

Journal of
Diego Palacios Cerezales
171–191 From Porto to Portadown: Portuguese workers in Northern Ireland’s
labour market
Martin Eaton

Social Science
193–211 Minority representation in Portuguese democracy
André Freire

213–219 Francisco Javier Luque Castillo

221 Index – Volume 6

The Portuguese Journal of Social Science is published by intellect in partnership with:

intellect Journals | Media & Culture

ISSN 1476-413X

9 771476 413007 www.intellectbooks.com

PJSS-6-3-00-FM 3/14/08 2:32 AM Page 135

Portuguese Journal of Social Science

Volume 6 Number 3

The Portuguese Journal of Social Science (PJSS) is an English language, peer Editor
reviewed scholarly journal by the Higher Institute for Labour and Business João Ferreira de Almeida
Studies (Instituto Superior de Ciências do Trabalho e das Empresas – ISCTE), a
state-funded university institute located in Lisbon, the largest in Portugal doing
social science research. Av. das Forças Armadas
The journal is dedicated primarily to introducing an international readership to Edifício ISCTE
the best work currently being produced by Portuguese scholarship in the social sci- 1649–026 Lisboa
ences. This work – until the publication of PJSS – was only available in Portuguese Portugal
or was scattered among a variety of different non-Portuguese language journals: Tel/Fax: +351 217 903 473
much important work – of general interest to the social science commu-nity as well E-mail: ferreira.almeida@iscte.pt
as of particular interest to specialists in Latin America and southern Europe – has
tended until now to be neglected by readers who do not readily read Portuguese or
who do not have access to Portuguese language journals.
Editorial Secretary
While preference will be given to original work, the PJSS will consider excep- Stewart Lloyd-Jones
tional pieces that have previously been published in languages other than CPHRC
English. All articles will be submitted to two referees, nominated by the Editorial 506 Strathmartine Road
Committee, for peer review. The principal academic disciplines covered by the Dundee DD3 9BR
PJSS include anthropology, economics, history, social psychology, sociology, Scotland UK
political science and social geography. Tel: +44 1382 832525
Fax: +44 1382 526643
Editorial Committee E-mail: stewart.lloyd-jones@iscte.pt
João Ferreira de Almeida
Maria Eduardo Gonçalves
Emanuel Leão
António Costa Pinto
José Manuel Leite Viegas

Editorial Board
Onésimo T. Almeida – Brown University, USA
Maurice Aymard – Maison des Sciences de l’Homme, Paris, France
Michael Billig – Loughborough University, UK
Robert Boyer – CEPREMAP, Paris, France
Willem Doise – University of Geneva, Switzerland
Schmuel Eisenstadt – Hebrew University of Jerusalem, Israel
Michael Herzfeld – Harvard University, USA
Max Kaase – International University Bremen, Germany
Kenneth Maxwell – Harvard University, USA
Phillipe C. Schmitter – European University Institute, Florence
Lady Margaret Sharp – University of Sussex, UK
Verena Stolcke – Autonomous University of Barcelona, Spain
Gilberto Velho – Federal University of Rio de Janeiro, Brazil
Stuart Woolf – University of Venice, Italy
Leonardo Morlino – University of Florence, Italy
Michael Burawoy – University of California, Berkeley, USA

ISSN 1476-413X
The Portuguese Journal of Social Science is published three times a year by Intellect Ltd,
The Mill, Parnall Road, Fishponds, Bristol, BS16 3JG, UK. The current subscription
rates are £30 (personal) and £210 (institutional). A postage charge of £10 is made Articles appearing in this journal are
for subscriptions outside of Europe. Subscriptions, enquiries and bookings for abstracted and indexed in Cambridge
advertising should be addressed to the: Marketing Manager, Intellect Ltd, The Mill, Scientific Abstracts, International
Parnall Road, Fishponds, Bristol, BS16 3JG, UK. Bibliography of the Social Sciences (ISBSS).
© 2007 Intellect Ltd. Authorisation to photocopy items for internal or personal use
or the internal or personal use of specific clients is granted by Intellect Ltd for
libraries and other users registered with the Copyright Licensing Agency (CLA) in the
UK or the Copyright Clearance Center (CCC) Transactional Reporting Service in the Printed and bound in Great Britain by
USA provided that the base fee is paid directly to the relevant organisation. 4edge, UK.
PJSS-6-3-00-FM 3/14/08 2:32 AM Page 136

Editorial Notes for

The aim of this Journal is to provide a diagrams, etc. should be supplied in a services. Submissions received without
medium for the publication of original camera-ready format as TIFF files with a keywords will be rejected.
papers covering Portuguese thought and minimum resolution of 300dpi. If in doubt,
research on social sciences. The Editorial then please contact the editorial secretary Notes
Committee is particularly keen to publish at stewart.lloyd-jones@iscte.pt who will be Notes appear at the side of appropriate
work on current developments in research able to advise you in the preparation of pages, but the numerical sequence runs
and analysis. camera-ready TIFF files. throughout the article. There should be no
All submissions should, in the All illustrations, photographs, diagrams, more than 12 notes in any article, and these
first instance, be sent as an email maps, etc. should follow the same numerical should be identified by a superscript numeral
attachment to the Editorial Secretary at sequence and be shown as Figure 1, immediately following the next occurring
stewart.lloyd-jones@iscte.pt. Correspondence Figure 2 etc. The source should be punctuation mark.
and books for review should be sent to the indicated below. Copyright clearance should
Editor: João Ferreira de Almeida, be indicated where appropriate by the
References and Bibliography
We use the Harvard system for bibliographi-
UNICS/ISCTE, Av. das Forças Armadas, contributor who has sole responsibility for
cal references. This means that all
1649–026 Lisbon, Portugal. ensuring the relevant clearances have been
quotations must be followed by the name of
received. All illustrations must be included
Referees the author, the date of the publication, and
as separate attachments to the submission,
The Portuguese Journal of Social Science is a the pagination, thus: (Santos 1995:254).
they must be clearly labelled and their
refereed journal. Strict anonymity is accorded PLEASE DO NOT use ‘(ibid.)’. Note that the
position within the text clearly identified.
to both authors and referees. There are punctuation should always follow the
normally two referees, chosen for their Quotations reference within brackets, whether a
expertise within the subject area. They are Within paragraphs, these should be used quotation is within the text or an indented
asked to comment on comprehensibility, sparingly, identified by single quotation quotation. Your references refer the reader
originality and scholarly worth of the article marks. Paragraph quotations must be to a bibliography at the end of the article.
submitted. indented with an additional one-line space The heading should be ‘References’. List the
above and below and without quotes. items alphabetically. Here are examples of
Length the most likely cases:
Articles should normally be between 6500 Captions
and 7500 words in length. All illustrations should be accompanied by a Almeida, J. Ferreira de (1982), Classes
caption, which should include the Fig. No., sociais nos campos: camponeses parciais numa
Submitting and acknowledge the holder of the copyright. região do noroeste, 2 vols., Lisbon: ICS.
Articles should be original and not be under The author has a responsibility to ensure Amâncio, L. (1997), ‘The importance of
consideration by any other English language that the proper permissions are obtained. being male: Ideology and context in gender
publication. Submissions must be in English
Identities’, Revue Internationale de Psychologie
and be written in a clear and concise style. Other Styles Sociale, no. 2, pp. 79–94.
In the first instance, contributions should Foreign words and phrases within the text
be submitted electronically in MS Word must be in italics. Jargon should be kept to Pinto, A. Costa (ed.), (1998), Modern
format to the editorial secretary at an absolute minimum. Invented words Portugal, Palo Alto: Society for the
stewart.lloyd-jones@iscte.pt. Submissions must should be avoided at all costs—unless they Promotion of Science and Scholarship.
include the following: a note on the each are in standard use within the field of study,
author, each author's institutional affiliation when they should be explained to a non- Web References
(including address, contact telephone and specialist audience either by the inclusion of These are no different from other references;
email), an abstract of no more than 150 a footnote, an explanation within the text, they must have an author, and the author
words, and up to six keywords. or a glossary at the end of the article. must be referenced according to the
Chicago author-date system within
Language Author(s) Note the text. Unlike paper references,
The Journal uses standard British English. A note on the author is an absolute however, web pages can change, so
The Editor reserves the right to alter usage requirement. The author's note must we need a date of access as well as
to these ends. Because of the interdiscipli- include the author’s name, institutional the full web reference. In the list of
nary nature of readership, jargon is to be affiliation and address. Papers received references at the end of your article, the
avoided. Simple sentence structures are of without this information will be rejected. item should read something like this:
great benefit to readers for whom English is
a second language. Abstract Freitas, E. de, and Ávila, P. (2000),
The abstract should not exceed 150 words ‘Inquérito à Cultura Científica dos
Format in length and should concentrate on the Portugueses 2000’,
All submissions must be in MS Word format, significant findings. Apart from its value for http://www.oct.mct.pt/actividades/cultura/
or in a format that can be read by MS Word. abstracting services, it should also make a cultura2000/contributos/inquerito/docs/
case for the article to be read by someone relatorio.doc, Lisboa: Observatório das
Illustrations from a quite different discipline. Ciências e Tecnologias. (Accessed 11
Generally only black and white reproduction Submissions received without an abstract December 2001.)
is available. Photographs should be black will be rejected.
and white and glossy. All slides should be If you have any queries about referencing,
printed as colour photographs or scanned as Keywords or any other matter of style, then please
a greyscale TIFF file with a minimum Provision of up to six key words is much contact the editorial secretary at
resolution of 600dpi. Line drawings, maps, appreciated by indexing and abstracting stewart.lloyd-jones@iscte.pt.

Any matters concerning the format and presentation of articles not covered by the above notes should be addressed to the editorial
secretary at stewart.lloyd-jones@iscte.pt. The guidance on this page is by no means comprehensive: it must be read in conjunction
with Intellect Notes for Contributors. These notes can be referred to by contributors to any of Intellect’s journals, and so are, in
turn, not sufficient; contributors will also need to refer to the guidance such as this given for each specific journal. Intellect Notes
for Contributors is obtainable from www.intellectbooks.com/journals, or on request from the Editor of this journal.
PJSS-6-3-01-Burawoy 3/15/08 2:18 PM Page 137

Portuguese Journal of Social Science Volume 6 Number 3. © Intellect Ltd 2007.

Article. English language. doi: 10.1386/pjss.6.3.137/1

Open the social sciences: To whom and

for what?*
Michael Burawoy University of California

Abstract Keywords
The Gulbenkian Commission Report (1996) on the restructuring of the social sci- social science
ences disavowed anachronistic disciplinary divisions, Western universalism and Gulbenkian
methodological positivism, and instead proposed the unification of all scientific Commission
knowledge under what it called ‘pluralistic universalism’. It exposed its own scholas- public sociology
ticism, however, in failing to address for whom and for what is scientific knowledge policy sociology
produced. With these two questions as points of departure, this article develops a dis- critical sociology
ciplinary division of labour, and thereby distinguishes among professional, policy, professional sociology
public and critical knowledge. Examining the form and relations among these four
types of knowledge allows one to recognise the real basis of divergences among disci-
plines, and within disciplines across nations and history. A global perspective on the
social sciences today examines the specific responses to market fundamentalism
from different disciplines and different places in the world system.

It is exactly ten years since the Gulbenkian Commission published its * This is a revised
version of the lecture
report on the restructuring of the social sciences. Chaired by Immanuel to the Portuguese
Wallerstein, the Commission consisted of ten distinguished scholars from Sociological
the natural sciences, humanities and the social sciences. Their report, Association delivered
at the University of
Open the Social Sciences, was widely publicised throughout the world as Lisbon’s Institute of
innovative, pointing towards a future that would dissolve outdated discipli- Social Science (ICS-
UL) on 30 March
nary divisions within the social sciences, while making their unification 2006. I would like to
the locus of an ambitious reconciliation of the humanities and natural sci- thank João Ferreira de
Almeida, Elisio
ences. The Commission attributed the backwardness of the social sciences Estanque, José Virgilio
to a lingering attachment to ideas, methodologies and divisions that Pereira, José
Madureira Pinto,
marked their birth in the 19th century. These antiquated notions, the Boaventura de Sousa
Commission noted, began to break down after 1945 laying the founda- Santos and Anália
tions for an anticipated integration of all scientific knowledge. Driving this Torres for their
rupture with the past would be the rational development of social science,
unhindered by false epistemologies and vested interests.
The Commission flattered scientific knowledge with its own autonomous
history. For such autonomy is illusory – a distorted expression of the privi-
leged existence that prevails only at the pinnacle of Western academe, and
of little relevance to most social scientists, embedded in contexts increas-
ingly driven by what I call third-wave marketisation. The Gulbenkian
Commission was the project of an elite cut off not only from the actual

PJSS 6 (3) 137–146 © Intellect Ltd 2007 137

PJSS-6-3-01-Burawoy 3/15/08 2:18 PM Page 138

practice of the social sciences, but also from the real world problems
those sciences are designed to investigate: not to mention from the people
affected by those problems. Rather than opening the social sciences,
the Gulbenkian Commission was effectively closing them off, not only to
the global south but also to most of the global north. Head stuck in the
sand, the Commission was disarming the social sciences as it faces search-
ing challenges to its viability.
Settling accounts with the Gulbenkian Commission is long overdue.
We need to rethink the social sciences, not from the top down but from the
ground up, rooting them in the multiple contexts of their production. We
need to dispense with imaginary utopias divorced from everyday practices
and explore the concrete division of labour within and between the social
sciences. We cannot quarantine the social sciences, refusing their dissec-
tion for fear of disturbing a hornet’s nest. We cannot exempt ourselves
from the investigative eye we so gleefully turn upon others. If sociology, in
particular, can disclose to others the public issues that underlie their
private troubles, why can it not do the same for itself, turning private
antagonisms into public debate. To transcend the divisions that divide us,
or, at least, turn those divisions in a constructive direction, we have to
trace them to different locations and trajectories within and through the
scientific field. Spelling out the parameters and dimensions, the patterns of
domination and interdependence within and among scientific fields
should foster a more effective presence in the world beyond.
We begin, therefore, by endorsing the Gulbenkian Commission’s identifi-
cation of three problems that beset the social sciences, and the Commission’s
identification of three corresponding empirical trends. We then reinterpret
those trends not from the rafters of the ivory tower but from the grounded
laboratories of social science production – laboratories understood as fields
of force operating in a world historical context.

Three problems, three trends and a totalising utopia

The Gulbenkian Commission identified three significant issues that must
be at the heart of any rethinking of the social sciences: (1) the false uni-
versalism of Western thought that had underpinned the social sciences;
(2) the anachronistic division of the social sciences divided by their objects
of knowledge; and (3) a misguided positivist methodology that still domi-
nated the practice of the social sciences.
These three problems were corroborated and accentuated by three cor-
responding historical tendencies identified by the Gulbenkian Commission.
First, feminism, anti-racism and anti-colonial thinking attacked the social
sciences as universalising the experiences of particular societies, namely
Europe and the United States, and even more narrowly of hegemonic
groups within these societies. Second, the advance of inter-disciplinary
programmes and journals as well as area studies signalled the anachro-
nism of divisions within the social sciences, divisions only maintained by
retrograde disciplinary organisations. Third, narrow positivist methodology,

138 Michael Burawoy

PJSS-6-3-01-Burawoy 3/15/08 2:18 PM Page 139

based on an imagination of Newtonian physics, with its predictable future

and reversible time, no longer pertained in the natural sciences, which
exhibited striking convergences with cultural studies in a common hostil-
ity to simple explanatory frameworks. Together, natural sciences and cul-
tural studies pointed to a new social scientific epistemology.
The Gulbenkian Commission’s crowning proposal was to unify discipli-
nary knowledge within which the social sciences, now combined into a
single historical science, would be the field of reconciliation of the natural
sciences and the humanities. With all fruitless oppositions thereby
resolved, the social sciences would march forward under the banner of an
unspecified ‘pluralistic universalism’. Paradoxically, this was not a move
beyond, but a programmatic return to the ambitions of 19th century
positivism – the unification of all scientific knowledge. We hear nothing
about how and where this new knowledge will be produced. Nor do we
hear for whom this knowledge will be produced, nor for what ends.
Instead we have an abstract and totalising utopia that reflects the con-
cerns of Western academics, perched high up in the ivory tower, seemingly
unaware that the fortress beneath them – supporting them – was under
siege. We need to transport the Gulbenkian Commission out of its ivory
tower, and bring the Commissioners down from heaven to earth. We need
to start with the actual relations of the material production of knowledge,
recognising how they vary by time and place. To advance the social sci-
ences, I shall argue, we must not dissolve them, but create alliances both
among them and between them and the public, around shared projects –
alliances stitched together from below rather than imposed programmati-
cally from above.

Knowledge for whom? Knowledge for what?

The Gulbenkian Commission suppressed two questions that provide a nec-
essary foundation for re-envisioning the practice and project of the social
sciences in the light of the tasks they face today. The two questions are:
knowledge for whom?; and knowledge for what? In the context of scientific
production we ask, first, whether knowledge is for an academic audience
or an extra-academic audience: that is, whether as social scientists we talk
to one another or to others. We ask, second, whether the knowledge con-
cerns the determination of the appropriate means to pursue a given,
taken-for-granted end, or whether it involves a discussion of those very
ends themselves: that is whether the knowledge is instrumental or
whether it is reflexive.
This gives rise to four types of knowledge that define a scientific field.
Policy knowledge is knowledge in the service of problems defined by clients.
This is, first and foremost, an instrumental relation in which expertise is
rendered in exchange for material or symbolic rewards. It depends upon
pre-existing scientific knowledge. This professional knowledge involves the
expansion of research programmes that are based on certain assumptions,
questions, methodologies and theories that advance through solving

Open the social sciences: To whom and for what? 139

PJSS-6-3-01-Burawoy 3/15/08 2:18 PM Page 140

external anomalies or resolving internal contradictions. It is instrumental

knowledge because puzzle-solving takes for granted the defining parame-
ters of the research programme. Critical knowledge is precisely the exami-
nation of the assumptions, often the value assumptions, of research
programmes, opening them up for discussion and debate within the com-
munity of scholars. This is reflexive knowledge, in that it involves dialogue
about the value relevance of the scientific projects we pursue. Finally,
public knowledge is also reflexive – dialogue between the scientist or scholar
and the public beyond the academy, dialogue around questions of societal
goals but also, as a subsidiary moment, the means for achieving those
goals. The result is the following matrix.

Division of disciplinary knowledge

Academic audience Extra-academic audience
Instrumental knowledge Professional Policy
Reflexive knowledge Critical Public

This matrix forms a division of disciplinary knowledge in which the four

types of knowledge are fundamentally different practices, with different
criteria of truth, modes of legitimation, notions of politics, regimes of
accountability and pathological tendencies. This division defines a scien-
tific field as a pattern of domination and inter-dependence among the four
different types of knowledge. In this view, what distinguishes the natural
sciences from the humanities is the former’s emphasis on instrumental
knowledge that is a concern with the development of scientific research
and its applications and the latter’s focus on reflexive knowledge: that is, a
concern with dialogue about meaning, the fundamental values of society.
The social sciences are not the reconciliation of natural sciences and
humanities, as the Gulbenkian Commission hoped; rather they lie at the
crossroads of these two opposed bodies of knowledge. That is, the social
sciences contain within them the contradictions and challenges of com-
bining instrumental and reflexive knowledge. From this perspective, the
commitment to methodological positivism represents the professional self-
misunderstanding of the nature of social science that sees it as value
neutral and context-free, which reduces the four-fold division of discipli-
nary knowledge to a single quadrant.
We can now turn to the second ill that was emphasised by Gulbenkian
Commission – the changing relation among the social sciences. In terms of
our scheme, the separate social sciences are marked by different configu-
rations and balance among the different types of knowledge. In the United
States, the paradigmatic social science of economics is marked by the dom-
ination of instrumental knowledge while, say, cultural anthropology
weights reflexivity more heavily. Political science is closer to economics,
while sociology is closer to anthropology. More fundamentally, however,
because of the importance of reflexivity, the social sciences should be

140 Michael Burawoy

PJSS-6-3-01-Burawoy 3/15/08 2:18 PM Page 141

distinguished by their configuration of value stances, or what we might

call their standpoint. Economics takes as its standpoint the market and its
expansion, political science takes as its standpoint the state and political
order, while sociology takes the standpoint of civil society and the
resilience of the social. Cultural anthropology and human geography are
potential allies in the defence of civil society. It would, of course, be a
mistake to homogenise disciplines as each is a field of power with subal-
tern groupings that challenge the dominant standpoint of the discipline.
Still, it would be no less an error to overlook the different interests that
divide the disciplines.
At the same time, we must not forget the importance of inter-disciplinary
or trans-disciplinary programmes that, at least in the United States, were
born out of the eruptions of society in the 1960s, and continue to maintain
close relations with their distinctive publics. They are not harbingers of some
new unity of the social sciences or of the social sciences with the humanities,
but, more usually, their appearance and then their persisting marginality
reflect the overweening power of the disciplines. Indeed, the dissolution of
disciplinary boundaries and the unification of the social sciences could only
be real in a totalitarian world in which there are no longer divisions among
state, economy and society. In present-day capitalism, a unification of the
disciplines would be artificial and coercive. It would necessarily reflect
the domination of the market economy and thus be the incorporation of the
social sciences under the hegemony of neo-classical economics.
We have discussed two of the issues identified by the Gulbenkian
Commission, the limits of methodological positivism and the relation among
the social sciences, and it remains only to consider the question of universal-
ism. In criticising the false universalism of European social sciences, the
Gulbenkian Commission created a new and elusive category – pluralistic uni-
versalism. We, however, approach the problem of universalism and pluralism
more concretely – universalistic questions with particularistic answers. Our
two questions, knowledge for whom and knowledge for what, generate four
types of knowledge that provide a general frame for expressing variations in
and inter-connections among local, national and regional divisions of discipli-
nary labour. It enables us not only to specify the differences among disci-
plines, but also the concrete manifestation of disciplines in different historical
times and geographical places. The rest of this article focuses on sociology, but
it applies equally, I would argue, to other disciplines.

The contribution of the semi-periphery: the case of

Portuguese sociology
At one pole of national variation stands US sociology with its elaborate
professionalisation, rooted in an enormously diverse and steeply hierarchi-
cal system of higher education. Professional knowledge did not always
dominate US sociology. Indeed, in its late 19th century origins US sociology,
like so many other sociologies in their inception, was predominantly public
in character, impassioned by social injustice and a champion of moral

Open the social sciences: To whom and for what? 141

PJSS-6-3-01-Burawoy 3/15/08 2:18 PM Page 142

reform. Indeed, in part because it had a more radical and public agenda, in
1905 it broke with the American Economics Association within which it
had developed. As the 20th century unfolded, however, sociology under-
went its own professionalisation, becoming ever more inner directed as it
competed with the other social sciences for a permanent place in the aca-
demic hierarchy. With notable exceptions, such as Edward A. Ross, sociol-
ogists removed themselves from the public eye as they became more
oriented to their peers.
The field of sociology has a different disciplinary configuration in other
countries, reflecting different historical trajectories, patterns of higher edu-
cation and relations among economy, state and civil society. Thus,
Scandinavian sociology possesses a strong policy moment compatible with
the demands of a welfare state. The sociology of some Soviet regimes, such
as Poland and Hungary, were marked by a subterranean critical moment.
Authoritarian regimes, such as those of South Africa and Brazil that fell to a
burgeoning civil society developed a powerful public sociology. Along these
lines the division of labour in Portuguese sociology is especially interesting.
As a late developer, sociology in Portugal shows an especially vibrant
relation among the four types of knowledge. Portuguese sociology began in
earnest towards the end of the Salazar dictatorship and really took off only
after 1974. Entering so late, it could borrow from the traditions of profes-
sional and critical knowledge in other countries, especially from France and
the United States. This was no mechanical adoption, however, but an imag-
inative adaptation to the Portuguese circumstances – circumstances that
called on sociology not only to tackle questions of policy, but also to foster a
societal self-consciousness. With alacrity, sociology took up the challenge to
reconstitute the very social fabric of post-revolutionary Portugal.
Some 30 years after the dictatorship sociology is still very much in the
public eye. Sociologists are regular commentators in the media: newspa-
pers, television and radio. Extended lecture series on sociology have
appeared on public radio. Especially interesting are the open city confer-
ences organised by the Portuguese Sociological Association, which bring
sociologists into dialogue and debate both with one another and with
diverse publics about local and national issues. Sociology’s high profile can
be attributed, at least in part, to the duality of professional sociology. A
sociology degree is not merely a stepping-stone to some other degree but
provides a meaningful identity and distinct occupation in all manner of
organisations: in municipalities, schools, trade unions, media and so forth.
In other words, sociologists are professionals not just in the academy or
research institutes, but in all realms of state and civil society.
Its close association with ‘socialist’ governments has advanced sociol-
ogy’s policy and public roles. Sociologists have entered the political arena as
ministers, parliamentary deputies, trade union leaders and at all levels of
the civil service, while those who remained in the academy became advi-
sors to the leaders of the country from the president down. Entry into the
European Union in 1985 gave rise to a new impetus for policy sociology –

142 Michael Burawoy

PJSS-6-3-01-Burawoy 3/15/08 2:18 PM Page 143

an avalanche of demands for mapping patterns of inequality, poverty, edu-

cation, and for diagnoses of social problems from drugs, to prisons to
mental health. The research is well-financed, but has to be delivered speed-
ily and according to detailed specifications. Still, this policy science then
becomes a potential vehicle for public discussion and the impetus for more
in-depth research. Policy sociology reverberates into and energises all
arenas of sociology.
Underpinning both public and policy sociology is a strong professional
sociology. I have already noted how the Portuguese Sociological Association
represents a certain civic professionalism. It is also particularly robust. It
has 2000 members, which in a society of ten million, represents a density
more than three times that of the United States. Moreover, sociology is
taught in universities and in high schools across the country. There are a
number of dynamic research centres, including those within the universi-
ties of Lisbon, Oporto and Coimbra as well as ISCTE: a founding centre of
Portuguese sociology and a university unto itself.
Institutionally robust, especially for a small semi-peripheral country,
the actual practice of Portuguese sociology has also a distinctive character.
Reflecting and reinforcing the permeable boundaries between sociology
and society is a proclivity towards ethnographic research – research that,
by definition, is at the interface of the academic and the public. Unlike the
majority of participant observation studies in the United States, which
have been steadfastly micro and ahistorical and riveted to the ethno-
graphic present, Portuguese ethnography – whether of urban or rural
areas, whether of family or of work – lays bare micro-processes in order to
gauge the character of the wider Portuguese society and its transforma-
tions. Indeed, ethnographic sites are regularly revisited and restudied to
mark such historical change.
Just as the dividing lines between professional, policy and public sociol-
ogy are quite blurred, similarly we cannot compartmentalise critical soci-
ology. Whether it flows from the French lineages of Touraine and Bourdieu
or from the American lineages of Wallerstein and Wright, critical sociol-
ogy is intimately bound up with professional and public sociology. The rel-
atively recent re-emergence of Portuguese society and the close links
between Portugal and the global south, especially ties to Africa and Latin
America inherited from the colonial era, have given a rare dynamism to
the critical-public nexus, ranging from the emancipatory projects of the
World Social Forum to international feminist projects to Bourdieu-style
critiques of social domination and symbolic violence.
To what can we attribute the multiple and fluid connections among
the four types of sociology? To what extent is Portugal replicating the same
relatively undifferentiated character that can be found in all newly emer-
gent sociologies? To what extent are we seeing the vibrancy of youth, to
what extent the legacy of a peculiar history and to what extent the effects
of a particular place in the world order? How did opposition to the colonial
war and dictatorship create the grounds for a flourishing sociology, whether

Open the social sciences: To whom and for what? 143

PJSS-6-3-01-Burawoy 3/15/08 2:18 PM Page 144

by preparing intellectuals in exile or the formation of a critical intelligentsia

at home? Did those same historical experiences lead to a self-conscious
placement within a global division of sociological labour, connecting criti-
cal voices in both north and south? Will its distinctive connection to
society and as a meeting place of intellectual currents from the world over
be threatened if Portuguese sociology becomes more professionalised and
its public become more cynical? Will the growing importance of policy
science, pressures from European Union for standardisation – the Bologna
process – the hegemonic currency of English draw sociology away from its
local roots and concerns? Can Portuguese sociology manage to maintain
its global profile without at the same time losing its national distinctive-
ness? Indeed, can it develop its specificity through its global connections?
To situate the promise and the challenge of Portuguese sociology, and
indeed other sociologies of the semi-periphery, in an international context
is my final task in this brief commentary.

The spectre of third-wave marketisation

Undoubtedly, Portuguese sociology is a product of its own history and
context that led to the selective appropriation of sociology from elsewhere,
but its late development also expresses something more general – the
potentiality of what I call third-wave sociology.
Sociology has gone through three waves. Its first wave emanated from
Europe. It was a response to the first wave of marketisation that threatened
the existence of the labouring classes, which, in turn, sought to install and
defend labour rights with trade unions, co-operatives, utopian communities
and political parties. This burgeoning civil society of the 19th century
grounded the first wave of sociology: a sociology with strong utopian flavour.
Second-wave sociology had its epicentre in the United States and
stretched from the First World War until the breakdown of the communist
regimes. It corresponded to second-wave marketisation, which began in
the late 19th century, was interrupted, and then burst forth again in the
1920s and 1930s, provoking reactions from nation-states that assumed
the forms of fascism, Stalinism, social democracy and, in the United States,
the New Deal. In each case the state sought to protect society from the
market through the (real or putative) guarantee of social rights. Sociology,
where it was allowed to exist, tried to strike a collaborative relation with
the state. Professional-academic sociology in the United States was given a
boost by policy science, whether the latter served foundations or the
federal government. At a global level this second-wave sociology lasted
symbolically until the last vestiges of planned economies had dissolved,
although in the West the assault on policy sociology began much earlier
with Thatcher and Reagan. From then on states became more inhospitable
to sociology and its project to defend and invigorate civil society. States
instead began to nurture the expansion of the market together with an
offensive against civil society. Economics became the favoured social
science – in some countries more than others.

144 Michael Burawoy

PJSS-6-3-01-Burawoy 3/15/08 2:18 PM Page 145

Sociology has now entered its third wave, a reaction to third-wave mar-
ketisation, more popularly known as neo-liberalism, and more euphemisti-
cally as globalisation. In the present era, defending civil society through
national social policy becomes less viable, and so sociology turns increas-
ingly to the public for its audience, not only on a national scale but also on
a local and global scale. With third-wave marketisation’s assault on
national civil societies, with the retrenchment of labour and social rights,
sociology’s task in its third wave, I argue, lies in the defence of human
rights (which includes labour and social rights) through the organisation
of a civil society of global proportions. This third-wave sociology does not
emanate from the advanced capitalist societies of the north, but from the
countries of the south – latecomers to sociology. Countries that look both
to the south and to the north, countries such as South Africa, Brazil and
Portugal become the fertile ground of a new publicly oriented sociology:
the epicentre of third-wave sociology.
The impetus for a third-wave sociology with its valorisation of public
sociology may spring from such semi-peripheral countries as Portugal, but
it must still operate under the hegemony of the United States and Western
Europe. The sociologies of these countries of advanced capitalism, especially
the United States, command enormous influence, prestige and resources
within the context of global sociology, and thereby shape the possible reali-
sation of public sociology on a world scale. It becomes especially important,
therefore, that alternative models for the division of sociological labour,
such as the one found in contemporary Portugal, gain recognition and
support within the United States for example, where sociologists think their
disciplinary model is the only one, and where those with critical and public
intent are overpowered by professional sociology. Third-wave sociology
must sweep back against the ramparts of second-wave sociology.
We can now restore the Gulbenkian Commission to its historical context
and recognise the source of its myopia. Even though it was written only ten
years ago the Commission’s academic detachment still reflected the period
of second-wave marketisation in which state regulated capitalism protected
the autonomy of universities and their disciplines. But this era has passed
as states are bent on fostering markets – the commodification of research
and the privatisation of higher education – and subjecting the academy to
political surveillance. The confidence in the resilience of academic auton-
omy, taken for granted by the Commission, now looks sadly misplaced as
universities across the globe come under assault from state and market. So
long as the social sciences are differentially implicated in this offensive their
unification becomes more remote and the proposals of the Gulbenkian
Commission more utopian. In an important sense, we are, ironically,
returning to the laissez-faire world of the 19th century and what seemed to
the Commissioners to be an anachronistic past is now a haunting present.
The Gulbenkian Commission’s linear history – social science before
1945, after 1945 converging on a unified historical science – has to be
replaced by a combined and uneven history. By its silence about the very

Open the social sciences: To whom and for what? 145

PJSS-6-3-01-Burawoy 3/15/08 2:18 PM Page 146

different conditions that pertain in different parts of the world, the

Commission assumes that all nations pass through the same phases of
development at the same time. This is obviously far from true. Selective
borrowings (and rejections) of knowledge from advanced countries
combine with indigenous forms and conditions to produce distinctive
national configurations of the division of disciplinary labour – configura-
tions that vary by geo-political region as well as by historical period.
Today these national specificities develop in the context of third-wave
marketisation, a phenomenon that creates divisions not only among coun-
tries but also among disciplines. Thus, economics and political science
have provided ideologies to justify third-wave marketisation although, to
repeat, neither discipline is a homogeneous field, but is internally divided
into dominant and subordinate segments, a division that varies between
countries. Sociology, cultural anthropology and human geography, on the
other hand, have defended civil society against markets and states,
although these disciplines, too, are more or less invaded by economics and,
moreover, mere promotion of civil society can often buttress the power of
state and market. Even if the configuration of the social sciences looks dif-
ferent in different societies, we can still surmise that third-wave marketisa-
tion is more likely to polarise than unify the social sciences.
To conclude, from the stand-point of opposition to third-wave marketi-
sation, there is now real urgency to open the social sciences. That is, to
open them first to reflexive thinking that thematises their relation to the
values and purposes of society, and second to extra-academic audiences, in
particular publics, and especially those publics threatened with the erosion
of autonomy and voice. By virtue of their history and their place in the
modern world system, social scientists of the semi-periphery are pointing
the way forward – not retreating behind the walls of academe, but advanc-
ing into the trenches of civil society. Countries with older and more estab-
lished disciplines would do well to take note of their example.

Wallerstein, I., Juma, C., Keller, E.F., Kocka, J., Lecourt, D., Mudkimbe, V.Y.,
Miushakoji, K., Prigogine, I., Taylor, P.J. and Trouillot, M.-R. (1996), Open the
social sciences: Report of the Gulbenkian Commission on the restructuring of the
social sciences, Stanford, CA: Stanford University Press.

Suggested citation
Burawoy, M. (2007), ‘Open the social sciences: To whom and for what?’, Portuguese
Journal of Social Science 6 (3): 137–46, doi: 10.1386/pjss.6.3.137/1

Contributor details
Michael Burawoy is a sociologist in the Department of Sociology at the University
of California, Berkeley. His interests are in work organisation and working class
consciousness under capitalism and socialism. Contact: Michael Burawoy,
Department of Sociology, University of California, Berkeley. CA 94720, USA.
Tel: +1 510 643 1958. Fax: +1 510 642 0659.
E-mail: burawoy@berkeley.edu

146 Michael Burawoy

PJSS-6-3-02-Pinto 3/15/08 2:19 PM Page 147

Portuguese Journal of Social Science Volume 6 Number 3. © Intellect Ltd 2007.
Review article. English language. doi: 10.1386/pjss.6.3.147/4

‘Open the social sciences: To whom and

for what?’, by Michael Burawoy
José Madureira Pinto Institute of Sociology of the Faculty
of Arts, University of Porto

Abstract Keywords
This text tries to demonstrate that Portuguese sociology has been built on a set of Portuguese sociology
virtuous relations between four poles of sociological activity: the theoretical prob- Portuguese
lematisation pole, the observational research pole, the reflexivity pole and the pro- sociological
fessionalisation pole. It is suggested that this specific dynamic was favoured by a association
series of political-institutional and organisational conditions (the dominance of a applied rationalism
critical/applied rationalism in university training, the active role of the Portuguese public sociology
Sociological Association in the promotion of a creative interaction between acade- social intervention
mics and ‘field professionals’, the political engagement of Portuguese sociologists, sociological training
the relatively successful opening-up of the labour market to professionally trained
sociologists, etc.) The text is, of course, punctuated with comments – largely con-
cordant, but sometimes critical – on Michael Burawoy’s theses about the evolu-
tion and specificity of Portuguese sociology and the need to re-invent public
sociology and reformulate the scientific agenda of the discipline.

Portuguese sociology started to take shape as a project of autonomous dis- * The commentary
status of this article
ciplinary affirmation about 40 years ago. The argument that will be devel- meant that its author
oped here, in close dialogue with the positions taken by Michael Burawoy could not develop
on the same issue, is that if this project had been achieved with a significant more fully some
angles of the
degree of overall success, this was due to a set of circumstances which can evolution of
be analysed by means of the relations indicated in Figure 1 below.1 Portuguese sociology,
which could have
The vertices of the square represent four poles of activity which, interact- better accommodated
ing with one another in the form of a virtuous tension, have in our view stim- the positions
supported here. For a
ulated a qualifying dynamic in the field of Portuguese sociology. They are: more thorough study,
see (Pinto 2007,
Chap. II), which
1. the theoretical problematisation pole (T), representing the set of efforts includes the analysis
which, in the scientific domain in question, seek to encourage theoret- of Michael Burawoy’s
ical updating and discussion in a systematic way; theses on the need to

PJSS 6 (3) 147–154 © Intellect Ltd 2007 147

PJSS-6-3-02-Pinto 3/15/08 2:19 PM Page 148

reformulate the
scientific agenda of 2. the observational research pole (O),
sociology (Burawoy relating to the analysis of concrete
social situations through theoreti-
1 The fact that Michael cally and methodologically informed
Burawoy himself
frequently makes use procedures for gathering and pro-
of this kind of graphic cessing empirical information;
devices in his texts
was an incentive to 3. the reflexivity pole (R), embracing
opt for this solution. critical and self-critical questioning
on positions of principle and founda-
tions of the methodological-theoreti-
Figure 1: Activity dynamics grid. cal options and technical operations
required by sociological work;
4. the professionalisation pole (P), over-determined by the demands of social
intervention in relatively circumscribed ‘practical’ contexts and in
contact with specific ‘lay publics’.

From the standpoint of scientific progress in sociology, we consider that

rather than developing each of these poles independently, it is above all
important to create conditions prone to exploring on a permanent basis
the connections they can establish with one another. Such conditions will
include, as we shall see, both the dissemination of predispositions and intel-
lectual instruments of a certain kind, and a series of political-institutional
and organisational requisites.

The TO and OT vectors represent two fundamental components of scientific
work and together they correspond to what Michael Burawoy regards as
the distinctive outlines of professional sociology (academic sociology), taken
as the place which guarantees the sustainable affirmation of a specific sci-
entific point of view and the institutional consolidation of a discipline.
The TO relation represents the epistemological principle which in
Portuguese sociology has been termed, coherently with its critical perspec-
tive on the empiricist model of knowledge, the command function of
theory in scientific research. Meanwhile, the reciprocal relation (OT
vector) indicates the demand – which is also a genetic mark and persistent
ambition of the ‘scientific spirit’ – to confront interpretative hypotheses
raised by the movement of theoretical problematisation with the results of
observational research of real social situations.
This engagement in controlled systematic observational tasks has been
one of the most important factors in the development of Portuguese soci-
ology, not just because it involves the ‘progressive’ reformulation of theo-
retical frames of reference (countering the ‘normalising’ tendency of
paradigmatic affirmation), but because it is a sort of reserve ready to act
against formalist theoreticism. The fact that research on concrete situa-
tions has been instituted in Portuguese universities as an essential condi-
tion for earning academic degrees and gaining scientific credit has

148 José Madureira Pinto

PJSS-6-3-02-Pinto 3/15/08 2:19 PM Page 149

undoubtedly played a significant role in ensuring that the movement rep-

resented by the vector in question has been in effective operation in
Portuguese sociology for decades.

It is widely agreed by authors engaged in the analysis of the origins of the
institutionalisation of Portuguese sociology that its protagonists partici-
pated and invested strongly in the epistemological debate (always political,
to some extent) which, since the mid-1960s particularly, agitated the field
of the social sciences as a whole. It has equally been noted in what
measure this ‘virtue’ actually arose out of ‘necessity’ (as so often occurs in
social life): in this case, in a country without sociology there was a need
for the group of candidate sociologists (coming from a wide diversity of dis-
ciplinary areas) to promptly and justifiably reconvert their original univer-
sity training.
Having adopted a highly critical perspective in relation to the principles
and procedures of a predominantly empiricist nature (then still very much
ingrained in this field of knowledge) and unreservedly accepting that the
scientific approach of social phenomena always contains a reference to
values and never exempts itself from the effects of partly insurmountable
theoretical-ideological conflicts, the heritage of reflections which was being
consolidated from this time (RT and RO vectors) found fertile ground for
dissemination among apprentices and practitioners of sociology, first at
graduate level and afterwards in postgraduate university courses.
Critical rationalism, as an epistemological model and as a practical
principle for producing knowledge, managed to assert itself as dominant
stream, notwithstanding the influence that ‘post-modernist’ hypercriti-
cism, and, at the other extreme, some positivist manifestations came to
exert in certain sectors of Portuguese sociology. On the other hand, an
(epistemologically non-ingenuous) opening-up to theoretical pluralism
was being imposed on the domestic scientific community, and this was still
anchored in the ‘cultural goodwill’ of the first apprentices.
The rate at which several works on epistemology and methodology,
guided simultaneously by the critique of empiricism and a prudent demar-
cation vis-à-vis hypercriticism, are being republished is in itself a fair indi-
cator of the degree of dissemination of the ‘automatisms of reflexivity’
(even if there is in this expression a contradiction in terms) in the socio-
logical practice of successive generations of Portuguese sociologists. But to
make a deeper assessment of the virtuous effects of these automatisms, it
is worthwhile to bear in mind the agility with which extensive and inten-
sive analytical methodologies intersect in Portuguese sociology – the
former, particularly prone to characterising the structural conditionings of
social practices, and the latter, close to the ethnographic observation pole,
more able to highlight relevant details and singularities.
Accepting this view leads us to believe, in light of the distinction proposed
by Michael Burawoy in the lecture commented on here, that in Portugal

‘Open the social sciences: To whom and for what?’, by Michael Burawoy 149
PJSS-6-3-02-Pinto 3/15/08 2:19 PM Page 150

critical sociology has been somewhat subsumed and embedded in profes-

sional (academic) sociology. It has therefore lost some of its useful meaning
as an autonomous arena for sociological production. We could go even
further, and say that it is precisely because of its mode of acting in practice,
that is to say, as an operator of sociologists’ professional habitus, both in the-
oretical discussion and above all in observational research, that the critical
perspective in sociology has become really effective. The TR and OR vectors
(symmetrical to the previous ones) represent precisely the containment
effect of the ‘abstract’ hypercriticism made possible by (critically embedded)
sociological practice. They also indicate why the agenda of the critique
of sociology is still essentially marked among us by (tacitly critical . . .)
academic sociology.

Another pertinent, and to some extent, original, aspect of Portuguese sociol-
ogy is undoubtedly the connection and reciprocal interaction, which occur
between the academic world and professional practice in (extra-academic)
organisations, in which the aims of social intervention tend to surpass those
of scientific interpretation/validation.
The adjustment between these two worlds is represented in the square
by the OP/PO and TP/PT vectors. It is due, as Michael Burawoy properly
notes, to the importance reached early on in shaping the field of sociology,
by the characterisation and promotion (mostly through the Portuguese
Sociological Association) of a professional culture among sociologists as a
culture that associated ‘science’ and ‘practice’. The popularity of postgrad-
uate training among ‘professionals’ has also operated in favour of the con-
vergence of working interests and environments mentioned earlier.

The reflexivity pole (R), in its essentially methodological and meta-theoretical
components (reflexive knowledge directed overwhelmingly at academic
audiences, that is, critical knowledge according to Burawoy), has always
been responsible for establishing certain criteria to protect scientific work
from coarse bias (here, every adjective is in a sense inadequate, and has to
be considered in relation to what the results of scientific practice indicate as
provisionally acceptable in the corresponding field of knowledge). But there
is nothing to stop it from also playing an active role in the definition –
subject to public scrutiny, and not only to the one of experts and peers – of
the relevant domains (problems) to be appropriately explored by scientific
work (public knowledge).
But if the idea can be ventured here that Portuguese sociology exhibits
some comparative advantages in relation to other national contexts, then
this is because it has, from the very start, ‘naturally’ incorporated into its
normal activity (particularly in mainstream academic sociology) both the
political dimension of reflexivity and a open-minded nearing to the speci-
ficities of social intervention (P).

150 José Madureira Pinto

PJSS-6-3-02-Pinto 3/15/08 2:19 PM Page 151

As already suggested, the professionalisation pole of Portuguese sociol-

ogy is characterised by more than the fact that it has been constituted as
an informed repository of academic knowledge (represented by TP and OP
vectors). Indeed, it has also contributed positively to the reformulation of
the theoretical agenda of the discipline (renewal of the relevant ‘sociologi-
cal problems’) and of the answers to questions on the meaning of sociologi-
cal knowledge – ‘sociology: to whom and for what?’
All the limitations that social intervention professionals have to cope
with are known. These limitations arise largely from the need to find
urgent answers, ‘on the ground’, to extremely complicated situations of
social dysfunction, whose structural causes are ‘remote’ and to some
degree ‘inevitable’. Even so, it must be stressed that, when it concerns
the active involvement of professionals who keep fairly strong links with
the centres of sociological academic production and reflexivity, the profes-
sional work of sociologists can give important contributions to scientific
advancement. Particularly, it can allow the public statement of social
problems bereft of any audible spokesperson, and thereby enable the iden-
tification of innovative lines of theoretical problematisation (PT), the revi-
sion of the accumulated empirical knowledge about societies (PO) and
even the critique of the ‘abstract’ hypercriticism of certain sociological
reflexivity exercises (PR).
But the ‘entry’ of public knowledge into the square of Portuguese sociol-
ogy has not been achieved via the professionalisation pole alone. It has
also come about, as has been mentioned in passing already, from the way
the political dimension of reflexivity was incorporated early on into the
regular activity of producing knowledge aimed basically at peers. Having
made its appearance in an intellectual context in which the wish to
question, sociologically, the social reality was an almost obvious extension
of the wish to put the dictatorship in check, virtually no issue on the
embryonic sociological agenda in the early 1970s escaped some form of
Marxism, as both an analytical tool and as an instrument of system-
atic criticism of the explicit or implicit assumptions of the sociological
frameworks prevailing at the time, was undeniably the most widely dis-
seminated ingredient of the politicised stance – and consequently open
to public discussion on the meaning and ethical-political relevance of
its knowledge content – which Portuguese sociology adopted since its
birth. And the fact that, contrary to what happened in other national
contexts, the presence of Marxism in the university teaching of sociology
remained and spread (albeit indirectly, through other theoretical frame-
works: critical theory, theory of practice, etc.) has ensured that the disci-
pline still retains a measure of analytical-interpretative non-conformism
(in light of the inevitable trends towards paradigmatic standardisation
appearing in the field) in how Portuguese society and the proper posi-
tion of sociology as an instrument of intervention and social change are
thought about.

‘Open the social sciences: To whom and for what?’, by Michael Burawoy 151
PJSS-6-3-02-Pinto 3/15/08 2:19 PM Page 152

Contrary to what the internalist visions of the history of science suggest,
there is no production of scientific knowledge that is led in a social
vacuum – that is, completely immune to the logic of restrictions or incen-
tives of a financial-economic nature, to the influence of ideological
assumptions, albeit implicit, or to the interplay of relatively dissimulated
political interests. In fact, none of the operations in concrete scientific
activity is sheltered from the influence of ‘external’ factors.
Keeping a constant eye on the positions of Michael Burawoy, we have
already pinpointed in this paper a certain number of political, institutional
and organisational conditions that favoured (or at least did not impede) a
sustainable ‘virtuous’ development of sociology in Portugal. Let us turn
them more explicit.
One of those ‘exogenous’ determinants refers to the nature of sociolog-
ical training at graduate level, almost always organised around solid
learning in the spheres of theory, epistemological reflection and the
methodology of observational research. With the replication of such a
demanding model at postgraduate level, it has been possible to reproduce
a set of research procedures and professional routines globally inspired by
an ‘applied rationalism’ adaptable to the specificities and great changeabil-
ity of Portugal’s social reality.
The role of the Portuguese Sociological Association (APS) is another
institutional ingredient which may be taken in account when we deal with
the specificity of the field of Portuguese sociology. A high proportion of
academics and professionals are members of this organisation, which reg-
ularly invites them to discuss the various implications and difficulties of
sociological work at well-attended conferences or seminars.
Another factor favouring the development of Portuguese sociology
concerns the opening-up of the labour market to professionally trained
sociologists: hard at first, but afterwards relatively successful. Contrary to
the somewhat pessimistic forecasts, employability in this area has in fact
remained at acceptable levels from the mid 1980s to the start of the new
century. The factors that helped here were access to European funds
linked to social intervention programmes, plus, later on, the political
option of national and local governments to broaden the spectrum of
measures and policies directed towards the building of a welfare state, at
the time still highly incipient, and, finally, the creation of a demand for
sociological knowledge based on movements and institutions of ‘civil
society’, itself in expansion due to the democratisation process underway
in Portugal.
Another of the forces that Portuguese sociology can rely on is the
consolidation of its research apparatus, at first closely linked to the university
system, but which has subsequently achieved a significant degree of eman-
cipation. Having started early on by seeking spontaneous paths of interna-
tionalisation (initially based on a desire for theoretical updating not
confined to any of the hegemonic centres of international sociological

152 José Madureira Pinto

PJSS-6-3-02-Pinto 3/15/08 2:19 PM Page 153

production), this apparatus is today organically connected to foreign net-

works and research centres ‘of excellence’.
This aspect is more encouraging the better sociology knows how to use
it without losing sight of the requirement to analyse Portuguese society in
all its specificity. Once again, both observational engagement and the
opening up to reflexivity that have been a feature of Portuguese sociology
can interact virtuously – now so as to ensure, as required of sciences that
have to face the historically situated character of their objects, the com-
patibility of analytical instruments of ‘universal reach’ with others capable
of restoring specifically concrete social combinations (that are, to a certain
extent, always unique).

VII: a final and very brief comment

Nothing guarantees that the (‘exogenous’) conditions that have been the
creative force behind the TOPR square will remain stable and keep inter-
vening in Portuguese society.
It is not certain, in the first place, that the graduate and postgraduate
training model for sociologists adopted by Portuguese universities in the
wake of the Bologna Process will ensure such consistent learning as was
achieved in the first decades as the discipline developed. Dominated in
practice by motives of ‘employability’ and ‘mobility’ (terms defined much
more in the register of the stereotype than in that of the sociological reflex-
ivity), this model may be at risk of compromising not only the preparation
of sociologists for the tasks of developing and renewing the paradigmatic
guidelines of the field, as is more obvious, but also the actual fundamental
training for a demanding professionalisation.
If we further agree that, with the generalisation of neo-liberal views
even within the ideological world of social-democracy, the opportunities for
employment and sociologically demanding social intervention in the state
apparatus will decline steadily, then it is foreseeable that the Portuguese
sociology square will become more and more permeable to disqualifying
logics. With greater reason, it is legitimate to expect that the ‘breathing
space’ introduced by the reflexivity pole in this square will be reduced.
In these circumstances, everything suggests that the mediating role of
APS in the qualification of the Portuguese sociological field will find itself
increasingly under threat. And maybe we will find in the future that the
discussions about how to gain and legitimate a certifiable professional
status at European level will prevail over the initiatives which encourage
the virtues of the culture of association between academics and profes-
sional sociologists.
Are we on the path to a new era for Portuguese sociology?

Pinto, J.M. (2007), Indagação científica, aprendizagens escolares, reflexividade social,
Oporto: Afrontamento.
Burawoy, M. (2005), ‘For public sociology’, American Sociological Review 70: 4–28.

‘Open the social sciences: To whom and for what?’, by Michael Burawoy 153
PJSS-6-3-02-Pinto 3/15/08 2:19 PM Page 154

Suggested citation
Pinto, J. (2007), ‘“Open the social sciences: To whom and for what?”, by Michael
Burawoy’, Portuguese Journal of Social Science 6 (3): 147–54, doi: 10.1386/

Contributor details
José Madureira Pinto is a professor at Faculty of Economics (Social Sciences
Department) and research member of the Institute of Sociology of the Faculty of Arts
(University of Porto). He has published several books on the methodology of social sci-
ences, sociological theory and sociological analysis of education, symbolic processes
and cultural practice. He is editor of Cadernos de Ciências Sociais. Contact: Faculdade de
Economia, Rua Dr. Roberto Frias, 4200-464 Porto, Portugal. Tel: +351 225 571 100.
Fax: +351 225 505 050.
E-mail: jmp@fep.up.pt

154 José Madureira Pinto

PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 155

Portuguese Journal of Social Science Volume 6 Number 3. © Intellect Ltd 2007.

Article. English language. doi: 10.1386/pjss.6.3.155/1

‘Fascist lackeys’? Dealing with the

police’s past during Portugal’s transition
to democracy (1974–1980)
Diego Palacios Cerezales UCM

Abstract Keywords
When a dictatorship is overthrown and a transition to democracy begins, the police revolution
force’s place in the new regime becomes a contested issue. Can they be trusted? Are police
they to be held responsible for having enforced the dictatorship’s rules? The April regime change
1974 Carnation Revolution put an end to Europe’s longest right-wing dictator- transition to
ship. The Armed Forces Movement, in order to consolidate its power after the revo- democracy
lution, dismantled the political police (PIDE) and imprisoned its officers. Other military
police forces were ordered to remain in their headquarters and wait for ‘democratic’ social upheaval
reorganisation. During the two revolutionary years that followed, the provisional
governments could not count on the police and did not exercise effective authority:
workers occupied factories, shanty town dwellers occupied empty houses and
angry mobs destroyed the headquarters of political parties. How could the new
authorities deal with the ‘people’s’ disruptive mobilisations if ‘repression’ was the
mark that stigmatised the overthrown ‘fascist’ dictatorship? The post-revolutionary
governments had to devise a new interpretation of the police’s repressive practices,
learning to distinguish which were a mark of ‘fascism’, and which could simply be
understood as the exercise of ordinary public order duties.

When a dictatorship is overthrown and a transition to democracy begins,

the police force’s place in the new regime becomes a contested issue. Can
they be trusted? Are they to be held responsible for having enforced the
dictatorship’s rules? How may their policing practices change in order to
achieve democratic standards of policing? (Kádar 2003). Portugal’s transi-
tion to democracy began with a coup d’état led by the Armed Forces
Movement (MFA, Movimento das Forças Armadas) on 25 April 1974. This
coup, which is also known as the Carnation Revolution, overthrew an
authoritarian regime institutionalised during the 1930s, which, from 1961
on entangled the armed forces in a protracted colonial war. Historians agree
that the police forces, particularly the political police (Polícia Internacional
para a Defesa do Estado – PIDE), were the backbone of the dictatorship.
While the police’s activities comprised a wide field of action – ranging from
crime detection to social services – every police force collaborated with
the feared PIDE and openly expressed their support for the regime’s most

PJSS 6 (3) 155–169 © Intellect Ltd 2007 155

PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 156

contested policies: the colonial war in Africa; moral traditionalism; and

the absence of opposition other than foreign-led communist agitation. No
police units took part in the Carnation Revolution’s military operations:
they sided with the regime and were practically the only force that
attempted to resist. During the ensuing two years, the definition of the role
and status of police forces in newly democratic Portugal became a con-
tested issue: were they criminal ‘fascist lackeys’, or could they be inte-
grated in the state’s bureaucracy as professional police forces?
The MFA programme spoke of democracy, economic development and
decolonisation; yet the MFA was not an ideologically coherent group: rather,
it was a loose coalition of about 400 junior officers, mostly captains, bound
by their formative years in the military academy and the comradeship
ethos of the war experience. Following the coup, 60 of the 62 most senior
officers were placed on the reserve (Pinto 2001); however, the MFA did not
represent the majority of the remaining officer corps, but an organised
minority that had to remain vigilant in order not to be swept away by the
military’s administrative machinery.
By 1974, Portugal’s armed forces had reached its maximum strength,
with over 25,000 officers and 150,000 men. Approximately 5250 of the
officers were professionals, whereas the remainder were conscripts (Carrillho
1988: 440–65). It was during the struggle to consolidate their power in
the aftermath of the coup that the MFA became politicised (Graham
1979). Additionally, General Spínola – the senior officer whom the MFA
had accepted as president in order not to stage too radical a break with the
state hierarchy – became an independent political figure who competed
with the MFA within both the armed forces and society in general.
The ensuing political transition accomplished the programme’s main
points. It certainly brought about democracy and made the independence
of Portugal’s colonies possible; however, the process was not directed.
During two years of transition there were six different provisional govern-
ments, which, in turn, conflicted with new military and political bodies
that also claimed their right to have a say in how the country was to be
ruled: the president, the National Salvation Committee (Junta da Salvação
Nacional – JSN), the Council of State, the MFA’s programme co-ordination
commission, the Continental Operational Command (Comando Operacional
do Continente – COPCON), the Council of the Revolution, the MFA Assembly,
the 5th Division, the democratically elected Constituent Assembly and
some irregular regional popular assemblies. As they all competed for
power, authority fragmented and each military clique organised its own
plot: three of which became actual coup attempts (Manuel 1995; Maxwell
1995; Sánchez Cervelló 1993).
The military was one of the key characters in the play, but as the pho-
tographic record of the Carnation Revolution reminds us, popular mobili-
sation was the other significant player in the political process. On the first
day of the revolution the cheering crowds in Lisbon framed a new form of
legitimacy that was ratified throughout the country by the huge May Day

156 Diego Palacios Cerezales

PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 157

demonstrations that followed; however, this popular mobilisation did not

only express the welcoming of new times, confidence in the MFA’s officers
and the will to experiment with civil and political rights, it also repre-
sented determination to participate in the country’s transformation and
the struggle for improved living conditions. Accordingly, the May Day
demonstrations were followed with the formation in every corner of the
country of popular assemblies and the election of representative commis-
sions in businesses, local councils, shanty towns and schools (Hammond
1988; Downs 1989). The mobilisation affected both the dismantling of the
‘fascist’ regime and the struggle for better wages and living conditions.
Throughout May and June 1974 most ‘fascist’ local and regional
authorities were dismissed by the popular assemblies to be replaced with
provisional administrative commissions or military appointees. Students
denied right-wing teachers, particularly those who were accused of collab-
oration with the PIDE, entry to the universities; whereas in medium and
large enterprises workers organised and held owners and managers
responsible for having based their negotiation power on the repressive
legislation that prohibited strikes and independent unions. The wave of
mobilisation lasted for two years, and became increasingly politicised
during 1975.
The different types of popular mobilisation that took place during 1974
and 1975 were enabled by the lack of authority exercised by the provisional
governments. Lawful procedures for conflict resolution were set aside as
direct action entered into the repertoire of the social and political move-
ments. In the socio-economic arena, industrial, rural and urban collective
movements adopted illegal forms of direct collective action, which included
occupying factories and the estates of large landowners. Thousands of
properties were occupied by groups of organised shanty town residents.
Meanwhile, the popular political struggle consisted of demonstrations,
counter-demonstrations, the boycotting of political meetings, the occupa-
tion of public buildings, road blockades, the destruction of the offices of
political parties, and so forth. Several business owners, town councillors,
and judges, ministers – including the prime minister: twice – were
detained in their offices by demonstrators who were determined not to free
them until their demands had been accepted. There was no police to
repress them.
One of the more dramatic episodes took place during November 1975,
when a large demonstration of workers besieged the Constituent Assembly
for more than 30 hours. The demonstrators refused to disperse and would
not allow food to enter the parliament building until the government
agreed to increase the wages of those employed in the building industry.
The government called for assistance from the police and the military, but
no security force was prepared to use violence to disperse the crowd. As a
result, the government had to submit to the demonstrators’ demands.
The database for 1974 and 1975 contains details of more than 600
episodes of collective law disruption that were not contested by any repressive

‘Fascist lackeys’? Dealing with the police’s past during Portugal’s transition to . . . 157
PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 158

1 Durán Muñoz (2000) public authority. The public order system had collapsed; making it possible
compares the workers’ for any group that was determined enough bring about almost any form of
repertoire of collective
action in the Spanish collective action (Durán Muñoz 2000).1
and Portuguese
transitions, noting
that the Portuguese Repression as stigma
repertoire was much In order to understand the collapse of the public order system we have to
more radical. After
examining some other take into account at least four related topics: the association of police coer-
hypothesis – political cion with ‘fascist’ repression; the competition among political actors in the
culture, organisational
strength – he
post-coup environment; the legitimating role of popular action; and
concludes that the the dilution of discipline within the armed forces. These issues resulted in
crisis of state the political cost of repression being dramatically increased.
authority in Portugal
was the main factor The police had been the dictatorship’s trademark, and its repressive
that explained activities were the stigma that demonstrated how the authoritarian
worker’s recourse to
radical means. regime was unpopular, unjust and based on the coercion rather than the
2 Regulamento de consent of the governed. In Caetano’s Portugal, the primary task of the
Informação da police forces was to maintain the political system and safeguard its opera-
Polícia de Segurança
Pública, despacho do
tion, while all democratic activities were denounced as international com-
ministro do interior, munist subversion.2
15 December 1962. There were three main police forces, all of which were militarised to
These confidential
instructions different degrees: the PIDE,3 the Public Security Police (Polícia de Segurança
determined the Publica – PSP) and the National Republican Guard (Guarda Nacional da
political role of the
PSP and also República – GNR). The PIDE had around 2500 agents, and relied on some
explained that one 20,000 informers (Gallagher 1979), the PSP had 10,500 agents com-
of the communist’s
main goals was to manded by 137 army officers, while the GNR had 9900 officers who
make people lose their patrolled rural areas and garrisoned the main cities with strong infantry
faith in the police.
and cavalry units. In addition to these main forces, there were also two small
3 PIDE changed its specialist forces: the Fiscal Police (Guarda Fiscal – GF) and the Judicial Police
name to the Security
Directorate General (Polícia Judiciária – PJ). The various police forces were complemented with
(DGS, Direcção Geral the regime’s party militia, the Portuguese Legion (Legião Portuguesa – LP),
de Segurança) in
1969, although its which, whilst it had its own security agency and shock troops, dealt mainly
previous name with civil defence matters.
continued to be
widely used.
Following the 1974 coup, Portuguese society experienced feelings of col-
lective liberation (Oliveira 2004). ‘I don’t know what democracy means,
what communism means’, one policeman declared to the press, ‘but every-
thing has changed in the last two days. It’s the first time I feel something like
that. It’s good: the people don’t need to be beaten in order to behave’ (Jornal
de Notícias, 2 May 1974). Nevertheless, the police was in shock and on the
losing side. Its world had turned upside down: the regime they had sworn to
defend had been overthrown and its political elite and senior officials were
being dismissed, while democratic, socialist and communist political activists
who had previously been persecuted were being released from prison,
returning from exile and being honoured as freedom fighters. These former
‘enemies of the state’ even sat in the cabinet and in provisional town coun-
cils. Would the dictatorship’s police ever find a place in this new Portugal?
One of the JSN’s first decisions was to dismantle both the LP and the
once all-powerful PIDE-DGS (Decree-Law 171/74, 25 April 1974). Both

158 Diego Palacios Cerezales

PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 159

these measures were largely symbolic and preventive: the institutions in

question were incompatible with democracy and could be used to co-ordinate
a reactionary counter-coup. This document also decreed that the PSP, the
GNR and the GF be reorganised along ‘democratic lines’. Twenty days later
the PSP’s anti-riot unit was disbanded and its officers distributed among
PSP offices in Lisbon (Portaria 413/74, 15 May 1974). This decision had
long-lasting implications, as it signalled that all – not just political –
repressive actions were under scrutiny.
The anti-riot mobile police unit, which was more commonly known as
the ‘shock police’ (polícia de choque) emulated France’s Republican Security
Company (Compagnie Republicaine de Sécurité – CRS). It was formed by
190 men and trained in the state of the art of riot policing of the 1960s.
The shock police’s equipment included less-lethal weapons such as batons,
dogs, water-cannons and tear gas. Its ruthless action against strikes and
demonstrations – both of which were illegal under the dictatorship – had
become the focus for the opposition’s antipathy, while its operational com-
mander, Captain Maltês, had become an infamous figure. This anti-riot
unit was harsh, hated and had developed a violent sub-culture: the men
commonly beat demonstrators with the metal grip of their batons, usually
causing wounds that bled profusely. Nevertheless, it has to be stated that
the shock police was capable of assessing the degree of force that would be
necessary to disperse crowds without causing fatalities during the cycle of
highly militant mobilisation between 1968 and 1974.
After the revolution, this well-trained and cohesive unit was regarded as
a potentially dangerous counter-revolutionary force, while the popular
view was that it was a ‘fascist police’. However, its demobilisation meant
the provisional authorities could not count on the services of a police force
that was specialised in dealing with crowds and capable of breaking-up a
demonstration or a picket line without injuring demonstrators. The police
and military units that were called to undertake public order control were
neither trained nor had they the specialised equipment. Consequently, their
involvement could quite easily result in politically expensive casualties.

Spínola and police demoralisation: May-September 1974

At first, General Spínola’s men – officers with a professional ethos and a
conservative outlook – were appointed commanders of the PSP and GNR.
The general sought to normalise the political situation. ‘As long as the
police is now liable to the new authorities’, the provisional governments
often declared, ‘the population is bound to obey their commands’. Police
bands took part in the 1974 May Day demonstrations, and policemen
wore carnations in their uniforms to demonstrate their support of the new
political situation. However, neither the reassuring words of the govern-
ment nor these token displays were enough to rescue the police’s reputa-
tion. In the eyes of most Portuguese, the police forces remained the same
‘fascist’ police forces they had been under the dictatorship. In the popular
view the issue was simple: ‘fascism’ had been defeated, policemen were

‘Fascist lackeys’? Dealing with the police’s past during Portugal’s transition to . . . 159
PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 160

4 After the Revolution, ‘fascist lackeys’ and therefore police officers were not to be obeyed. In
previous resistance order to be integrated in the new political system, the police would have
to the ‘fascist regime’
became a legitimating to receive the full support of the new authorities, and words and speeches
factor, with even are not enough when actions suggested otherwise. In fact, neither the gov-
the democratic
Constitution stating ernment nor President Spínola was in real control of the state machinery:
the ‘MFA overthrew the MFA co-ordinating committee maintained its structure and wanted to
the regime as a
culmination of the guarantee the implementation of its programme and prevent the emergence
Portuguese people’s of any counter-revolutionary movements. This represented a significant
resistance to fascism’
(CIHAEP-SP, 1992:
problem for the police: the new civil and military authorities were fighting
291). one another in a battle in which anti-fascist credentials were an essential
5 Interview with asset in order to participate in the new political scene. Consequently,
author, 20 June no-one wanted to taint their symbolic democratic capital by associating
themselves with the remnants of a ‘fascist’ police.4
Although police officers, with the exception of those who served with
PIDE, were neither imprisoned nor dismissed, and the demands for a political
trial were relatively few, the police forces were nevertheless on their own: a
fact they were to learn very quickly. When they first attempted to deal
with picketing strikers or popular commissions that intended to occupy
empty houses, they discovered the public’s usual obedience to police
orders had transformed into open resistance. This was not unexpected;
however, if they attempted to use force to overcome resistance, their supe-
riors in the civil authorities and the MFA would denounce them as ‘fascist
brutes’ and demand they be held responsible. The traditional cover-up or
‘grey check’ the police used to perform their duties no longer functioned,
meaning that they learned it was wiser and safer to remain passive.
Two months after the revolution, Luis Filipe Madeira, who was a young
lawyer and a well-known democratic activist who became civil governor
for the Algarve, discovered that the PSP agents in the region had all become
demoralised. The public no longer obeyed police instructions, made fun of
them and denounced them as ‘fascists’ whenever they became involved in
any civil dispute. In these conditions, police officers simply refused to
patrol the streets. In the words of Madeira, ‘they were not policemen, but
empty uniforms’.5 There are similar accounts of police demoralisation all
over Portugal. On the other hand, the appearance of the army at a trouble
spot, where they were quite often sent to rescue police officers, was gener-
ally met with cheers. The military’s involvement in social conflicts was
usually welcomed – at least until the summer of 1975, when political
polarisation began to affect even its popularity.
President Spínola and the MFA first clashed on the issue of decolonisa-
tion, and the police were part of this power struggle. Spínola wanted to put
an end to the social turmoil and ensure the rapid reconstruction of law
and order. He counted on the support of the police. With the intention of
reinstating police authority, Spínola had to display his confidence in the
police. In early June 1974 he ordered that the police be supplied with mil-
itary issue automatic rifles; however, the chief of staff of the armed forces,
General Costa Gomes, sided with the MFA and blocked Spínola’s decision.

160 Diego Palacios Cerezales

PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 161

Costa Gomes argued that ‘the GNR and the PSP are not mentally fit to 6 Spinola’s own project
participate in the revolution’, insinuating that they could be used in a for Africa was a kind
of confederation.
counter-coup: he suggested that the PSP be disarmed and provided with He was opposed to
wooden truncheons instead (Spínola 1978: 149–50). the recognition of
the armed liberation
Spínola began a campaign to rally conservative social forces and mili- movements as the
tary officers. The MFA feared that if Spinola managed to rebuild the state’s former colonies’
representatives. His
authority, he would be able to impose his own programme for a presiden- plan to organise a
tial constitution, no economic transformation and a long and controlled political consultation
in the colonies needed
process of decolonisation along federal lines that would result in the army the backing of an
remaining in Africa for many years to come.6 In July 1974, Costa Gomes armed force capable
and the MFA outflanked Spínola on the matter of maintaining public order of controlling the
situation, but after
when they obtained his support for the creation of a special operations the coup, soldiers
commission, COPCON, which was to be responsible for public order. The refused to combat.
armed forces thus became the principal force for public order. The police 7 Mainland Portugal
was divided in four
was effectively cast aside – insulted even – because the decree mentioned military regions:
their ‘inability’ and ‘inconvenience’ in respect of maintaining public order: North (based in
Oporto), Centre
it was clear that the police did not enjoy the confidence of the political (Coimbra), South
leadership (Decree-Law 310/74, 8 July 1974). From then on military units, (Évora) and Lisbon.
such as the Military Police (Polícia Militar – PM), Lisbon’s artillery regi-
ment, the naval infantry and Queluz operational infantry became the
forces of public order in the capital, while regional military commanders
were attributed with the responsibility to maintain public order in their
zones.7 COPCON also controlled Portugal’s elite commandos, paratroops
and marines: forces that would be critical in putting down any attempted
counter-coup. All this power became tangible when, in late September
1974, Spínola attempted to strengthen his position by organising the
demonstration of the ‘silent majority’, in emulation of De Gaulle’s moves
following the events in Paris in May 1968. Order-loving Portuguese from
throughout the country were called to a rally in Lisbon where they could
show their support for Spínola’s policies, their rejection of the strikes and
social disorder and to show that the communists and the socialists who
were monopolising the public arena were not representative of the
Portuguese people.
This call worried the left-wing groups, particularly since there were
rumours that Spinola was seeking to ban the Portuguese Communist
Party (Partido Comunista Portuguesa – PCP) and that he would use the
meeting to launch a counter-coup. In response, the popular commissions
and trade unions mobilised against this demonstration by barricading the
entrances to Lisbon. Cars were stopped and searched for weapons, while
people on their way to attend the demonstration were turned back. In
order to counter these illegal popular blockades, Spinola asked COPCON to
clear the roads and guarantee the rights of the demonstrators; however,
COPCON took the side of the ‘people’, reinforced the barricades and over-
whelmed those military and police forces that had followed Spinola’s
instructions. The general was forced to resign, his supporters were dis-
missed and General Costa Gomes was appointed president.

‘Fascist lackeys’? Dealing with the police’s past during Portugal’s transition to . . . 161
PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 162

The police forces on stand-by: September 1974–March 1975

Spínola’s defeat was also a defeat for the police. With some PSP and GNR
units having supported Spínola during the events of September, angry
mobs sacked many police stations. Both the GNR and PSP were made
responsible to the armed forces chief of staff rather than to the ministry of
internal affairs. They were to be reorganised.
Throughout the following months, the police maintained a low profile
presence in social conflict, while COPCON was in charge of public order.
The military met their new police role imaginatively, and tended to fraternise
with the ‘popular masses’. Workers, the rural proletariat and shanty town
dwellers cheered military forces as they approached. Left-wing officers and
non-commissioned officers had full authority to deal with popular strug-
gles, and began co-operating with ‘the people’. When workers occupied a
factory, they asked the armed forces to conduct an inventory and mediate
with the owners and the political authorities; when rural workers began
occupying the large estates, they could count on the military to supply
trucks for transportation. One MP official recalled, ‘we were told to go
to troubled spots, such as factories that were being taken over by the
workers, but when we arrived we asked the workers why were they doing
what they were doing, and most often they were right and we let them do
as they wanted’ (Domingos, Gago and Matos 1977).
On 11 March 1975 forces loyal to the deposed President Spínola
made a counter-coup attempt. The conspirators retained MFA-appointed
police commanders and managed to mobilise some GNR and PSP units.
Nevertheless, the MFA managed to put the attempt down, while self-styled
anti-fascist crowds were mobilised in many towns and, in several locations,
demonstrated in front of PSP and GNR stations to prevent the officers
within the buildings from taking part in the counter-coup. Once it was
clear that the attempt had failed, police officers were so afraid of being
attacked by the crowd that they had to be rescued in armoured vehicles by
the military.

A police force for the revolution: March–July 1975

This failed Spínolist coup gave new strength to the MFA radicals, who
passed their favoured policies and assured the MFA’s institutionalisation in
the future constitutional order: banks and major industrial firms were
nationalised, the MFA Assembly appointed a Council of the Revolution
with constitutional prerogatives and the new provisional government
announced that Portugal had begun a transition to socialism. At last, a
major police reform was announced: Pinto Ferreira, a left-wing colonel who
had been appointed GNR commander in February, was also appointed
commander of the PSP. In order to create a new democratic police, both
forces were going to be merged in six months time.
What could it be: a democratic police? Or, as it was sometimes asked:
what was the role of a police force in the construction of socialism? The
re-organisation of the police was not an easy task. If the new police was to

162 Diego Palacios Cerezales

PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 163

be democratic, it had to change its internal organisation, its culture and the 8 The few military units
way it dealt with the public. This re-definition attempt was also taking place that were politicised
contrary to the
in the armed forces. As we have already noted, the MFA was neither a prevailing ideology
coherent nor a disciplined body, but a coalition of junior officers. It had to of the surrounding
social area – such
undermine the traditional hierarchies and purge ‘counter-revolutionary’ as Vila Real Infantry
officers in order to maintain their grip on the armed forces, so they created Regiment (left-wing in
a conservative area)
co-ordinating structures that bypassed formal command chains and several or the commandos
participatory mechanisms inside the units which diluted the discipline. The in the Lisbon area
(right-wing in the
MFA fostered internal assemblies in every military unit, in which officers, ‘Lisbon Commune’)
non-commissioned officers and men were supposed to discuss the unit’s life had a special part to
and promote the political awareness of the military. Young, educated, leftist play in the following
events (Palacios
officers used to dominate the assemblies, in particular when the unit had Cerezales 2003:
its barracks in urban, industrial areas and in areas of large agricultural 167–9, 182–9).
estates, while in northern rural Portugal the conservative outlook of the 9 The PPD later
changed its name to
surrounding society limited the leftist appeal.8 A similar participatory Social Democratic
process was promoted within the police: non-hierarchical assemblies were Party (PSD – Partido
Democrático Social).
held in most districts, with each nominating delegates to a national assem-
bly at which re-organisation was to be discussed. The first police assembly
at a national level took place 11 June 1975. The order of the day shows
that they discussed political purges of fascist police officers, the principles of
the fusion with the GNR and the way in which there could be built a so-
called ‘alliance of the security forces with the people’. As long as the issues
at stake were the destruction of fascism, popular mobilisation and the con-
struction of a new society, those in command were no longer to exercise
their functions by virtue of the authority ascribed to them from above, but
by the consent of those serving under them and the will expressed by the
so-called ‘popular masses’. A far-right critic would latter recall these experi-
ences as a process of terrorising police officers, undermining hierarchy and
handing over the Police to the PCP (Barreto 1978). A second gathering was
scheduled for August, but the political process was changing at a very fast
pace and it never took place.

The hot summer of 1975

In April 1975 Portugal held its first democratic elections. Participation
was high (91 per cent), with the Socialist Party (Partido Socialista – PS)
winning overall while the centre Popular Democratic Party (Partido Popular
Democrático – PPD) and the centre-right Social Democratic Centre (Centro
Democrático Social – CDS) parties obtained the majority in the north
of the country.9 Radical parties like the PCP and the Popular Democratic
Movement (Movimento Democrático Popular – MDP) showed their strength
amongst the industrial and rural proletariat, although they only managed
to obtain 15 per cent of the national vote.
Although the election was for the constituent assembly that was to
prepare and approve a new constitution, the election result changed the
political balance. While Mário Soares’ PS discovered its massive political
appeal, it struggled to convert its electoral success into government power.

‘Fascist lackeys’? Dealing with the police’s past during Portugal’s transition to . . . 163
PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 164

Vasco Gonçalves and the MFA radicals who were in office resisted, with the
PCP supporting their position: they claimed the MFA embodied a ‘revolu-
tionary legitimacy’ that was higher than the political parties’ ‘electoral
legitimacy’. Some military officials even went so far as to claim the MFA
would have won the election had it had presented its own candidates,
and dismissed the high turnout, claiming that since the rural poor and
illiterate masses of the north did not yet understand the principles of the
revolution, their votes were not as meaningful as the votes of the class-
conscious industrial areas. The north may have voted for centre-right
parties, but that was only the dead weight of 40 years of obscurantist dic-
tatorship (Correia, Soldado and Marujo undated).
This clash of legitimacies – electoral and revolutionary – broke the
broad political coalition that had supported the MFA. Far-left organisations
and some members of the armed forces accused the communist-leaning
government of trying to control the popular struggles, so they began to
fight for popular power and to organise autonomous workers’ and soldiers’
assemblies. The PS also began to organise large demonstrations against the
government, and called for the primacy of electoral legitimacy. The centre-
right parties followed this strategy, taking part in PS initiatives. Throughout
this ‘hot summer’ of 1975, the hitherto silent conservative majority in the
north began to mobilise under the leadership of the Catholic Church, which
organised meetings, pilgrimages and demonstrations. Anti-communism
became the rallying cry for a new broad political coalition of conserva-
tives, democratic socialists and former Salazarists, while angry crowds
sacked the offices of left-wing local authorities throughout the region.
During two months of violence, approximately 80 of the PCP’s local offices
were destroyed (Palacios Cerezales 2003: 141–73). This radicalisation was
on the increase, and the government and the MFA began to lose control of
the northern districts. Some hard-line MFA officials pressed for ‘strong
repressive action’ in order to save the revolution from ‘reactionary’ and
‘terrorist’ forces. Nevertheless, most of the military did not want to resort
to shootings or to be associated with violent repression.
The maintenance of public order was a hot issue, and some regional mil-
itary commanders tried to transfer those duties to the police forces; however,
as one GNR commander explained, he could not ask his men to stop the
angry crowds, because force could be necessary and if a guard killed
someone in the line of duty, the political authorities would charge him with
murder and label him a fascist (Comércio do Porto, 23 September 1975).
COPCON purchased anti-riot equipment, including tear-gas, rubber
bullets, protection shields and water tanks, but did not have sufficient time
to train its troops how to use it. It also sent some hard-line marine units to
the most violently anti-communist districts, but after killing two demon-
strators who attacked a PCP headquarter in Fafe, near Braga, they were
asked to withdraw. Most of the MFA’s men were unwilling to assume the
costs of repression since, as General Costa Gomes claimed, ‘[it] did not
have a repressive vocation’.

164 Diego Palacios Cerezales

PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 165

At this point the MFA split into three factions: radicals, who favoured a
military-controlled transition to socialism; populists, who supported the
autonomous popular struggles; and moderates, who called for an agree-
ment with the PS while manoeuvring to form an alliance with conserva-
tive officers and take control of the military apparatus. As most military
units demonstrated their lack of repressive will against anti-communist
crowds in the north, the radicals discovered they were unable to govern.
The 5th provisional government, which had been the most radical to date,
was forced to resign in favour of the 6th and final provisional government
that counted on support from the moderate parties.

Moderate government against the ‘Lisbon Commune’:

September–November 1975
The new government was backed by the majority of the Constituent
Assembly, and was received with the abrupt halt of popular anti-commu-
nist violence. However, the social turmoil continued: left-wing soldiers
who disagreed with the moderate turn began organising demonstrations
of enlisted men, declaring that ‘soldiers would always be on the side of the
people’. Indiscipline was widespread, and the troops refused to obey orders
when they were asked to stop the illegal collective actions being carried
out by workers and landless peasants. The government and the MFA’s
moderates decided that in order to re-establish authority they had to rely
on the police. The unification of the GNR and PSP was halted, and a new
commander was nominated for each. The government took powers away
from COPCON and created the Military Intervention Group (Agrupamento
Militar de Intervenção – AMI), a co-ordinating structure in which the few
disciplined military units and police were integrated. Both the PSP and
GNR were provided with heavy weapons, such as the G3 automatic rifle, a
decision that both dramatised the new political confidence in the police
and had powerful symbolism. The infamous anti-riot shock Mobile Police
unit was also reorganised.
The reconstruction of a disciplined and respected public order system
was at stake, and the new authorities enacted their will to re-commit
themselves to the police. Nevertheless, the political situation remained
uncertain. In Oporto and other northern cities the police quickly realised
they could obtain the confidence of the public authorities, and began once
more to use coercion when they believed it to be necessary. In the ‘Lisbon
Commune’ and other southern districts, the balance of power was not yet
in favour of the government: some leftist military units remained active, the
marines were as powerful as ever and Major Otelo Saraiva de Carvalho,
the darling of the left, was still Lisbon’s military governor.
Lisbon became a theatre for enormous daily demonstrations and counter-
demonstrations and rumours of imminent coups were commonplace. On
20 November, following a 30-hour siege of the Constituent Assembly, the
government declared itself to be on strike and announced it was going to
move to Oporto until such times as the armed forces had re-established the

‘Fascist lackeys’? Dealing with the police’s past during Portugal’s transition to . . . 165
PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 166

conditions necessary for effective government. The moderate faction of the

MFA had everything prepared in their plan to outwit the radicals: Otelo
was dismissed as the capital’s military commander; troops serving in the
more radical units were placed on leave; and the veteran commandos were
re-enlisted and called to active duty. Finally, on 25 November 1975 an
illegal move by paratroopers was interpreted as an attempted coup by the
far-left, enabling the pro-government military to seize the advantage and
mobilise the commandos to take control of Lisbon.
The country had been on the verge of civil war for four months, and
this could have been its first episode; however, the radical officers refused
to mobilise their forces, which mainly consisted of marines, and accepted
democratic legitimacy must prevail.

The police come-back: the end of the revolution

The government returned to Lisbon and resumed office under its new strong
man, Colonel Eanes, who had been commander of operations that took place
on 25 November. Eanes was appointed army chief of staff and oversaw the
normalisation of the hierarchy within the armed forces. Left-wing units
were disbanded and the assemblies and soldiers’ participation in them
were suppressed. At least 100 left-wing officers were imprisoned and left-
wing publications were closed. This was not a Chile-style coup; however,
these measures only lasted a few months – just enough time to re-establish
the authority of the state. The popular movements soon discovered the
political situation had changed and radical collective action quickly disap-
peared, allowing Portugal to once more become a demobilised and under-
politicised society.
The political experience of these months of social unrest had made clear
that democratic legitimacy also meant the primacy of public authority by
whatever means necessary, and supported by whatever civil or military
enforcement required. As long as the association of coercion with fascism
had reduced the state’s capacity to govern, the democratic need for
authority had to be stressed. The governing politicians and MFA moder-
ates had learned a lesson: the consent of the governed was not enough to
govern, because a complex society comprises competing interests and
political wills. They also learned they had to accept some degree of police
violence, and, at least in principle, to publicly back police actions.
In a very significant episode, on 1 January 1976, the GNR shot over
the heads of people taking part in a far-left demonstration in Oporto, and
managed to kill four people in the process. Both the journalists at the
scene and the subsequent Russell Commission report stated the guards
had lost their temper; yet both the government and the Council of the
Revolution sided with the GNR and held the demonstrators responsible
for the violence (Diário de Lisboa, 2–4 January 1976). The restoration of
political confidence in the police was thereby explicitly stated, encouraging
the public to once more respect the indications and instructions of police

166 Diego Palacios Cerezales

PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 167

The police resumed its usual functions: former Spínolists were nominated
as police commanders, and the anti-riot equipment purchased by COPCON
was transferred to the police. In March 1976 the new anti-riot police, the
Intervention Corps (Corpo de Intervenção – CI) was publicly announced,
although its existence was yet to be approved by law. Most of the officers
who were employed in this corps had previously been officials of the
authoritarian regime’s anti-riot Mobile Police. The commando regiment,
which had been in charge of public order in Lisbon since the 25 November
coup, was allowed to return to barracks: the return of the police thus may
well be the moment symbolising the end of the revolution.

During the years that followed, Portugal’s democratic regime became
increasingly stable. The new lines of police modernisation in Portugal
would take Western European standards as its benchmark. During the late
1970s, the PSP was demilitarised and a special school for police officers
was established. The GNR evolved more slowly, and continued to use
firearms against collective rural collective throughout the period during
which the agrarian reform was dismantled. Nevertheless, in the early
1980s they were provided with less-lethal weapons, and in 1986 the
GNR’s special anti-riot unit was created.
After the political transition, the police forces had to deal with a double
legacy: that of the dictatorship and that of the revolution (Pinto 2001).
In the public’s reconstructed recollection, only the PIDE remained associ-
ated with the fascist repression: the PSP and GNR were integrated as pro-
fessional police forces that were responsible to the law rather than to any
particular political regime. Moreover, in the sub-culture of the police, the
memory of the revolutionary period’s social turmoil became a kind of alibi
for their previous devotion to the authoritarian regime.
Thirty years on, the generational replacement and dramatic increase
in police numbers have diluted the presence of old-school police officers.
Those police officers who had willingly participated in the revolutionary
reorganisations of 1975 were cast aside, but there remained a desire to
change some internal hierarchical and professional problems that expressed
itself in the long-standing struggle for union rights within the police
(Colaço and Gomes 2001). The aspiration of police officers for civic rights
is often expressed as the need to ‘bring the revolutionary principles of
25 April’ into the police forces.

Barreto, M. (1978), História da polícia em Portugal: Polícia e sociedade, Braga: Braga
Carrilho, M. (1985), Forças armadas e mudança política em Portugal no século XX: Para
uma explicação sociológica do papel dos militares, Lisbon: INCM.
Colaço, A. Bernardo and Gomes, A.C. (2001), Sindicalismo na PSP: Medos e fantas-
mas em regime democrático, Lisboa: Cosmos.

‘Fascist lackeys’? Dealing with the police’s past during Portugal’s transition to . . . 167
PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 168

Comércio do Porto, 23 September 1975.

Commissão Internacional para a História da Assembleia de Estados e dos
Parlamentos—Secção Portugusa (CIHAEP-SP) (1992), Constituições por-
tuguesas, Lisbon: Assembleia da República.
Correia, R., Soldado, P. and Marujo, J. (forthcoming), MFA e luta de classes: Subsídios
para a compreensão do processo histórico português, L: Ulmeiro.
Decree-Law 171/74, 25 April 1974.
Decree-Law 310/74, 8 July 1974.
Díario de Lisboa, 2–4 January 1976.
Domingos, H., Gago, J.S. and Matos, L.S. de (1977), A revolução num regimento: A
polícia militar em 1975, Lisboa: Armazém das Letras.
Downs, Ch. (1989), Revolution at the grassroots: Community organisations in the
Portuguese revolution, New York, NY: State University of New York Press.
Durán Muñoz, R. (2000), Contención y trasgresión: Las movilizaciones sociales y el
estado en las transiciones española y portuguesa, Madrid: Centro de Estudios
Políticos y Constitucionales.
Gallagher, T. (1979), ‘Controlled repression in Salazar’s Portugal’, Journal of Con-
temporary History 14: 385–402.
Graham, L.S. (1979), ‘The military in politics: The politicisation of the Armed
Forces Movement’, in Graham L.S. and Makler H.M. (eds), Contemporary Portu-
gal: The revolution and its antecedents, Austin, TX: University of Texas Press.
Hammond, J.L. (1988), Building popular power: Workers and neighbourhood move-
ments in the Portuguese revolution, New York, NY: Monthly Review Press.
Jornal de Notícias, 2 May 1974.
Kádar, A. (2001), Police in transition, Budapest: Central European University Press.
Manuel, P.C. (1995), Uncertain outcome: The politics of Portugal's transition to democracy,
Lanham, MD: University Press of America.
Maxwell, K. (1995), The making of Portuguese democracy, Cambridge: Cambridge
University Press.
Oliveira, L.T. de (2004), Estudantes e povo na revolução: O serviço cívico estudantil
(1974–1977), Oeiras: Celta.
Palacios Cerezales, D. (2003), O poder caiu na rua: Crise de estado e acções colectivas na
revolução portuguesa, Lisbon: Imprensa de Ciências Sociais.
Pinto, A.C. (2001), ‘Settling accounts with the past in a troubled transition to
democracy: The Portuguese case’, in Brito, A.B. de, C. González Enríquez, and
P. Aguilar (eds), The politics of memory and democratisation, Oxford: Oxford
University Press.
Portaria 413/74, 15 May 1974.
Regulamento de Informação da Polícia de Segurança Pública (1962), Despacho do
Ministro do Interior, 15 December 1962.
Sánchez Cervelló, J. (1993), A revolução portuguesa e a sua influência na transição
espanhola 1961–1976, Lisbon: Assírio e Alvim.
Spínola, A. (1978), País sem rumo: Contributo para a história de uma revolução,
Mirandela: SIRCE.

Suggested citation
Cerezales, D. (2007), ‘“Fascist lackeys”? Dealing with the police’s past during
Portugal’s transition to democracy (1974–1980)’, Portuguese Journal of Social
Science 6 (3): 155–69, doi: 10.1386/pjss.6.3.155/1

168 Diego Palacios Cerezales

PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 169

Contributor details
Diego Palacios Cerezales is an assistant professor and doctoral researcher at
Madrid’s Complutense University (UCM). He obtained his masters degree in social
sciences at the University of Lisbon’s Institute of Social Sciences (ICS/UL). He has
published O poder caiu na rua: Crise de estado e acções colectivas na revolução portuguesa
(Lisbon: Imprensa de Ciências Sociais, 2003), and several articles on contempo-
rary Portuguese history. He is currently completing his doctorate on the history of
public order policing in Contemporary Portugal (1834–2000). More information
can be obtained at http://www.historiadelpensamiento.es/dpc.html. Contact: Diego
Palacios Cerezales, Departamento de Historia del Pensamiento y de los Movimientos
Sociales y Politicos, Facultad de Ciencias Políticos y Sociología de la Universidad
Complutense de Madrid, Campus de Somosaguas, 28223 Madrid, Spain. Tel:
+34 91 394 27 83. Fax: +34 91 394 28 57.
E-mail: dpalacio@cps.ucm.es

‘Fascist lackeys’? Dealing with the police’s past during Portugal’s transition to . . . 169
PJSS-6-3-03-Cerezales 3/14/08 2:34 AM Page 170
PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 171

Portuguese Journal of Social Science Volume 6 Number 3. © Intellect Ltd 2007.

Article. English language. doi: 10.1386/pjss.6.3.171/1

From Porto to Portadown: Portuguese

workers in Northern Ireland’s
labour market
Martin Eaton University of Ulster

Abstract Keywords
While North-Western Europe remains the principal destination for Portuguese Portuguese
emigrants, post-millennium flow has seen the United Kingdom (UK) and North- migrant worker
ern Ireland (NI), in particular, emerging as a focal point. As part of a changing employment agent
labour market demand and supply process, several thousand migrants have now Northern Ireland
been recruited by agencies to work in the region’s rurally based food processing rural labour market
industries. This article quantifies the resurgence of Portuguese emigration trails,
explores their recent distribution patterns, and evaluates the role of employment
intermediaries in facilitating the flow. Using qualitative discursive techniques the
experiences of these players are examined before determining their impacts on the
local labour market. Results show that benefits have been brought to a number of
localised economies suffering from shortages and working patterns based on sub-
stitution and segmentation have been fundamentally altered. At the same time,
some small towns have struggled to adapt to this influx and concerns have been
raised in relation to work-based problems and the pace of developmental change
associated with the growing numbers of Portuguese emigrants in Northern

In summer 2006 as the Portuguese soccer team enjoyed success at the
World Cup, residents of the small provincial town of Portadown in
Northern Ireland gathered to cheer them on. In a display of inter-community
support, both locals and immigrant workers came together in their
support for a team that carried the hopes of two, small, semi-peripheral,
part industrialised countries located on the fringes of the European Union
(EU). It was a union, in part, inspired by their mutual rivalry with the
England football team and demonstrated some of the progress that has
been made in integrating Portuguese workers into Northern Irish society.
The Portuguese in this part of Ulster represent a small proportion of an
emigrant community numbering at least 4.5 million worldwide (Lawless
2005). This mobilisation was based upon exploration, colonisation, and
more recently, economic emigration to seek a better life. Migrant flows
have curtailed since the peak period of the 1970s when hundreds of

PJSS 6 (3) 171–191 © Intellect Ltd 2007 171

PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 172

thousands fled from authoritarian Portugal. Nevertheless, there has been

a post-millennium, domestic recession-induced resurgence in emigration
and it continues to be a fundamental factor shaping Portugal’s socio-
demographic evolution (Arroteia 2001). In turn, the United Kingdom has
emerged as an important destination with immigrants now estimated to
number between 110,000 and 250,000 (Anon 2005; Almeida 2006).
While their nuclei focused upon London and the Channel Islands, there
have been significant influxes of emigrants into Britain’s peripheral, semi-
rural regions. These secondary flows included East Anglia, north-west
England, Wales, the Scottish Borders and Northern Ireland.
With this background in mind, this article presents an overview of the
Portuguese emigrant flows in north-western Europe and more specifically
towards the United Kingdom. It attempts to quantify recent influx, deter-
mines their location patterns and the reasons for recent re-distributions of
emigrantes towards peripheral parts of Britain. The study reviews sec-
ondary literature and statistical data before utilising empirical research to
focus upon the working experiences of Portuguese migrants and the views
of interested third parties, related local employers and emergent support
organisations. Our spatial emphasis is the Northern Ireland labour market
where many migrants have been recruited by employment agencies to
work in the region’s agricultural harvesting, meat processing and food
packing sectors. These industries, in turn, are located in small rural towns
often distant from the region’s main population centres. While the
numbers of foreign minorities in Northern Ireland are relatively small,
many Portuguese (along with recent influxes of eastern European
workers) have congregated together in expressions of human gregarious-
ness and shared economic interests. These concentrations have brought
benefits to local economies but have also lead to problems at the micro-
scale. As a result, issues relating to the workers’ motivations, and includ-
ing the phenomena of ‘trade-off ’, competition and discrimination are
discussed, together with a determination of what the future might hold for
these itinerant workers.

Migrant workers, labour markets and population mobility

The majority of contemporary Portuguese migrants have been labelled as
neo-classical labourers perpetually moving in order to find jobs, secure
salaries and remit their savings back to their families (Castles 2000). This
pattern has, however, become more complicated since the emigrants have
been further motivated to travel and find a better level of remuneration
and more secure conditions of employment in their chosen destination
countries. In relative terms, therefore, workers availed of more advanta-
geous labour market conditions to the ones they may have been used to in
their country of origin or previous destination society and this is now a
key factor influencing their decision-making, and in turn, their exodus.
Although this mobility was normally an individual decision-making
process, in more recent years, Portuguese emigrants have become part of

172 Martin Eaton

PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 173

a collective labour market process. Employment agencies have been set up

that were designed to engage workers for the benefit of specific sectors of
industry and these have proliferated with recruitment branches being
established in both host and destination countries. The agents have
actively sought out candidates and then moved migrants from either lower
cost/lower wage areas (such as Portugal) or from employment in similar
sectors of industry (in north-west Europe) to semi-peripheral areas of the
European Union (such as Northern Ireland). In so doing, they have
induced more complex local labour market issues (Mulholland 2005),
which are, in part, based upon the perception of migrant labourers replac-
ing the indigenous workers. This idea that the immigrants ‘are taking
(our) jobs (which should be) for local workers’ (the ‘TOJ’ syndrome), is a
prejudiced but very real concern (Hayes and Dowds 2006) in many
advanced industrialised countries where immigration is often near the top
of the political agenda (Borjas 1999; Spencer 2003). As a result, the
notion of strain on social and economic integration into local job markets
has emerged as a contentious issue for state governance, public organisa-
tions and private agencies. In an Irish context this issue was further com-
plicated by a recent history of conflict within its divided communities and
spatial boundaries, making the process of Portuguese immigration unique,
and integration into the local labour market, an outwardly difficult
prospect fraught with hazard (Borooah and Mangan 2007).
These theoretical conditions meant that Northern Ireland represented
a microcosm of what was happening in other, similar, rural, labour migra-
tion-dependent and peripheral parts of the United Kingdom. Given that
little research has been carried out into these job markets so this contribu-
tion becomes an important if tentative starting point. It is an issue deserv-
ing of attention from politicians, economists, demographers, geographers
and social commentators interested in determining the impacts of rela-
tively large numbers of often averagely educated, transient, single young
people seeking work and financial remuneration at a level that would be
impossible to achieve in an equivalent type of job in Portugal. As such, the
theoretical and practical implications for both host and destination soci-
eties should not be underestimated.
The diáspora associated with Portugal’s population has been an endur-
ing feature of the country’s social, cultural and demographic evolutions.
Since the age of the discoveries, and for much of the 20th century, emi-
grantes have travelled in a worldwide search for heightened economic
opportunity (Moreira 2005). As a result, sizeable Portuguese communities
are now established in South and North America, South Africa, Australia
and much of the rest of the European Union (Baganha 2003). The largest
groups can be found in Brazil and the United States with approximately
1.2 million, in each case (Lopes 1997). Traditionally, Brazil was a favoured
destination (Volpi Scott 1999) but towards the latter part of Portugal’s
authoritarian era (during the 1960s and 70s), young Portuguese men
were emigrating in unprecedented numbers towards north-western

From Porto to Portadown: Portuguese workers in Northern Ireland’s labour market 173
PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 174

1 The longevity Europe: driven out by a repressive political regime. At its height in 1970,
associated with 173,000 individuals (Serrão 1977) left this part of the Iberian Peninsula;
immigrants and their most heading for jobs, and potentially more rewarding and safer destina-
relative integration tions in France, Germany, Luxembourg and Switzerland (Branco 2001).
into British society is
reflected in the second In addition, untold numbers emigrated illegally, anxious to avoid conscrip-
highest level of British tion and embroilment in the ultimately ill-fated attempt to control
citizenship being
granted. In 2004, for Portugal’s African colonies (Corkill 1999: 25). After the revolution in
example, 545 1974, more relaxed societal controls led to large numbers of emigrants
Portuguese nationals
were confirmed with
returning home. Some were forced back (as retornados) from the former
this status (Home colonies whilst others travelled home voluntarily (as regressos) from north-
Office 2005). ern Europe (Rato 2001). These returnees, along with increasing numbers
of immigrants from the Cape Verde islands, Brazil, Angola, China, and
parts of north-western, central and eastern Europe, contributed to a turn-
around in the country’s migration balance. Researcher’s attentions shifted
to evaluate this net-inward migration of flows (Fonseca 2001; Eaton
2002) while generally ignoring the continued outflow of the domestic
population. This was unfortunate because as a renewal of economic reces-
sionary conditions impacted in the first half of the current decade, emigra-
tion from Portugal re-emerged at rates of between 21,000 and 27,000 each
year. Indeed, since the start of 2000, over 96,000 Portuguese (almost 1 per
cent of the total domestic population) have left their country of birth (INE
2000–2003). Many have been forced to leave as a result of home labour
market difficulties including growing unemployment, limited job opportu-
nities, higher interest rates, the rising cost of living, wage freezes and other
austerity measures imposed by successive governments (Economist
Intelligence Unit 2004) (Table 1).
As a result, Table One shows that, in the new millennium, almost
83,000 (86 per cent) Portuguese emigrants continued to follow the modern
route by migrating to central North-Western Europe. Switzerland and
France remained the principal destinations (Malaurie 1998; Marques
2001) but the United Kingdom had come to account for one in ten of all
emigrants. Indeed, the United Kingdom outstripped Germany (Bauer et al.
2002), Spain and Luxembourg as a main receiver of Portuguese migrants.
Seventy-three per cent of emigrants were classified as temporary (less than
one year), short term or seasonal migrants, but in the British case, the
ratio was significantly different with 43 per cent being labelled as long-
term, permanently settling (more than one year) emigrants. Almost 10,000
migrants were recorded as having travelled to the United Kingdom between
2000 and 2003, and the rate of outward movement was accelerating. When
coupled with voluntary Portuguese consular registrations it was clear that
Britain had gained significantly in its attraction.1 Nevertheless, it is impos-
sible to derive an accurate total not least because Portuguese citizens are
allowed to circulate freely around the EU. Almeida (2006: 6) recognised
the usefulness of the British labour force survey, which suggested that
there were 85,000 Portuguese citizens living in the United Kingdom in
2005. However, some commentators believed there to be nearly 110,000

174 Martin Eaton


2000 2001 2002 2003 Total (2000–2003)
Area/Country of

Destination Tem. Per. Tot. Tem. Per. Tot. Tem. Per. Tot. Tem. Per. Tot. Tem. Per. Tot.
North Western Europe 13,561 3855 17,416 13,473 4359 17,832 15,308 6922 22,230 19,347 5909 25,256 61,689 21,045 82,734
Switzerland 4718 1113 5831 3558 247 3805 6038 2240 8278 3879 907 4785 18,193 4507 22,699
France 2431 609 3040 3444 2229 5673 4124 1838 5962 6550 849 7399 16,549 5525 22,074
2:27 PM

United Kingdom 1416 675 2091 1441 501 1943 984 881 1865 1705 2187 3893 5546 4245 9792
Germany 1495 1064 2559 X X 1970 X X 986 1443 955 2398 2938 2019 7913
Spain X X 1177 X X 1175 2524 404 2928 X X 2247 2524 404 7527
Luxembourg X X X X X 1415 X X 704 1266 770 2036 1266 770 4155
Page 175

Other N. West
European countries 3501 394 2718 5030 1382 1851 1638 1559 1507 4504 241 2498 14,673 5380 8574
Other countries
worldwide 3080 837 3917 1354 1403 2757 3237 1891 5128 974 778 1752 8645 4909 13,554
Total for all emigrant
destinations 16,641 4692 21,333 14,827 5762 20,589 18,545 8813 27,358 20,321 6687 27,008 70,334 25,954 96,288
Tem., temporary emigration for a period of less than one year; Per., permanent emigration for a period of more than one year; Tot., total of temporary and permanent emigration;
X, information not available.
Source: Instituto Nacional de Estatística (2000–2003).

Table 1: Portuguese emigration to North Western Europe, 2000–2003.

From Porto to Portadown: Portuguese workers in Northern Ireland’s labour market

PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 176

2 Registration of Portuguese resident (Anon 2005) although, in turn, the real figure
Portuguese nationals (according to the Portuguese Consulate General in London2) could
at the Consulate
General (PCG) is a be more than twice as high at around 250,000 nationals (Almeida
voluntary activity and 2006: 12).
the figure of 250,000
immigrants reflects Whatever the true figure, it is clear that the Portuguese have been spa-
speculation on the tially drawn to two main locations. The first of these is the Channel
part of the PCG
(Almeida 2006). Islands. Most emigrants found jobs in the horticultural and tourism indus-
tries of Jersey and Guernsey, where employment of Madeiran emigrants in
the hotel trade remains important (Anon 2004; Beswick 2005). Today,
Jersey has a population of around 6000 Portuguese, which grows to
10,000 annually as a result of seasonal employment fluctuations. There
has also been a pattern of emigrants locating in central London, and more
especially the boroughs of Kennington, Lambeth and Stockwell, where
employment in cleaning and domestic service remains a significant
feature (Campos and Botelho 2001: 3). This capital based community is
well established and can be considered a socially coherent entity number-
ing up to 27,000 (Benedictus 2005). Indeed, it mimics traditional
Portuguese enclaves found in parts of France (Volovitch-Tavares 1999)
even to the extent that they have produced their own version of the Yellow
Pages commercial telephone directory – as Páginas Portuguesas – detailing a
myriad of Portuguese owned but British based services (Ramalho 2006).
These included café-bars, restaurants, delicatessens, lawyers, doctors and
hairdressers, as well as fostering community centres, social clubs and an
expatriate football league. Many luso-families have produced first and
second generation offspring who attend British schools (Abreu et al.
2003) and are culturally and dialectically assimilated into the host com-
munity. More recently still, there has been spatial distribution of the
Portuguese emigrant population towards peripheral regions of Britain.
Trails have developed, for example, towards East Anglia to help in agricul-
ture (John 2003), and towards north-west England (around Manchester)
where the Portuguese work mainly in food production factories. In a post
millennium shift, workers have also begun gravitating towards Wales, the
Scottish Borders and Northern Ireland (Corkill and Almeida 2007).

Portuguese migrants in Northern Ireland

Northern Ireland’s foreign population is dominated by Chinese (estimated
at 7–8000 individuals), Indian (1500), and African (1600) communities
(Multi-cultural Resource Centre 2002). More recently, east European
immigrants have proliferated with significant numbers travelling from
Lithuania, Russia and, in particular, from Poland (STEP 2006). Nevertheless,
as part of a provincial population of almost 1.7 million, foreigners remain
a small minority. Given the paucity of information on the region’s foreign
population and the failure of the NI Census 2001 to delve beyond generic
ethnic groupings, once again, accurate migration data on national group-
ings was difficult to attain. Conservative estimates placed the region’s
Portuguese population somewhere between 700 (MCRC 2002) and

176 Martin Eaton

PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 177

1000 individuals (Holder 2003: 74). In turn, most were Portuguese

nationals with a small minority (of 10 per cent) being Portuguese speaking-
individuals from third countries such as Angola, Brazil, Mozambique and
East Timor. Data relating to the allocation of national insurance numbers
showed that between 2002 and 2005 a total of 1630 Portuguese were
registered with the Social Security Agency (DSDNI 2006: 22). Given that
the mobilisation was a recent phenomenon and the situation constantly
changing, as well as there being no requirement to register (or de-register)
with the Portuguese Consulate (in Manchester), then the true figure
for Portuguese immigration was perhaps higher still (at approximately
2500 nationals). The question marks surrounding these figures reflected
the weak statistical hold of government agencies and the unrecorded
flux associated with the migratory flows, both into and out of the
province. Nevertheless, what is clear is that the Portuguese represented
a minor grouping within a small foreign minority population in Northern
In spite of this limited status, recent media attention highlighted the
Portuguese community but delivered mixed messages as to their impact.
On one hand, they were viewed as a positive player in the regional
economy, helping to support agriculturally based industries that have suf-
fered from acute labour shortages in recent years. At the same time, some
media sources have painted them as a disruptive group (Tyrone Today
2006) citing problems such as anti-social behaviour, abuse, harassment,
intimidation and violence (issues reported in Público 2002) in the work-
place. In autumn 2002, a locally produced TV documentary exposed some
of the difficult conditions and experiences of Portuguese emigrants
working in Dungannon’s meat processing factories (Collins 2002). The
programme alleged that wages paid to the Portuguese workers were gen-
erally lower than the salaries paid to local Northern Irish employees
working in the same factory. Survey of the Portuguese community in
Northern Ireland by Soares (2002) found that the recruitment process
was demographic and gender specific with almost nine out of ten workers
being young, single males, between 22 and 31 years of age. Most returned
to Portugal at the end of their six-month temporary work contracts.
Educational attainment amongst these immigrants was of an average
standard, with over three-quarters having completed their secondary level
school education. These were a replication of the neo-classical migration
chains established by temporary labour migrants in other semi-industri-
alised areas, and particularly those previously moving from, into, and via
Portugal (Castles 2002). Indeed, almost seven out of ten migrant labour-
ers interviewed had worked in other north European countries, namely
France, Germany, and Switzerland (Soares 2002: 78). Many workers,
therefore, had experienced similar work posts and labour schemes operat-
ing in these countries. Consequently, Northern Ireland could be seen as
another cog in the wheel of migratory circulation that now typifies this
sort of semi-rural labour market arrangement that is pan-West European

From Porto to Portadown: Portuguese workers in Northern Ireland’s labour market 177
PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 178

in scope, with workers moving as a reaction to where the job demands

were emanating from.
In spatial terms, the Portuguese were found located in small but highly
concentrated numbers in the new/market towns of mid-Ulster. These
included the localities of Portadown and Craigavon (each conservatively
estimated to contain around 200 Portuguese nationals), and in the initial
focus of location – Dungannon (estimated at 300) (McGreevy and Bayne
2001). More recently, Portuguese workers were drawn into the smaller
outlying market towns of Antrim, Ballymena, Ballymoney, Banbridge,
Coalisland, Coleraine, Cookstown, Kilkeel, Limavady, Magherafelt, Omagh,
Newtownards and Rathfriland (NISRA 2006). What was unusual about
this distribution was that, the main population centres of Belfast, Lisburn
and Londonderry/Derry were largely ignored by Portuguese emigrants.
This was because migrant workers were responding to change and short-
fall in the rural economy and filling an: ‘unmet demand for low paid
labour’ (TUC 2004: 2): effectively substituting themselves into the local,
agriculturally based labour markets of Ulster where activities including
mushroom picking, sandwich-making and potato-packing factories
remained important. Indeed, Northern Ireland’s food processing industry
was the third largest manufacturing sector in the region, employing
19,000 workers and producing £2.5 billion sales in 2005 (Anon 2007).
Moreover, much of the produce was exported and this value-added trade
accounted for: ‘almost two-thirds of the food processing sector’s output’
(OFMDFM 2004), thus helping to maintain its position as a staple indus-
try, alongside tourism, financial services, and electronics manufacturing
sectors. Northern Ireland remains a relatively buoyant regional economy
with low unemployment (4.2 per cent – April 2007), increasing levels of
employment activity (an EAP of 786,000), and falling numbers of persons
claiming unemployment related/incapacity benefits (around 24,000) in
spring 2007 (DETI 2007).

Portuguese workers’ impacts

Because of the sensitivity associated with this topic area, language con-
straints and the levels of suspicion now surrounding the role of agencies, as
well as alleged interference on the part of supervisors, it proved impossible
to conduct a quantitative inquiry aimed at the workers. However, to coun-
teract these difficulties and begin examining the experiences of these labour
migrants a series of in depth, semi-structured exploratory discussions were
held with representatives closely associated with the Portuguese working in
Northern Ireland. These respondents included a female migrant worker
from Lisbon (Interviewee A), a local supermarket line manager (B), two
food processing factory production managers (C and D), a local commu-
nity/church worker (E), the managing director (F) and manager (G) of
a sandwich making factory, a Brazilian born, male immigrant worker (H), a
Portuguese family support worker (I), a local newspaper editor (J) and
a former recruitment agent (K). Our informants were carefully selected as

178 Martin Eaton

PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 179

possessing knowledge of the labour market issues specifically surrounding

the migrants in two of the main destination towns: Portadown and
Dungannon. This important and detailed qualitative information was col-
lected in the spring of 2006 and focused upon the segmentation of the local
labour market, the problems and benefits emerging, and the integrative
process undertaken by migrant workers and local communities alike. We
should, of course, be aware that the empirical base was narrow and opin-
ions put forward were value laden. Nevertheless, quality and integrity of
responses was high and allowed for tentative comment to be made.
Our discussions showed, for example, that at the micro scale the influx
of Portuguese workers into Northern Ireland has been rapid and a very
recent phenomenon.

Interviewee E gave an indication of the timeline whereby: The first group of

people to come were the Portuguese appearing here in 2000. At the start it
was all males arriving most of which were aged between 20 and 30. In the
last few years (however) there have been many families and middle-aged
people coming to live and work here.

Not surprisingly, our investigation showed that Portugal’s immigrant com-

munity is a small and largely hidden group whose relative ‘invisibility’ was
based upon their inherent desire not to draw attention to themselves.
While their numbers were the subjects of debate it was quite likely that
current estimates of between 1000 and 2500 newcomers were conserva-
tive. Taking account of unrecorded immigration and family reconciliation
the real figure was much higher at anything up to 6000 individuals
(reflecting speculation on the part of ground worker E). In spite of this
uncertainty, most Portuguese immigrants were now part of a distinctive
labour market where the immigrant’s importance was repeatedly stressed.
Indeed, without the migrant labourers then, it was likely the meat pro-
cessing factories (and the rural agricultural economy) in many parts of the
region would struggle to survive.

Interviewee C, for example, was unequivocal on this topic: Local people don’t
want to work in this (chicken processing plant) environment. Moreover: We
can’t get local people to do the work so if it weren’t for foreign workers the
company would not be able to operate. We have such high productivity and
market demand that if we failed to meet it, the factory would close down and
(all of us) would be out of jobs.

The inherent flexibility, ready compliance and low skill requirements

afforded by the immigrant worker (in comparison to a local employee) also
lay at the heart of the business decision to take on foreign nationals. This
was particularly true in terms of the seasonally fluctuating, generally
long, hard and unsocial hours that were associated with shift working pat-
terns in these sectors. To this end, G attested that:

From Porto to Portadown: Portuguese workers in Northern Ireland’s labour market 179
PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 180

The food industry has unpredictable hours at best . . . (this factory) starts at
seven and does not finish until production has stopped. For local workers
with families to support, this is not seen as an acceptable condition.
However, it is perfect for foreign nationals who (in his perception) . . . have
no families to support in the immediate area.

D argued:

the migrants have been able to work overtime in the past when it was not
wanted (by the local workers), for instance, at Christmas and New Year holidays.

This demand-led argument meant that the employers were generally

happy with their supply relationships with the labour agencies (Cains
2004: 5) and were, therefore, reluctant to intervene to make agents change
their ways or regulate their activities. Indeed, the factories had garnered
the fruits of a well-motivated, highly productive and relatively docile immi-
grant labour force that was easily recruited and praised for its intrinsic
work ethic. This positive image was one reiterated by:

F who said that his factory had been able: to use a large population of
(Portuguese) workers keen to work at any time, any holiday to maximum
effect, and as:
B affirmed: we have (Portuguese) agency workers who come in to do
shifts we have trouble filling at night.

The role of the labour agents and the agencies they represented was
clearly very strong and often extended from pre-arrival through to their
initial location in Northern Ireland:

A’s narrative appeared typical of many and explained that: I had to give
(name of employment agency) a cheque for £250 in order to secure my
plane ticket . . . On arrival, (name of agency) had organized accommodation
for us but it was rough with no heating, oil, electric and very little

A synthesis of information supplied by the interviewees and K (in particular)

showed that these employment agencies were both internationally and
locally based, in Oporto (Portugal), and in the island of Ireland (in Belfast
and Dublin). A typical agency attracted workers to factories across Northern
Ireland (Bell et al. 2004: 54) through adverts placed in Portuguese daily
newspapers (e.g. Correio de Manha). The agency interviewed potential
workers and completed medical checks before signing them up and flying
them to Belfast. Once in the region, they were accommodated in shared
lodgings of variable quality with several other agency employees and then
assigned to meat processing companies and food packing firms in one of the
main foci/semi-rural towns mentioned previously.

180 Martin Eaton

PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 181

Source: Author’s schematic interpretation (2007).

Figure 1: Portuguese immigrant insertion in the Northern Ireland Labour Market.

Figure 1 interpreted the segmented nature of this labour arrangement

and the way in which rental and transport costs were deducted from the
employee’s wages, which in turn were paid by the agency and not directly
by the factory of employment. In turn again, the agency was paid by the host-
company at minimum working wage (MWW) level for the production work
that was provided by the migrant. The agency then paid the migrant labourer
a wage (minus deductions) at a basic level that it determined. Often, the
money was paid ‘cash in hand’ and via a second intermediary who was
normally a native, Portuguese speaking, charge-hand. It was along this
continuum that the labour market changed from formalised to informal and
into a ‘grey’ area where regulatory controls were more difficult to enforce.
Unsurprisingly, this sub-contracting arrangement became increasingly com-
plicated in its operation and led to allegations that migrant workers were
being exploited. Indeed, many were considered victims of a ‘long hours/low
pay’ syndrome (TUC 2003) that has proliferated elsewhere in Britain.

From Porto to Portadown: Portuguese workers in Northern Ireland’s labour market 181
PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 182

Problems in the labour market

The major problem to emerge from the investigation, therefore, related to
the employment situation, and more specifically, the issues of exploitation,
discrimination, and physical/verbal abuse in the workplace. These hazards
were often perpetuated by intimidation and fear on the part of intermedi-
aries. Indeed, in many respects these workers were dependent upon the
‘generosity’ of their labour agents but vulnerable to (unfair) dismissal and
ejection from tied accommodation at short notice.

Interviewee A, for example, alleged that: Portuguese people and other for-
eigners . . . were complaining that (Mr X) was bullying and threatening
them . . . I knew of them to put Portuguese out of their homes which they
were renting off (Mr X), during the night if they were going to shift jobs that
were not a part of the (named) agency.

Rivalry and competitiveness amongst the controlling agents was a factor,

therefore, and echoed a similar system operating in Portugal but relating
to some Portuguese agent’s nefarious treatment of Lusophone African
immigrants (Eaton 2003: 108) during the 1990s. Moreover, there were
parallels with immigrants entering Portugal that were carrying similar
life experience backgrounds, since the Portuguese emigrants possessed a
strong work ethic and had family values at the root of their decision to
emigrate (McGreevy and Bayne 2001). These drivers included a desire
to earn money, remit savings, improve their lives and those of their
family, educate their children and broaden their horizons. Many of these
migrants, therefore, made independent choices in order to maximise their
incomes and other opportunities within the constraints that they faced.
This reflected a dual frame of reference that they carried with them (King
1998: 270) or a trade-off, whereby poor wages and working conditions
abroad were tolerated because the wages earned in Northern Ireland were
higher than any potential earnings in their home country. Portuguese
workers appeared to accept their circumstances in deference to the remu-
neration that they received: sterling, which could be converted into a rela-
tive euro fortune when repatriated, thus helping to improve familial
circumstances, in their source areas. However, in extreme cases and
according to:

C, for example: If foreign workers have trained to a higher level and are
working through an agency they still only receive(d) the basic rate, so in
some cases they were exploited through their agency.

This scenario was, however, complex and relative to the individual’s cir-
cumstances since one person’s ‘self-exploitation’ represented another
person’s ‘trade-off ’. Interviewee H, for instance, was previously a car
mechanic in his native Brazil and:

182 Martin Eaton

PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 183

made more money than most but this only allowed (him) to have a basic
standard of living (at home). It is nothing compared to what (he made) now
(in Northern Ireland). H explained that he currently had a specialist (food
processing) job that meant he: received a higher salary.

A related feature was the integration process that these immigrant

workers underwent in the face of wider societal issues such as racial
tension, verbal abuse, and occasional physical violence that was often a
reciprocal process. There was, for example, some evidence of intimidation
and violence directed against fellow nationals. Indeed, Portuguese ‘work
supervisors’ were alleged to be coercing the contracted immigrant work-
force into rejecting trade union membership and encouraging the
Portuguese to inform on each other with respect to misdemeanours, both
trivial and serious. The consequences were sometimes extreme:

A: I know a Portuguese guy who had a bottle smashed over his head. Also
(name of NI labour agent deleted) and his crowd treat Portuguese like slaves
and carry out wrongful beatings which I would class as racial.

Equally, there was some observational evidence of intimidation of the

indigenous community (e.g. street begging, vagrancy and casual violence),
and conversely instances of harassment of workers by locals, particularly,
in social interaction arenas such as public houses and nightclubs. Experience
amongst the respondent workers was mixed but the underlying theme was
one of conflict.

H, for example, stated that: (he) had one or two problems with local people . . .
mainly occurring on nights out in (a nearby) town.
A claimed that: one (Portuguese) guy was called a monkey because he
was black and people call us smelly and think we are diseased. (Some) homes
are targeted with petrol bombs.
D spoke of: some resentment towards the migrant workforce.

It appears that the harassment was not exclusively, therefore, the domain
of the local resident/worker against the immigrant but more a two-way
process reflecting confrontational attitudes on the part of some of the
immigrant population (agents/supervisors/workers) against each other. A
complicated picture was clouded further by the alleged involvement of
local paramilitary vigilantes (BBC 2004) looking to ‘control’ what they
considered to be ‘their’ communities.

Interviewee E stated: Most attacks are from . . . youth mobs that are linked to
paramilitaries and it is the paramilitaries who control the attacks. Some of
the houses, which are rented to Portuguese people, are paramilitary owned
and these people . . . (then) demand £20 per week from the Portuguese who

From Porto to Portadown: Portuguese workers in Northern Ireland’s labour market 183
PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 184

live in them to ensure their windows stay in . . . and that they receive no

This, of course, had a multiple depressant effect since the Portuguese

workers had monies deducted at source by the labour agent and allegedly
by their supervisor, paid rent to the paramilitary ‘landlord’ and then had
to pay protection money to the same ‘landlord’ to secure a safe living exis-
tence. From a spatial perspective, many immigrant workers lived in
rundown, dilapidated interface areas between Unionist and Nationalist
communities and were, therefore, often on the ‘front-line’ at times of
heightened tension such as during the Orange Order parading season.

Community responses
A further complication emerged, since the process of adaptation by
Portuguese immigrants to the local society, and indeed, by local communi-
ties to the Portuguese (and other immigrant communities) has been slow.
Mobilisations by organisations such as the Northern Ireland district coun-
cils and agencies like the Citizens Advice Bureaux, to help local integration
efforts were only a very recent development. Relative invisibility within the
community meant it was difficult for the region’s social services, for
example, to help the workers. A simple lack of knowledge of an exact
number of immigrants resulted in resource issues often being ill informed
and poorly determined. It was initially difficult to overcome factors such as
efficient provision of English language classes, the proper distribution of
interpreting and translating services, or improved access to health ser-
vices, education systems, welfare benefit offices, and so forth.
The situation has, however, changed with greater recognition and
involvement through, for example, local councils providing translators
and translations of documents. Several simple but far-reaching transitory
arrangements emerged.

Interviewee B (referring to a national supermarket retailer) stated that:

In the Dungannon store they put up signs in Portuguese for taxis and
(one) manager was sent to learn Portuguese. In this . . . store we have an
outside firm (which) acts as a translator and is . . . used when the cus-
tomer requires one.

Other retailers specifically employed Portuguese in their human resource

departments to assist with their recruitment process and facilitate direct
employment strategies, thus removing the agent from the employment

C noted that: Within personnel we have a Portuguese girl who assists in the
(worker) interviews and also checks identity cards (against false representa-
tion and fraud).

184 Martin Eaton

PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 185

In addition, the Police Service of Northern Ireland employed an interpreter

for the mid-Ulster area and community funded initiatives such as the South
Tyrone Empowerment Programme (STEP) was utilised to support the
immigrant community (STEP 2005). Migrant workers’ forums were set up
in Craigavon and a community/voluntary partnership called ANIMATE
(Action Now to Integrate Minority Access to Equality) was working on
migrant worker issues in the Dungannon, Craigavon and Cookstown areas
(Craigavon Borough Council 2005). Bus-mounted advertising campaigns
aimed at disseminating an anti-racism message in the workplace, were also
instigated in Dungannon in the spring of 2006 (DSTBC 2006: 10).
Moreover, a local newspaper, the Tyrone Courier, published a weekly column
in Portuguese devoted to matters of local interest to the immigrants. This
was tangible recognition of the contribution that the Portuguese workers
have made: a point noted by J when he stated that:

There are more than two communities in Dungannon: the Portuguese are
now a sizeable group and an important part of life in Dungannon.

In some respect the Portuguese emigrant experience in the province mim-

icked that found in rural parts of mainland Britain such as Lincolnshire,
Norfolk and Suffolk. The situation they found themselves in was fluid and
the immigrant group was rapidly expanding. Most new entrants were
trying to ‘shuffle up the socio-economic pyramid’ (Eaton 2003), which
was based upon relative levels of aspiration, wages and living conditions
found in different parts of North-Western Europe.

When Interviewee A was asked: what factors made her decide to leave
Portugal, her answer was emphatic and echoed the views of most: money.

In turn, even the minimum-working wage (minus agent’s deductions)/basic

wage earned locally was higher than what could be made in Portugal where:

everything keeps going up in price . . . but the salary stays the same so it
becomes unaffordable to live there (Interviewee A, again).

Northern Ireland was, ergo, an outwardly attractive location pulling

workers into the province and providing jobs and rewards, with only the
climate negating against even greater levels of satisfaction since according
to Interviewee A: it is very cold here. Equally, the local education system
(once the initial English language barrier has been overcome) offered first
and second-generation immigrant worker children a much better prospect
of schooling than similar opportunities found back home (Newnham
2003). Indeed, the Portuguese in Northern Ireland quickly moved beyond
a pioneer stage and were now settling with spouses/partners and/or chil-
dren. As E attested:

From Porto to Portadown: Portuguese workers in Northern Ireland’s labour market 185
PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 186

There are 70 (Portuguese) kids now in school in Portadown, mostly in

primary schools but there are a few attending secondary school.

All of these favourable conditions, therefore, had the potential to allow further
improvement of the familial situation. Once again, they helped to justify the
trade-off associated with many of the immigrant’s work experiences and their
apparent willingness to tolerate the more negative aspects of their existence.

In conclusion, we have tentatively examined the scale, experiences and
impacts of Portuguese migrant workers in Northern Ireland’s labour
market. Our analysis argued that employment of Portuguese migrants, in
Portadown, Dungannon and other regional market towns, was signifi-
cantly driven by employment agencies. This process has provided factories
with a stable workforce and counteracted many of the labour shortages
associated with the rurally based food processing sectors. As a result,
working patterns based upon substitution and segmentation, have been
fundamentally altered. Equally, the role of agencies have brought adverse
consequences for some migrants in terms of reduced pay and inferior
working and living conditions compared to other (non-migrant) workers.
However, it appeared that this was a conscious decision-making process on
the part of the migrant worker to ‘trade-off ’ personal inconveniences in
deference to remunerative reward, remittances, and savings, which could
all be used to improve familial circumstances, both in Portugal and in
Northern Ireland. As such, many Portuguese appeared to tolerate the
dualistic operating conditions that they faced as well as the slowness asso-
ciated with the pace of developmental change and the process of adapta-
tion to, and on the part of, many local communities.
Consequently, the future for these types of immigrant is difficult to
surmise. It may be that with growing levels of migrant labourers and
increasing evidence of family reconciliation then greater integration can
be expected. Integration can take two forms, first, in terms of the commu-
nity. As we have observed, first-generation immigrant children are now
settling in Northern Ireland’s primary schools and with time will move
into the secondary (and tertiary) education sectors. It is likely that com-
munity groups/associations will be established, and continue to grow.
Hypothetically, they may come to mirror (on a smaller scale) the estab-
lished Portuguese social communities found in London. Fledgling exam-
ples already exist in Dungannon where the weekend use of a local
community centre together with a Portuguese owned restaurant and a
managed public house forms the hub for a local socialisation/integration
process to take place. Portadown has a public house with a strong
Portuguese clientele base, and a coffee house and shop selling Portuguese
goods, which acts as an informal drop-in support-centre offering mutual
advice and translation services. More importantly, these initiatives are
contributing to a relatively positive information chain that constitutes a

186 Martin Eaton

PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 187

key part of the strong worker migration trail that has now developed 3 There is debate over
between Portugal and Northern Ireland. whether there are a
finite number of jobs
Second, greater levels of integration can be anticipated in terms of the local available in the food-
labour market. Progress has been made with some migrant workers now processing sector of
Northern Ireland. It
being directly employed by the factory (rather than continuing to be linked to may be that
a labour agency), thus benefiting from bonus payments, training opportuni- saturation point is
being reached.
ties, language attainment, access to trade union membership, and closer Equally there were
immersion in the workforce and local economy. However, there is also a down- inferences from the
survey (interviewees
side because if labour agencies are not carefully regulated (Concordia 2006: B and E) that some
12) then problems of discrimination/exploitation demonstrated in the article local areas have been
could render the Portuguese immigrants as a vulnerable group in a society regenerated and the
contributions of
not characterised by its tolerance of ‘outsiders’. The omnipresent spectre of migrant workers have
violence is unlikely to go away completely and it is a disturbing prospect; one helped to improve
trade and produce re-
orchestrated by criminal paramilitary elements (both Republican and Loyalist) investment, increases
exerting what they see is ‘control’ over ‘their’ communities. in jobs and sustained
growth of factories
Equally, as segmentation in the labour market continues and potential (DETI 2007).
saturation point3 is neared with migrant workers continuing to enter
Northern Ireland and take up jobs that local workers are reluctant to take
on, then it is possible that conflict will develop. There is already some
observational evidence of friction developing between different groups of
immigrant workers (Portuguese and eastern Europeans, for example)
competing for the same job vacancies in mid-Ulster and this may be exac-
erbated in the longer term in a three way-internecine tension between the
local, Portuguese and East European working groups.
On the other hand, with time, co-operation and a level of tolerance, the
Portuguese workers and their families could be a welcome addition to
the establishment of multiculturalism and a multi-ethnic society within
the region. Such a community already typifies large urban centres in the
rest of the United Kingdom (i.e. London and to a lesser extent, the Channel
Islands and around Manchester) but is a process still in its infancy in
Northern Ireland. Moreover, it is a largely unknown concept in many
rural market towns, and more economically peripheral parts of the
province. This lack of experience of immigrant labourers and their contri-
butions will be a key factor in changing community relations and percep-
tions of the Portuguese workers. It is a feature that time will change but
one which will require all parties to come together to discuss their similar-
ities and differences. Given past experience, there is no guarantee that this
will happen. As a result, Portuguese workers in Northern Ireland remain
in a classical state of migratory flux. Many live in a hidden, partially
understood, sometimes abused, but important, gradually evolving, and at
the micro-employment scale, an increasingly influential community.

Abreu, G., de, Cline, T. and Lambert, H. (eds) (2003), The education of Portuguese
children: Insights from research and practice in England and overseas, Psychology
Department Research Report. Oxford: Oxford Brookes University.

From Porto to Portadown: Portuguese workers in Northern Ireland’s labour market 187
PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 188

Almeida, J.C.P. (2006), ‘Citizens of the world: Migration and citizenship of the
Portuguese in the UK’, mimeo provided by author (Migration Research Unit:
Anon (2004), ‘Portuguese community develops’, BBC News, http://www.bbc.
Accessed 12 May 2005.
——— (2005), ‘Portuguese invading Britain!’ The Resident – Weekly Algarve Edition,
Accessed 12 May 2005.
——— (2007), ‘Economy minister praises Kerry group’, Northern Irish News,
www.4ni.co.uk/northern_ireland_news.asp?id=62823. Accessed 6 July 2007.
Arroteia, J.C. (2001), ‘As comunidades portuguesas no mundo’, Janus, http://
www.janusonline.pt. Accessed 17 August 2005.
Baganha, M.I.B. (2003), ‘Portuguese emigration after World War II’, in Pinto, A.C.
(ed.), Contemporary Portugal: Politics, society and culture, New York, NY: SSM.
Bauer, T., Pereira, P.T., Vogler, M. and Zimmermann, K.F. (2002), ‘Portuguese
migrants in the German labour market: Selection and performance’, Interna-
tional Migration Review, 36 (2): 467–91.
Bell, K., Jarman, N. and Lefebvre, T. (2004), Migrant workers in Northern Ireland,
Belfast: Institute for Conflict Research.
Benedictus, L. (2005), ‘Every race, colour, nation and religion on earth: Part 3’,
Guardian Unlimited, Friday 21 January, http://www.guardian.co.uk/print/
0,3858,5108900-103680,00.html. Accessed 11 October 2005.
Beswick, J. (2005), ‘The Portuguese diaspora in Jersey’, in Preisler, B., Fabricius, A.,
Haberland, H., Kjaerbeck, S. and Risager, K. (eds), The consequences of mobility,
Roskilde: Roskilde University.
Borjas, G. (1999), Heaven’s door: Immigration policy and the American economy,
Princeton, NJ: Princeton University Press.
Borooah, V.K. and Mangan, J. (forthcoming), ‘Love thy neighbour: how much
bigotry is there in Western countries?’ Kyklos.
Branco, J.P. (2001), ‘A comunidade portuguesa em França’, Janus, http://www.
janusonline.pt. Accessed 17 August 2005.
British Broadcasting Company (2004), ‘Petrol bomb attack was racist’, BBC News,
northern_ireland. Accessed 16 August 2005.
Cains, B. (2004), ‘Managing a non-national workforce: Employers' perspectives’, in
Equality Commission for Northern Ireland, Employing migrant workers in North-
ern Ireland, Belfast: ECNI.
Campos, A. and Botelho, L. (2001), ‘O trabalho temporário e as novas formas de
clandestinidade’, Janus, http://www.janusonline.pt. Accessed 17 August 2005.
Castles, S. (2000), ‘International migration at the beginning of the 21st century:
Global trends and issues’, International Social Science Journal, 165: 269–81.
——— (2002), ‘Migration and community formation under conditions of globali-
sation’, International Migration Review, 36 (4): 1143–68.
Collins, E. (2002), Migrant workers and the work of the Equality Commission for NI,
Belfast: ECNI.
Concordia (2006), Migrant workers in Northern Ireland, Dungannon: Concordia.
Corkill, D. (1999), The development of the Portuguese economy: A case of Europeaniza-
tion, London: Routledge.

188 Martin Eaton

PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 189

Corkill, D. and Almeida, J.C. (2007), ‘The three Ps: Portuguese migrant labour in East
Anglia’, Paper to Conference on the Portuguese-Speaking Diaspora in the UK,
Manchester Metropolitan University, 28 June.
Craigavon Borough Council (2005), CBC community relations plan 2005–2006,
Craigavon: CBC.
Department of Enterprise, Trade and Investment (2007), ‘Northern Ireland’s
unemployment rate remains low’, http://www.detini.gov.uk/cgi-bin/morenews?
utilid=1068. Accessed 18 July 2007.
Department for Social Development Northern Ireland (2006), National Insurance
number allocations to overseas national entering Northern Ireland, Belfast: NISRA.
Dungannon and South Tyrone Borough Council (2006), Good relations audit and
strategic plan 2006–2007, Derry: Holywell Consultancy.
Eaton, M.D. (2002), ‘International population mobility: Immigration and labour
market change in Portugal’, in Syrett, S. (ed.), Contemporary Portugal: Dimen-
sions of economic and political change, Aldershot: Ashgate.
——— (2003), ‘Portugal’s Lusophone African immigrants: Colonial legacy in a
contemporary labour market’, in Lloyd-Jones, S. and Pinto, A.C. (eds), The last
empire: Thirty years of Portuguese decolonisation, Bristol: Intellect.
Economist Intelligence Unit (2004), Country profile 2004: Portugal, London: EIU.
Fonseca, M.L. (2001), ‘The geography of recent immigration to Portugal’, in
King, R. and Beck, J.M. (eds), Geography, environment and development in the
Mediterranean, Brighton: Sussex Academic Press.
Hayes, B.C. and L. Dowds (2006), ‘Social contact: Cultural marginality or economic
self-interest? Attitudes towards immigrants in Northern Ireland’, Journal of
Ethnic and Migration Studies, 32 (3): 455–76.
Holder, D. (2003), In other words? Mapping minority ethnic languages in Northern
Ireland, Belfast: Multi-Cultural Resource Centre.
Home Office (2005), ‘Persons granted British citizenship: United Kingdom, 2004’,
Home Office Statistical Bulletin, 08/05, 17 May.
Instituto Nacional de Estatística (2000–2003), Estatísticas demográficas, Lisboa: INE.
John, C. (2003), ‘Woes of Boston’s gang workers’, BBC News UK, http://news.bbc.
co.uk/1/hi/uk/3223560.stm. Accessed 16 November 2004.
King, R. (1998), ‘From guest workers to immigrants: Labour migration from the
Mediterranean periphery’, in Pinder, D. (ed.), The New Europe: Economy, society
and environment, Chichester: Wiley.
Lawless, J. (2005), ‘Remoteness lures immigrats to Iceland’, Guardian Unlimited,
Accessed 12 May 2005.
Lopes, P. (1997), ‘Emigração e communidades portuguesas no estrangeiro’, Janus,
http://janusonline.pt/1997/1997_3_15.html. Accessed 9 August 2005.
Malaurie, G. (1998), ‘La fierté retrouvée des portugais de France’, L’Européen, 8:
Marques, J.C.L. (2001), ‘A emigração portuguesa para a Europa: Desenvolvimentos
recentes’, Janus, http://janusonline.pt. Accessed 9 August 2005.
McGreevy, L. and Bayne, P. (2001), ‘Portuguese pouring into Dungannon’,
Tyrone Courier, 5 Spetember, http://www.ulsternet-ni.co.uk/cour3601/cpages/
CMAIN.htm. Accessed 16 August 2005.
Moreira, H. (2005), ‘Emigração portuguesa (estatísticas retrospectivas e reflexões
temáticas)’, Revista de Estudos Demográficos, 38: 47–65.

From Porto to Portadown: Portuguese workers in Northern Ireland’s labour market 189
PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 190

Mulholland, C. (2005), ‘Defending the rights of migrant workers’, Workplace News –

North, March, www.socialistparty.net/pub/pages/socialist004mar05/13.htm.
Accessed 25 October 2006.
Multi-Cultural Resource Centre (2002), Estimated populations of minority ethnic
communities in Northern Ireland, Belfast: MCRC.
Newnham, D. (2003), ‘Between two worlds’, Times Educational Supplement, 28
Northern Ireland Statistical Research Agency (2006), Components of population
change by administrative area, 2004 to 2005, Belfast: NISRA.
Office of the First Minister and Deputy First Minister (2004), Northern Ireland: Take
a closer look, Belfast: OFMDFM.
Público (2002), ‘Portugueses vítimas de ataques racistas na Irlanda do Norte’,
Público, 29 April.
Ramalho, J. (ed.) (2006), Páginas portuguesas: A lista telefónica portuguesa publicada
no Reino Unido, Pinner: JR Publications.
Rato, H. (2001), ‘O retorno dos emigrantes’, Janus, http://www.janusonline.pt.
Accessed 17 August 2005.
Serrão, J. (1977), A emigração portuguesa, Lisboa: Horizonte.
Soares, A. (2002), Relatório sobre trabalhadores portugueses na Irlanda do Norte,
Belfast: MCRC.
South Tyrone Empowerment Programme (2006), ‘Linguistic diversity in Dungannon’,
Accessed 25 October 2006.
South Tyrone Empowerment Programme (2005), ‘Portuguese praised for
contribution’, www.communityni.org/index.cfm/section/News/key/8A728155-
1143-D8AC-6. Accessed 25 October 2006.
Spencer, S. (2003), ‘The challenges of integration for the EU’, Migration Information
Source, http://www.migrationinformation.org/Feature/display.cfm?ID=170.
Accessed 18 August 2005.
Trade Union Congress (2003), Overworked, underpaid and over here: Migrant workers
in Britain, London: TUC.
Trade Union Congress (2004), Migrant workers are propping-up rural and small-town
Britain, London: TUC.
Tyrone Today (2006), ‘Stabbing appeal to foreign nationals’, Tyrone Today, http://
www.tyronetoday.co.uk/ViewArticle2.aspx?sectionID=2617&Article ID=13394.
Accessed 25 July 2006.
Volovitch-Tavares, M.C. (1999), ‘Les années fondatrices des portugais de France:
1947–1974’, Migrance, 15: 44–59.
Volpi Scott, A.S. (1999), ‘Une histoire d'adieux: L'émigration portugaise vers le
Brésil (1822–1914)’, Migrance, 15: 32–43.

Suggested citation
Eaton, M. (2007), ‘From Porto to Portadown: Portuguese workers in Northern
Ireland’s labour market’, Portuguese Journal of Social Science 6 (3): 171–91, doi:

Contributor details
Martin Eaton is a Reader in the University of Ulster’s School of Environmental Sciences
and since 2003 has been an International Fellow of the Gilbert M. Grosvenor Center

190 Martin Eaton

PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 191

for Geographic Education at Texas State University and International Scholar at

the Universidad de La Serena in Chile. His research interests include EU regional
development processes, industrial geography, analyses of the periphery (especially
Iberia) and migration processes in Portugal and Northern Ireland. He is the author
of several articles on Portuguese immigration and demography. Contact:
Dr. Martin Eaton, Reader in Human Geography, School of Environmental Sciences,
University of Ulster, Coleraine, Northern Ireland, BT52 1SA, UK. Tel: 02870 324663.
Fax: 02870 324911.
E-mail: M.Eaton@ulster.ac.uk

From Porto to Portadown: Portuguese workers in Northern Ireland’s labour market 191
PJSS-6-3-04-Eaton 3/15/08 2:27 PM Page 192
PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 193

Portuguese Journal of Social Science Volume 6 Number 3. © Intellect Ltd 2007.

Article. English language. doi: 10.1386/pjss.6.3.193/1

Minority representation in
Portuguese democracy
André Freire CIES-ISCTE

Abstract Keywords
Using Lijphart’s framework concerning different models of democracy, the objec- representation
tive is to provide a very brief overview of the main social and political divisions in minorities
Portuguese society, and to present the main institutional features of Portuguese democratic model
democracy and the possibilities they offer for minority representation. The article Portugal
starts by looking at the main social and political divisions in Portuguese society.
Then, the main institutional characteristics of Portuguese democracy, as regards
the ‘executive-parties dimension’ and the ‘federal-unitary dimension’, and the
opportunities they offer for representation of minorities in Portugal, are presented.
The article uses a perspective which is both longitudinal (1974 to the present)
and comparative (Portugal in the context of Western Europe). The article ends
with some brief conclusions.

Various different models can be found for democracy, both in normative * Paper prepared for
theories for democracy, and in the domain of empirical political science presentation at the
Seminar on ‘Minority
(Dahl 1998; Lijphart 1977, 1989, 1999; Beethem 2005; Held 2005). representation in
However, considering that our topic here is ‘minority representation in national parliaments’
organised by the
Portuguese democracy’, the theoretical model best suited to this type of Council of Europe,
analysis is that conceived by Arend Lijphart on majoritarian and consen- Committee on Rules
of Procedure and
sus democracies (Lijphart 1977, 1989, 1999; Horowitz 1985; Reynolds Immunities,
1999; Reilly 2001; O’Flynn and Russell 2005). Assembleia da
República, Sala do
In his various works on the subject, Lijphart starts out from the premise Senado, Lisbon, 29
that modern democracies are fundamentally representative. So if government September 2006. The
is not exercised (directly) by the people, the types of democratic regimes are author wishes to
thank the Deputy
basically differentiated by the varying answers to the question ‘who should Ana Catarina Mendes
govern’? According to the ‘majoritarian model’ of the Westminster type, for her invitation to
take part in this
the answer is the representatives of the majority of the voters – ‘the major- seminar.
ity’. According to the ‘consensus democracy model’, the answer is the rep-
resentatives of the largest possible part of the various segments into which
the electorate is divided – ‘as many people as possible’.
Each type of democracy is associated with an integrated set of political
institutions, which function as a system of incentives and constraints for
the activities of social and political actors. With regard to the election of
representatives and the decision-making process at central government

PJSS 6 (3) 193–211 © Intellect Ltd 2007 193

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 194

Consensus – ‘Who
Majoritarian – ‘Who governs? As many
Empirical dimensions governs? The majority’ people as possible’
Executive-parties Concentration of power Power-sharing within the
in the executive executive (‘enlarged’
(single-party coalition government)
government) Power-sharing between
Executive dominance of executive and
the legislature legislature
Bipartisan system Multi-party system
Majoritarian electoral Proportional electoral
system system (PR)
Pluralism of interest groups Neo-corporatism
Paradigmatic examples: UK, New Zealand (until Switzerland, Belgium,
countries, etc. 1996), Barbados European Union
Type of societies usually Non-plural societies (i.e. Plural societies (i.e.,
associated with each homogenous societies societies highly divided
type of democracy such as those divided along ethnic, religious
only along socio- and/or linguistic lines
economic or territorial ‘number of groups
lines) and their relative
Source: Lijphart (1999).

Table 1a: Models of democracy: majoritarian and consensus.

level, which the author calls the ‘executive-parties dimension’ of the insti-
tutional model (see Table 1a), majoritarian democracy is fundamentally
associated with the following characteristics: first, a majoritarian electoral
system and a political arena dominated by two major parties which take
turns in government and, second, the prevalence of executive power, nor-
mally exercised by a single party, over legislative power. Politics is here
envisaged as a zero-sum game in which the winner (in each election) takes
all (i.e. control of the fundamental decision making processes in central
government). This solution is therefore particularly well suited to homoge-
nous societies, in other words, societies with few ethnic, linguistic, reli-
gious or other divisions, and where the main dividing lines through the
electorate are socioeconomic. This is the sort of society where today’s
losers are most likely to be tomorrow’s winners, that is, where a system of
two alternating parties will work.
However, we should note that majoritarian systems at central govern-
ment level may co-exist with consensual type solutions at the level of the
state’s territorial organisation: the prime example of this is the federal
solution (as opposed to a unitary and centralised state) (see Table 1b). In
accordance with the second empirical dimension of his models of democ-
racy, which Lijphart calls the ‘federal-unitary dimension’, majoritarian
democracy tends to feature the following characteristics: unitary and cen-
tralised government (both as regards the powers conferred by the constitu-
tion, and as concerns the allocation of state revenues and expenditure), a

194 André Freire

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 195

Consensus – ‘Who
Majoritarian – ‘Who governs? As many
Empirical dimensions governs? The majority’ people as possible’
Federal-unitary Unitary and centralised Federal and decentralised
government government
Concentration of power in Strong bicameralism
a unicameral parliament Constitutional rigidity
Constitutional flexibility Judicial review (and
Absence of judicial review strong amendment
(soft amendment procedures)
procedures) Independence of the
Central bank controlled by Central Bank
the executive Switzerland, Belgium,
Paradigmatic examples: UK (until devolution in European Union (USA,
countries, etc. 1998), New Zealand Canada)
(until 1996), Barbados Idem
Type of societies usually Idem
associated with each
type of democracy
Source: Lijphart (1999).

Table 1b: Models of democracy: majoritarian and consensus.

single-chamber parliament (or a highly asymmetrical two-chamber system,

where the second chamber has every limited powers), constitutional flexi-
bility (the national executive is clearly supreme, and only a simple majority
is required for constitutional amendments) and a central bank which is
dependant on executive power.
Majoritarian democracy offers various advantages, including the possi-
bility of producing stable governments which may be more easily held to
account by the electorate. But when the floating electorate is relatively
small because the different segments are fairly rigid and faithful (to their
respective social and party political groups), alternation may be jeopar-
dised. What is more, under these conditions, minority segments are denied
access to power and are therefore tempted to bring the regime down by
force (or else to foster sentiment in favour of territorial secession). So a
stable government might actually deliver an unstable regime.
In these cases, the solution has involved introducing institutional rules
of a consensus type. In other words, as regards the ‘executive-parties
dimension’, these are systems with proportional representation (electoral
system); multi-party systems (often fragmented); coalition governments
(often oversized) and a certain balance between executive and legislative
power (see Table 1a). These are generally joined by other features relating to
the ‘federal-unitary dimension’, such as federalism, strong bicameralism
(i.e. where the two chambers are elected in fairly different ways, generally
with one representing individuals and the other the States, and with identical
powers) and constitutional rigidity (i.e. very broad majorities are needed to
alter the constitutional architecture), and so forth (see Table 1b).

Minority representation in Portuguese democracy 195

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 196

We should note that, as in the previous case of majoritarian democracy,

the executive-parties and the federal-unitary dimensions are analytically
and empirically distinct. In consensus democracy, politics is conceived
essentially as power sharing and minority rights are highly protected,
namely through a kind of right of veto. Power sharing is regarded as the
price to pay for political stability in deeply divided societies. Consensus
solutions may also be especially suited to processes of recent democratisa-
tion and/or to societies emerging from military conflicts (especially where
the conflict is rooted in ethnicity).
Using the theoretical framework developed by Arend Lijphart, my
objective here is to provide a very brief overview of the main social and
political divisions in Portuguese society, and to present the main institu-
tional features of Portuguese democracy and the possibilities they offer for
minority representation. We will start by looking at the main social and
political divisions in Portuguese society, in the first and second sections. In
the third section we will present the main institutional characteristics of
Portuguese democracy, as regards the ‘executive-parties dimension’ and
the ‘federal-unitary dimension’, and the opportunities they offer for repre-
sentation of minorities in Portugal. We will use a perspective which is both
longitudinal (for the democratic period, 1974 to the present) and compar-
ative (viewing the Portuguese experience in the context of Western
Europe). We will end with some brief conclusions.

Social and territorial divisions in Portuguese society

Portugal has a long and unbroken history as an independent state (its
frontiers have been relatively stable since the middle-ages) and, in addition
to this, it has no significant minorities, from an ethnic, religious or linguis-
tic point of view. So, unlike in Spain and various other new democracies in
Eastern Europe, studies of Portuguese political culture have revealed
strong feelings of national identity shared by a more or less homogenous
population (Cruz 1989; Reis and Dias 1993; Pinto and Núñez 1997; Freire,
Magalhães and Espírito-Santo 2003).
As we can see in Table 2, Portugal is the most ethnically concentrated
country of the thirteen Western European countries presented. In 1999,
the dominant ethnic group accounted for 99 per cent of the population.
Despite the growth in immigration since then (Pires 2002), the situation
has not significantly changed in this respect. The level of ethnic homo-
geneity in Portugal is similar to that of countries such as Denmark, the
United Kingdom or Greece, amongst others, and contrasts particularly
with the situation experienced in Europe’s ethnically divided or frag-
mented societies (France, Spain and Belgium).
In terms of religious structure, Portugal belongs to the group of pre-
dominantly Catholic countries in Western Europe (France, Austria, Italy,
Belgium, Ireland and Spain) (see Table 3). Indeed, Portugal is joint leader
with Spain of the table for religious homogeneity with the lowest level of
religious fragmentation. In addition, of the thirteen Western European

196 André Freire

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 197

Countries Dominant ethno-linguistic group (per cent)

Portugal (POR) 99
Denmark (DK) 97
United Kingdom (UK) 97
Greece (GRE) 96
Ireland (IRL) 95
Austria (AUS) 94
Italy (IT) 94
Germany (GER) 93
Netherlands (NL) 92
Sweden (SWE) 91
France (FRA) 87
Spain (SP) 80
Belgium (BEL) 59
Source: Lane and Ersson (1999: 55).
Note: The countries are listed in a decreasing order of ethnic homogeneity.

Table 2: Ethnic concentration in Western Europe, 1990.

Index of
No religious
Countries Catholic Protestant Orthodox Others Religion fragmentation
FRA 73.9 0.0 0.0 10.5 15.6 0.42
AUS 78.0 4.9 0.0 8.6 8.6 0.38
ITA 83.1 0.0 0.0 0.7 16.2 0.28
BEL 90.0 0.0 0.0 10.0 0.0 0.18
IRL 93.1 0.0 0.0 6.0 0.0 0.13
SP 94.9 0.0 0.0 5.1 0.0 0.10
POR 94.5 0.0 0.0 5.5 0.0 0.10
NL 33.0 23.0 0.0 5.0 39.0 0.68
GER 35.3 40.2 0.0 2.1 22.3 0.66
UK 13.1 72.0 0.6 4.8 9.5 0.45
DK 0.0 88.2 0.0 11.8 0.0 0.21
SWE 0.0 88.2 0.0 11.8 0.0 0.21
GRE 0.0 0.0 97.6 2.4 0.0 0.05
Source: Lane and Ersson (1999: 46).
Notes: Catholic countries; religiously mixed countries; protestant countries; Orthodox countries;
within each group, the countries are listed in a descending order of religious fragmentation.

Table 3: Confessional structure in Western Europe, 1995.

countries considered, with a variety of religious structures (predominantly

Catholic, Protestant or Orthodox; religiously mixed), the only country to
present a level of religious fragmentation lower than Portugal and Spain is
Greece (overwhelmingly Orthodox) (see Table 3).
Whilst in Portugal there is very little potential for political polarisation
along religious or ethnic lines, as the divisions are very small and society is
very homogenous in these fields, the same is not true in terms of the levels

Minority representation in Portuguese democracy 197

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 198

1 Note that the data on Never or Go to Never or Go to

the Gini Index for the
1980s is problematic rarely go church rarely go church
in various respects to church often to church often
because the reference (1990) (1990) (1999) (1999)
date and the
methodology are SWE 77.0 IRL 87.7 FRA 73.6 IRL 74.7
not the same for all DK 71.6 ITA 51.2 GB 70.8 ITA 53.6
countries, as in some GER 66.1 POR 47.5 DK 67.1 POR 53.2
cases the reference
FRA 66.1 AUS 44.1 NL 60.8 AUS 42.9
period is the 1970s.
On this point, see GB 63.9 SP 39.7 SWE 58.8 SP 36.0
Lane and Ersson NL 53.0 BEL 34.8 GER 58.1 GRE 33.6
(1999), Gunther and BEL 51.4 NL 30.5 BEL 56.0 BEL 27.8
Montero (2001) and SP 44.1 GER 26.5 SP 47.3 NL 25.0
Freire (2006b).
POR 44.0 GB 24.5 AUS 37.7 GER 24.2
AUS 37.9 GRE 22.7 POR 36.9 GB 18.7
ITA 24.5 FRA 16.9 ITA 20.9 FRA 12.3
GRE 20.5 DK 10.8 IRL 14.8 DK 11.9
IRL 6.1 SWE 10.3 GRE 13.1 SWE 9.1
Sources: World Values Survey 1999; European Values Study 1999/2000; Eurobarometer 1990
(for Greece).
Notes: The countries are listed in decreasing order (either for ‘secularisation’ or for ‘religious
integration’). GB, Great Britain.

Table 4: Levels of secularisation and religious integration in Western Europe,

1990 and 1999 (%).

of secularisation versus religious integration (see Table 4). Portugal is one

of the least secularised countries in Western Europe, together with the
other Catholic countries (Austria, Italy, Ireland and Spain) and Orthodox
Greece. The percentage of people who ‘rarely’ or ‘never’ go to church is
comparatively low, both in 1990 (44 per cent), and in 1999 (36.9 per cent),
putting the country in ninth and tenth place, respectively, out of the 13
countries in descending order of secularisation. On the other hand, the per-
centage of people who ‘often’ go to church is one of the highest for the thir-
teen European countries considered: 47.5 per cent (1990) and 53.2 per
cent (1999), putting the country in third place in descending order of reli-
gious integration. There is consequently rather more potential for political
polarisation due to the social division between a more religious majority
and a more secularised minority, given that the two groups have relatively
equal weight in society as a whole.
Of the 13 Western European countries presented in Table 5, Portugal
has the highest level of socioeconomic inequality, although in the 1980s
it shared this position with Spain.1 These findings persist in the 1990s,
regardless of whether we use the S80/S20 or the Gini Index as the indicator
for 1996 and 1999, respectively.
Portugal remains an extremely divided country in terms of the distrib-
ution of socioeconomic resources and the system of material rewards. In
other words, the divisions between a minority that controls most of the
socioeconomic resources, and a majority that has fairly limited access to
these resources, are the deepest of the 13 Western European countries in

198 André Freire

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 199

Gini Index (1980s) S80/S20 (1996) Gini Index (1999)

SP 0.39 POR 6.6 POR 0.36
POR 0.39 GRE 6.1 GRE 0.34
DK 0.38 ITA 6.0 SP 0.33
AUS 0.37 SP 5.9 UK 0.32
IRL 0.33 UK 5.6 IRL 0.32
ITA 0.31 IRL 5.6 ITA 0.30
FRA 0.30 GER 4.7 BEL 0.29
UK 0.30 AUS 4.7 FRA 0.29
GRE 0.30 FRA 4.5 AUS 0.26
NL 0.27 NL 4.5 NL 0.26
GER 0.25 SWE 3.7 GER 0.25
BEL 0.24 DK 2.9 DK 0.23
SWE 0.22 BEL - SWE 0.23
Sources: Gini Index, 1980s (Lane and Ersson 1999: 69 and 70) (except for Austria and Ireland
(income of the richest 20 per cent—the data is from the 1960s—and Denmark (Gini Index)—the
data is from the 1970s. All other country data is from the 1980s; Gunther and Montero (2001)—
Portugal, mid-1970s, and Greece, 1982; S80/S20 (1996) (Matsaganis et al. 2003); Gini Index
(1999) (Dennis and Guio 2003: 6).
Notes: Gini index: 0 (maximum equality) and 1 (maximum inequality); S80/S20—the ratio
represents the income received by the richest 20 per cent vis-à-vis the poorest 20 per cent; The
countries are listed in a decreasing order of social inequalities.

Table 5: Social inequalities in Western Europe, 1980–1999.

analysis. More recent data, from the United Nations Development Program,
show Portugal to be the most unequal country in the EU-25 (UNDP 2005).
This situation would appear to be related not only to the actual level of
development in the country (other fairly unequal countries are Spain and
Greece), but also to the population’s actual tolerance vis-à-vis inequality
(Cabral, Vala and Freire 2003) and, above all, to the ideological orienta-
tion, and consequent public policies, of the governing class (see in this
respect the position of the United Kingdom).
Socioeconomic inequalities in Portugal are not only fairly significant;
they also correspond to a fairly precise geographical demarcation. On the
one hand, there are areas of the country (Lisbon and the Tagus Valley,
Madeira and the Algarve) with greater control of socioeconomic resources
and above average earnings levels, whether measured in terms of
per capita GDP or in terms of per capita household disposable income
(see Table 6). On the other hand, with less control over socioeconomic
resources there are other areas of the country (north, centre, Alentejo and
the Azores), where earning levels are below average. Regions in the first
group are – as a rule – more densely populated, although certain coastal
areas of the north and centre also have a high population density (Freire
2001). Indeed, if the statistical data at our disposal allowed us to break
down the northern and central regions into coastal and interior sub-
regions, we would have an even clearer picture of the geographical pattern
of socioeconomic inequality: the interior outlying areas have much more
limited control over socioeconomic resources than the coastal areas,

Minority representation in Portuguese democracy 199

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 200

2 The PS has always GDP per capita 2001 DHI per capita 2001
been a member of the
Socialist International NUTS (Thousand euros)
(Sablosky 1997: PORTUGAL 11.9 8.0
North 9.6 6.7
3 Until the 1990s, the Centre 9.7 7.2
PSD had been Lisbon and Tagus Valley 15.8 9.8
associated with the
Alentejo 9.6 7.0
European Liberal
Democratic and Algarve 12.4 8.6
Reformist Group Azores 9.4 6.9
(ELDR) in the Madeira 13.4 8.5
European Parliament.
Since the beginning of Source: Instituto Nacional de Estatística, Contas regionais (www.ine.pt).
the 1990s, however, it
has aligned itself with Table 6: Territorial inequalities in Portugal: GDP per capita and disposable
the conservative household income (DHI) per capita by NUTS II, 2001.
European People’s
Party (EPP) (Frain which are more central (from a social and political perspective, although
1997: 80ff.)
not necessarily from a geographical perspective) (Freire 2001). Finally,
4 Founded in 1921, the there is also a contrast between the two most remote Atlantic island
PCP was a member of
the Comintern until regions: Madeira and the Azores. Madeira has above average levels of
the collapse of this income, especially in terms of per capita GDP, whilst the Azores is the
organisation (Cunha
1997: 37). In the region trailing farthest behind the national average, in terms of per capita
European Parliament, GDP, and the second farthest behind the national average in terms of per
the PCP is a member
of the United capita disposable household income.
European Left/Nordic
Green Left (UEL/NGL)
parliamentary group.
Issue dimensions of partisan conflict in the Portuguese
party system
Prior to the relatively bloodless Carnation Revolution of 25 April 1974 that
initiated the so-called ‘third wave’ of worldwide democratisation, free and
fair elections with universal suffrage and a competitive party system were
unheard of in Portugal. Portugal’s transition was initiated by a coup led by
junior officers (Freire 2005, 2006a). Whilst the coup may have been
planned as a political revolution to liberalise society – overthrow a decrepit
regime and end the interminable colonial wars – it is important to note that
the military remained committed to holding constituent elections one year
from the date of coup. These elections were held on schedule on 25 April
1975, and obtained a 92 per cent turnout. One year later, on 25 April
1976, the first constitutional parliamentary elections took place.
A stable party system quickly emerged, and by 1976 four parties rep-
resented almost 90 per cent of the electorate. Apart from a brief period
during the mid-1980s when the centre-left Partido Renovador Democrático
(PRD) emerged and disappeared, the party system has remained relatively
stable. The general tendency is for the vote to concentrate with the two
centrist ‘catch-all’ parties: the centre-left Partido Socialista (Socialist Party –
PS),2 and the centre-right Partido Social Democrata (Social Democratic
Party – PSD), which, despite its name is a liberal party and not social
democratic.3 Alongside the PS and the PSD, the Partido Comunista
Português (Portuguese Communist Party – PCP)4 and the conservative
Centro Democrático Social (Social Democratic Centre – CDS) have become

200 André Freire

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 201

the system’s main parties. Following its defeat in the 1991 legislative elec- 5 The CDS was founded
tions, the CDS changed its leadership, its ideological profile and its name, as a Christian
democratic party.
becoming the Partido Popular (Popular Party – PP).5 Some smaller parties Following accession
have obtained seats in parliament during the democratic period (Freire to the EU it joined the
EPP. In the early
2005, 2006a). Among these parties it is worth mentioning the Bloco do 1990s it began
Esquerda (Left Bloc – BE). This left-libertarian organisation was originally promoting an anti-EU
stance, leading to its
a coalition of two old extreme-left wing parties and a political movement expulsion from the
that was formed to compete in the 1999 legislative elections. Over the past EPP in 1992.
Following this, it
few years, however, it has come to be viewed as a single political party.6 joined the Union for
When competing for voters’ support, parties present different packages Europe of the Nations
of public policies, each with different levels of priority. Both the packages of Group (UPE). After
1997, the party’s
public policies and their relative priority are related to the issue dimen- stance on the EU
sions of partisan conflict. Lijphart emphasises the need to distinguish changed, culminating
with their return to
between the dimensions of policy competition, and ‘the characteristics of the EPP in July 2004.
the voters that parties represent’. In this respect, it is important to recall 6 The BE elected its first
the difference between ‘domain of identification’ and ‘space of competi- MEP at the 2004
European Elections.
tion’ that were introduced by Sani and Sartori (1983: 330). The former In the European
refers to which electors identify with the different parties, and which Parliament, the BE
dimensions of identification (ideological, religious, ethnic, linguistic, terri- (like PCP) is an
associated member
torial, etc.) are relevant in each case, while ‘space of competition’ ulti- of the UEL/NGL
mately addresses the query, along which dimensions lay the non-identified parliamentary group.
partisan or floating voters for which it is rewarding to compete (Sani and
Sartori 1983: 330). The two dimensions are complementary, but what it
does mean is that electors are usually distributed along multiple dimensions
of identification; however, this does not necessarily mean that political
parties compete along the same dimensions. Moreover, in spite of multiple
dimensions of identification, the space of competition can be one-
dimensional. Lijphart (1999, 1989) defines seven issue dimensions of policy
competition. Additionally, for each country and epoch, he classifies each of
them according to their importance for policy competition. In Table 7,
Lijphart’s analysis of the dimensions of policy competition in the Portuguese
case is presented for the periods 1975–86 and 1975–96. Updated data for
the period 1996–2004 has been added from our own analyses.

Issue dimensions and their salience

Socio- Cultural Urban- Regime Foreign Post- Number of
Years economic Religious Ethnic Rural Support Policy materialism dimensions
1975–86 H H – – M H – 3.5
1975–96 H M – – M M – 2.5
1996–2004 H M – – M M M 3.0
Source: Adapted from Lijphart (1989: 279) for the period 1975–1986 and (1999: 80ff) for the period 1975–1996. Information
for the period 1996–2004 represents the present author’s evaluation.
Notes: H ⫽ issue dimension with high salience (counts as 1 for the number of dimensions); M ⫽ issue dimension with medium
salience (counts as 0.5 for the number of dimensions).

Table 7: Issue dimensions of partisan conflict in the Portuguese party system, 1975–2004.

Minority representation in Portuguese democracy 201

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 202

In the Portuguese case, the cultural-ethnic dimension is not relevant

for either policy competition or as a domain of identification. The same
can also be said about the urban-rural dimension (Pinto and Núñez 1997;
André and Gaspar 1989; Freire 2001).
For the period 1975–1996, the post-materialist issue dimension was
irrelevant both as a domain of competition and of identification. For the
period for which appropriate survey data is available (1990–2002), it
can be seen that the electorate has very little support for post-materialist
values, and that, in general, there is practically no difference between
the supporters of each of the parties (Freire 2003; Jalali 2004). Until the
end of the 1990s, parties had hardly competed on this issue dimension.
With the emergence of the BE as a parliamentary force, however, new
political issues have became a domain of competition between the left (par-
ticularly the BE, but also the PCP and PS) and the right (PSD and particu-
larly the PP). From 1996 until at least 2004, post-materialism has been a
pertinent dimension of policy competition, although only with medium-
level significance.
During the early phase of Portugal’s transition to democracy, from
25 April 1974 to 25 November 1975, regime support was a highly con-
tentious issue that placed the PCP and several other extreme-left parties in
opposition to the pro-liberal democratic parties (Freire 2005, 2006a). The
PCP advocated a Soviet-style popular democracy, while the extreme-left
parties defended Third World Communist models. The PS, PSD and CDS,
on the other hand, advocated following the Western democratic model. On
25 November 1975, a counter-coup by moderate elements within the
MFA, who had foiled a coup attempt by the extreme-left, established a
durable liberal democracy. Since then, the PCP has normalised its relation-
ship with parliamentary democracy (Cunha 1997), and the issue has lost
most of its previous significance. In any event, this matter has very little
importance for our present analysis of the period 1996–2004, with the
only relevant point concerning the PCP’s and BE’s reservations regarding
the capitalist system, which is a model of society accepted, to varying
degrees, by the other three parliamentary parties.
In terms of foreign policy, the major issues of competition have been
concerned with the alignment of political parties in terms of the two Cold
War political and military blocs and European integration. With respect
to the former, the democratic pro-liberal PS, PSD and CDS supported the
West and its military organisations, while the PCP sympathised with the
Soviet bloc and its military organisation. As we have seen in respect of
regime support, this policy dimension of competition cuts across the left-
right divide. With the fall of the Berlin Wall and the end of the Cold War,
this divide lost most of its significance, although its continued presence
remains apparent in relation to certain international issues, such as the
NATO bombing of Serbia in 1999 and the 2003 Gulf War. These divisions
sometimes have the power to force ideological alignment that reinforces
the left-right political divide. One example of this can be seen in the

202 André Freire

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 203

political response to the 2003 Gulf War, which was opposed by all of the
left-wing parties and supported by the right-wing parties.
European integration is an issue that cuts across the ideological divide,
albeit in a rather less than straightforward manner. During the transition
to democracy, the PCP and other left-wing groups proposed alternative
socialist and Third World paths. This explains why European integration
was to become a major policy goal of the PS, PSD and CDS. From the mid-
1970s until 1992, political support for Europe was monopolised by these
largely pro-European parties, with the result that, from 1988 onwards
(the year of the first direct elections to the European Parliament), the PCP
was forced to significantly moderate its resistance to Europe (Cunha 1993).
Following its resounding defeat at the 1991 elections, however, the newly
renamed PP followed its new leader in adopting a much more sceptical
position towards the European Union and its proposals for a single European
currency. This change in direction was short-lived; however, following the
election of a new leader in 1997, the party accepted the inevitability of the
new currency. With the PP’s subsequent rise to power as part of the PSD-
PP coalition that formed the government in 2002, the party has assumed
a more prudent position. The position of the PP notwithstanding, it is a
fact that there is very little to separate the PSD and the PS on European
issues (Freire 2005, 2006a). One new element of left-right division over
European matters came to light in the wake of the European parliamen-
tary elections of 2004 when the opposition left-wing parties rejected the
EU stability pact that was defended by the governing right-wing parties.
The issues that provide the best overlaps in the left-right divide in
Portugal are, first, socioeconomic matters, and, second, religious affairs.
Whether as a domain of competition or of identification, both issues enable
us to split the parties into left- and right-wing, and to further order them
in a left-right continuum that ranges from the PCP on the left, through the
BE, PS and PSD to the PP on the right (Freire 2005, 2006a). In terms of
the domain of competition, the socioeconomic dimension (i.e., controver-
sies concerning socioeconomic equality and the role of the state in the
economy and society) is the most significant, with the religious dimension
having only medium significance. During the democratic transition, the
Catholic Church aligned itself with the pro-liberal democratic parties
against the radical left. During that period, religious polarisation was high.
Since then the religious dimension has barely registered as a domain of
policy competition except when policies concerning moral issues and/or
the Church’s interests are debated. This has been the case with proposals to
liberalise abortion legislation (which is supported by the left), or the proposal
to provide state finance for the Catholic University (which is supported by
the right) (Freire 2005, 2006a). As a dimension of identification, however,
the religious issue has always proved more significant than the socioeco-
nomic issue, with some studies of Portuguese electoral behaviour reveal-
ing that church attendance is a better vote predictor than social class
(Freire 2005, 2006a). Post-materialist issues are more pertinent to the

Minority representation in Portuguese democracy 203

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 204

7 Madeira has six competition domain than they are to that of identification. They more or
deputies in the Lisbon less permit us to range the parties from left to right in terms of their policy
parliament, while the
Azores have five. proposals; however, they are a very poor predictor of voting alignments
8 Portugal, Spain, or of the individual citizen’s position on the political spectrum (Freire
Greece, France, 2005, 2006a).
United Kingdom,
Germany, Austria, On the one hand there is a very low potential for partisan conflict based
Italy, Holland, on ethnic and linguistic issues in the Portuguese party system, and indeed
Denmark, Belgium,
Sweden and Ireland.
these are not relevant dimensions of partisan conflict: on the other hand
The 1989 figures are social inequalities with a territorial base are fairly deep. Moreover, Portugal
based on the has two ultra-peripheral regions (the Azores and of Madeira – although
Eurobarometer 31A
survey data for all Madeira now has a level of individual and familial per capita incomes that
countries except are above the national average). However, even the issues related to these
Austria and Sweden,
where data from the latter divisions (i.e. social inequalities with a territorial base) are not rele-
‘Party Manifestos’ vant issues of partisan conflict at least at the national level. On the one
was used. The figures
for all countries in hand, this might be due to the fact that regional parties are forbidden by the
1999 are based on Portuguese constitution. On the other hand, although Portugal has a cen-
data from the
European Election
tralised system of government, there is a regionalised system of government
Study of that year (see for the Azores and of Madeira, with a regional parliament and government
http://www.european in each. Thus, these minorities – at least from the standpoint of the number
EES%201999.htm). of inhabitants and the distance of the regions from the mainland – can
9 Only in Ireland we express their interests through the regional system of government, whilst
considered the two also being represented within the national parliament in Lisbon.7
major parties tout
court (Fianna Fail and
Leaving aside those issues that normally cut across the left-right divide,
Fine Gael) due to the we are left with the socioeconomic, the religious and the post-materialist
fact that these are the issue dimensions. What, however, is the strength of the left-right divide in
parties that usually
lead government’s Portugal in a comparative perspective in terms of the policy competition
alternation. domain? By using the electorate’s perception of the position of political
parties on the political spectrum in 13 countries in 1989 and 1999,8 the
following became apparent: in terms of standardised ideological distances
(i.e. the absolute distance between parties on the political spectrum,
divided by the maximum distance) between the two most extreme parties
represented in their respective parliament, France, Portugal and Greece
had the most polarised political systems both in 1989 and 1999. In both
Portugal and Greece this is due mainly to the presence of orthodox
Communist parties. When considering the standardised ideological dis-
tance between the two major parties (one from the left ideological bloc,
and the other from the right9) in these countries, the opposite conclusions
can be drawn for the Portuguese case. In 1989, Portugal, Austria and
Ireland were the least polarised systems, while in 1999 Portugal, Belgium,
the United Kingdom, Austria and Ireland were the least polarised. Using
data from surveys conducted during 1989 and 1993, the conclusions arrived
at were very similar (Freire 2005, 2006a).

Political institutions and the representation of minorities

On the basis of data gathered and processed by Arend Lijphart (1999) and
presented by Bruneau et al. (2001), Table 8 presents the relative position

204 André Freire

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 205

The executive-parties’ dimension The federal-unitary dimension

(high, majoritarian; low, consensus) (high, unitary; low, federal)
New Zealand 1.55 UK 1.50
Canada 1.45 Iceland 1.45
UK 1.45 Luxembourg 1.22
Australia 1.41 Portugal (1976–95) 1.00
USA 0.78 Greece (1974-96) 0.90
Spain (1977–96) 0.72 France, 4th Rep. 0.78
Greece (1974–96) 0.66 Sweden 0.54
Austria 0.53 Finland 0.47
Ireland 0.34 France, 5th Rep. 0.45
Luxembourg 0.29 Belgium 0.40
France, 5th Rep. 0.28 Ireland 0.34
Japan 0.20 Norway 0.34
Iceland –0.04 Denmark 0.23
Portugal (1976–95) –0.14 Austria –0.10
Germany –0.15 Spain (1977–96) –0.23
Norway –0.20 Italy (1946–96) –0.35
Sweden –0.25 Netherlands –0.38
Denmark –0.76 Canada –0.97
Belgium –0.92 Australia –1.33
Netherlands –1.06 Switzerland –1.39
Italy (1946–96) –1.16 Germany –1.50
Finland –1.42 Japan –1.52
Switzerland –1.59 USA –1.93
France, 4 Rep. –1.80 – –
Source: Bruneau et al. (2001).

Table 8: The executive-parties’ and the federal-unitary dimensions: the

Portuguese case in comparative perspective, 1945–1996.

of 23 countries and 24 political systems (France counts twice: the 4th and
5th Republics) in terms of the profile of the respective political systems in
the ‘executive-parties’ dimension and in the ‘federal-unitary’ dimension.
There is no need here to go into methodological details, suffice to say that
for each dimension the authors constructed a compound index that aggre-
gates the information for the various items characterising each of the two
dimensions (see Tables 1a and 1b above). Both as regards the ‘executive-
parties’ dimension and the ‘federal-unitary dimension’, the highest scores
refer to the majoritarian and unitary characteristics, respectively. At the
other end of the scale, the lowest scores point to profiles characteristic of
consensus and federal democracies, respectively.
Before continuing with an analysis of the data in Table 8, we should
point out the d’Hondt highest average system of proportional representa-
tion is used in Portuguese parliamentary elections. Votes are converted
into parliamentary seats by party with a view to ensuring a balanced
match between the percentage of votes and the percentage of seats
obtained by each party. Votes are counted for each parliament in each of
the 22 multi-member constituencies (18 corresponding to the 18 mainland
districts, one each for the Azores and Madeira, and two for expatriates,

Minority representation in Portuguese democracy 205

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 206

divided into ‘Europe’ and ‘rest of the world’). There has been only one
major change to the Portuguese electoral system since it was established
in 1975 – the reduction in the number of deputies from 250 to 230 as
part of the 1989 constitutional review, which took effect from the 1991
parliamentary election and which had a slight impact on the average size
of constituencies (reduced from 11.4 to 10.5). Under the electoral law, the
number of members per constituency is proportional to the number of reg-
istered voters.
In terms of the ‘executive-parties’ dimension, the average figures shown
in Table 8 for the entire democratic period under consideration (1976–95)
show that Portugal occupied a middle-of-the-road position amongst the
24 political systems, rather closer to the ‘consensus model’ of democracy
than to the ‘majoritarian model’. In other words, Portugal’s institutional
arrangements significantly facilitate the representation of minorities. This
means that, although Portuguese society is relatively homogenous (except
in terms of religious integration, socioeconomic inequalities and political
divisions), the design of the political institutions with regard to the ‘executive-
parties’ dimension is substantially favourable to the expression and repre-
sentation of minority identities and interests.
However, we know that during the democratic period (1974 to the
present), the Portuguese political system has presented differing charac-
teristics in terms of the ‘executive-parties’ dimension, depending on the
period in question – before or after 1987 (Lopes and Freire 2002; Lobo
2001, 2003; Magalhães 2003; Freire 2005, 2006a).
The changes in the Portuguese party system can be divided into three
distinct phases (Freire 2005, 2006a). The first of these was the period
from 1976 to 1987, which is characterised by a fragmented multiparty
system with highly unstable cabinets. During this phase, the role of each
of the different major political institutions (president, government and
parliament) was more balanced. The second period, from 1987 to 2002,
was one in which a strong bipartisan trend within the party system was
evident. This trend impelled change towards single party and increasingly
stable governments with power being concentrated with the prime minis-
ter. Between 2002 and 2005 the system appeared to have entered a
third phase, one in which the concentration of the vote in the two major
parties persisted, although not sufficiently strong to obviate the need to
form coalitions. We now know that the period between 2002 and 2005
represented only a brief interregnum in the majoritarian trend in the
Portuguese party system. The PS won the 2005 legislative elections with a
quasi-majority of votes (around 45 per cent) and an absolute majority of
seats that created the conditions for the country to return to stable single
party government. However, as can be seen in Figure 1, and as has been
argued elsewhere (Freire 2006a), the majoritarian trend in the political
and party systems is not due to any major change in the institutional
format of the regime, namely the electoral system, but mainly to a con-
centration of the vote in the two major parties (see Freire 2005, 2006a).

206 André Freire

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 207

Note: The AD coalition was considered as a single organisation in calculating both the ENEP and the
‘Vote percentage of largest party divided by 10’ for the years 1979 and 1980. In these years, the AD
is considered the largest party.
Sources: ENEP, and the ‘Vote percentage of largest party divided by 10’ calculated by the author
from official electoral data (www.cne.pt); Effective threshold (Lijphart) (Magalhães 2003: 189);
Disproportionality (Gallagher’s Least Squares Index) (Magalhães 2003: 189).

Figure 1: Electoral system properties and party system change, 1975–2002.

Whilst on average Portuguese democracy is closer to the consensus

model in terms of the executive-parties dimension, when it comes to the
federal-unitary dimension, it has always been closer to the majoritarian
model (see Table 8). Indeed, of all the political systems under analysis,
Portugal has the fourth most majoritarian and centralised system,
exceeded only by the United Kingdom, Iceland and Luxembourg. What is
more, in this dimension evolution over the entire democratic period points
to great stability in this respect (Bruneau et al. 2001). In other words, the
highly unitary and centralised character of the Portuguese political system
has been a stable and lasting trait over approximately thirty years of
democracy. This fact is all the more curious when we consider that the
greatest divisions in Portuguese society have to do with socioeconomic
inequalities and that these are tied, to a fairly significant extent, to clearly
defined territorial factors.
However, we should note that Portugal has a devolved system of gov-
ernment for each of the two island regions: the Azores and Madeira. This
system consists of a parliament and government, the former being elected
and the second chosen in the light of the outcome of these elections (Morais,
Araújo and Freire 2003). Each of these two regions has fairly wide-
ranging legislative and administrative powers. We should note that the
500,000 inhabitants of these regions represent only around five per cent
of Portugal’s total population, meaning that these devolved and decen-
tralised sub-systems have no real bearing on the fundamentally unitary
and centralist character of the Portuguese state.
Madeira consists of three islands, of which only two are inhabited. At
a distance of approximately 978 km from Lisbon it has a population of
253,482 (INE 2001). The region has a parliament with 68 deputies

Minority representation in Portuguese democracy 207

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 208

10 Although in 1998 a elected by proportional representation (for the next parliament, the
referendum was held number of deputies will be cut to 47, all elected for a single constituency,
to find out whether
the Portuguese instead of various constituencies as previously).
wanted to adopt a The Azores are located at around 2000 km from the Iberian Peninsula
regionalised system of
government for the and have a population of 241,763 (INE 2001). The region consists of nine
country as a whole. islands, each of which corresponds to a constituency (an extra compen-
The proposal from the
Socialist government satory constituency will be added at the next elections, as a result of the
was also supported by recent reform of the electoral system). The region’s parliament has a total
the Communist Party.
The right wing parties
of 52 deputies, elected by proportional representation.
(PSD and CDS-PP)
opposed the plans. Brief conclusions
The proposal was
rejected by the Portugal is a fundamentally homogenous country in terms of ethnicity
electorate, with and language, and also as regards religious faith. In other words, the
63.5 per cent against
regionalisation, and country has no significant minorities in these areas. However, there are
only 36.5 per cent large socioeconomic inequalities, which have remained largely unchanged
in favour (Freire and
Baum 2003). since at least the 1980s. What is more, these inequalities quite clearly fit a
geographical pattern. In addition to socioeconomic issues, and above all
those of inequality, the level of religious integration also has significant
potential for political polarisation. Curiously, however, these territorially
based social inequalities are not a great source of political controversy.10
This may be because the country’s institutional system (unitary and highly
centralised state), which includes no regionalisation or a parliamentary
second chamber to represent the regions, provides no channel for the
expression of any (potential) claims. Another reason may be the constitution
prohibition of regionalist political parties. In addition to these social sources
of political polarisation, there are also the political divisions as such, which
provide the framework for conflict within the party political system.
In terms of the executive-parties dimension, Portuguese democracy is
basically close to the consensus model. In other words, it provides fairly
favourable conditions for the expression of minority identities and interests.
However, from 1987 onwards there has been a shift towards majoritarian
democracy, but this was due more to changing voter behaviour than to
any real alteration in the design of political institutions, namely in the
electoral system.
Despite the existence of devolved government for the Azores and Madeira
regions, Portugal has always (since the transition to democracy) had a
political system closer to the majoritarian model with regard to the federal-
unitary dimension. Indeed, from a comparative standpoint, the political
system is extremely centralised and unitary, not least because the
Autonomous Regions of the Azores and Madeira account for no more than
around five per cent of the country’s population. In other words, if we
exclude these regions, from the point of view of expression of the interests
and identities of the population’s resident in outlying regions with limited
control over economic resources, the Portuguese political system does not
facilitate representation of territorially based minorities.

208 André Freire

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 209

The existence of a more decentralised and regionally based political

system would help to give a political voice to the population of more outlying
areas of mainland Portugal with less control over socioeconomic resources,
and could therefore serve as an effective weapon in combating social
inequalities, especially those tied to territorial factors. It is highly likely
that Portugal’s political centralism has been partly to blame for the persis-
tence of significant social inequalities from the 1980s to the present day.
Madeira’s current position in the breakdown of national earnings, as we
saw above, shows that more political decentralisation can be an effective
form of reducing inequalities, as this region started out from a position
significantly behind that of mainland Portugal. However, it is also true
that the Azores show that regionalisation is not enough in itself for
achieving this goal.

André, I. and Gaspar, J. (1989), ‘Portugal, geografia eleitoral: 1975 e 1987’, in
Coelho, M.B. (ed.), Portugal: O sistema político e constitucional 1974/1987,
Lisbon: Universidade de Lisboa, pp. 257–7.
Beethem, D. (2005), Democracy: A beginner’s guide, Oxford: Oneworld.
Bruneau, T.C., Diamandouros, P.N., Gunther, R., Lijphart, A., Morlino, L. and
Brooks, R. (2001), ‘Democracy, southern European style’, in Diamandouros, P. N.
and Gunther, R. (eds), Parties, politics and democracy in new southern Europe,
Baltimore, MD: The Johns Hopkins University Press, pp. 16–82.
Cabral, M.V., Vala, J. and Freire, A. (eds) (2003), Desigualdades sociais e percepções de
justiça, Lisboa: Imprensa de Ciências Sociais.
Cruz, M.B. (1989), ‘Nacionalismo e patriotismo na sociedade portuguesa actual’,
Nação e Defesa 49: 3–22.
Cunha, C. (1997), ‘The Portuguese Communist Party’, in Bruneau, T.C. (ed.),
Political parties and democracy in Portugal: Organizations, elections and public
opinion, Boulder, CO: Westview Press, pp. 23–54.
——— (1993), ‘L’opposition du parti communiste portugais à l’adhésion à la CEE’,
in Delwit, P. and De Waele, J.-M. (eds), La gauche face aux mutations en Europe,
Brussels: Université de Bruxelles, pp. 119–32.
Dahl, R. (1998), On democracy, New Haven, CT: Yale Universtity Press.
Dennis, I. and Guio, A.-C. (2003), ‘Poverty and social exclusion in the EU after
Laeken: Part 1’, Statistics in focus: Population and social conditions: Theme 3:
8/2003, Eurostat, pp. 1–7.
Frain, M. (1997), ‘The right in Portugal: The PSD and the CDS/PP’, in Bruneau,
T.C. (ed.), Political parties and democracy in Portugal: Organizations, elections and
public opinion, Boulder, CO: Westview Press, pp. 77–111.
Freire, A. (2001), Mudança eleitoral em Portugal: Clivagens, economia e voto em eleições
legislativas, 1983–1999, Oeiras: Celta.
——— (2003), ‘Pós-materialismo e comportamentos políticos: O caso português
em perspectiva comparada’, in Vala, J., Cabral, M.V. and Ramos, A. (eds),
Valores sociais: Mudanças e contrastes em Portugal e na Europa, Lisbon: Imprensa
de Ciências Sociais, pp. 295–362.
——— (2005), ‘Party system change in Portugal, 1974–2005: The role of social,
political and ideological factors’, Portuguese Journal of Social Science 4(2): 21–40.

Minority representation in Portuguese democracy 209

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 210

——— (2006a), ‘The party system of Portugal’, in Niedermayer, O., Stöss, R. and
Haas, M. (eds), Die Parteiensysteme Westeuropas, Wiesbaden: VS Verlag für
——— (2006b), Esquerda e direita na política europeia: Portugal, Espanha e Grécia em
perspectiva comparada, Lisbon: Imprensa de Ciências Sociais.
Freire, A. and Baum, M. (2003), ‘Referenda voting in Portugal, 1998: The effects
of party sympathies, social structure and pressure groups’, European Journal of
Political Research, 42(1): 135–61.
Freire, A., Magalhães, P. and Espirito-Santo, A. (2003), ‘Thinner than thin?
Political culture and political action in Portugal’, IOC Discussion Papers 18,
Institute of Oriental Culture, University of Tokyo.
Gunther, R. and Montero, J.R. (2001), ‘The anchors of partisanship: A compara-
tive analysis of voting behaviour in four southern European countries’, in
Diamandouros, N.P. and Gunther, R. (eds), Parties, politics and democracy in new
southern Europe, Baltimore, CT: The Johns Hopkins University Press, pp. 83–152.
Held, D. (2005), Models of democracy, London: Polity Press.
Horowitz, D. (1985), Ethnic groups in conflict, Berkeley, CA: University of California
INE (National Institute for Statistics) (2001), Population census 2001, Lisbon: INE.
Jalali, V.C. (2004), ‘As mesmas clivagens de sempre? Velhas clivagens e novos
valores no comportamento eleitoral português’, in Freire, A., Lobo, M.C. and
Magalhães, P.C. (eds), Portugal a votos: As eleições legislativas de 2002, Lisbon:
Imprensa de Ciências Sociais, pp. 87–124.
Lane, J.-E. and Ersson, S. (1999), Politics and society in Western Europe, London: Sage.
Lijphart, A. (1977), Democracy in plural societies, New Haven, CT: Yale University Press.
——— (1989), As democracias contemporâneas, Lisbon: Gradiva.
——— (1999), Patterns of democracy: Government forms and performance in 36 coun-
tries, New Haven, CT: Yale University Press.
Lobo, M.C. (2001), ‘The role of political parties in Portuguese democratic consoli-
dation’, Party Politics 7: 643–53.
——— (2003), ‘El incremento del poder del primer ministro en Portugal desde
1976’, in Barreto, A., Gomez, B. and Magalhães, P.C. (eds), Portugal: Democracia
y sistema político, Madrid: Siglo XXI, pp. 175–204.
Lopes, F.F. and Freire, A. (2002), Partidos políticos e sistemas eleitorais: uma intro-
dução, Oeiras: Celta.
Magalhães, P.C. (2003), ‘Elections, parties and policy-making institutions in demo-
cratic Portugal’, in Pinto, A. C. (ed.), Contemporary Portugal: Politics, society and
culture, New York, NY: Social Science Monographs, pp. 183–202.
Morais, C.B., Araújo, A. de and Freire, A. (2004), Entre a representação desigual e a
derrota dos vencedores: Estudo sobre a reforma do sistema eleitoral dos Açores,
Lisbon: Imprensa de Ciências Sociais.
Matsaganis, M., Ferrera, M., Capucha, L. and Moreno, L. (2003), ‘Mending nets in
the south: Anti-poverty policies in Greece, Italy, Portugal and Spain’, Social
Policy and Administration, 37(6): 639–55.
O’Flynn, I. and Russell, D. (2005), Power sharing: New challenges for divided societies,
London: Pluto Press.
Pinto, A.C. and Núñez, X. M. (1997), ‘Portugal and Spain’, in Eatwell, R. (ed.), Euro-
pean political cultures: Conflict or convergence?, London: Routledge, pp. 172–92.
Pires, R.P. (2002), ‘Mudanças na imigração: Uma análise das estatísticas sobre a
população estrangeira em Portugal, 1998–2001’, Sociologia Problemas & Práti-
cas, 39: 151–66.

210 André Freire

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 211

UNDP (United Nations Development Program), Human Development Report 2005,

in http://hdr.undp.org/reports/global/2005/
Reilly, B. (2001), Democracy in divided societies: Electoral engineering for conflict man-
agement, Cambridge: Cambridge University Press.
Reis, L.B. and Dias, M. (1993), ‘Grupos de referência socio-politicos’, in França, L.
(ed.), Portugal: Valores europeus, identidade cultural, Lisbon: IED.
Reynolds, A. (1999), Electoral systems and democratization in South Africa, Oxford:
Oxford University Press.
Sablosky, J.A. (1997), ‘The Portuguese Socialist Party’, in Bruneau T.C. (ed.),
Political parties and democracy in Portugal: Organizations, elections and public
opinion, Boulder, CO: Westview Press, pp. 55–76.
Sani, G. and Sartori, G. (1983), ‘Polarization, fragmentation and competition in
Western democracies’, in Daalder, H. and Mair, P. (eds), Western European party
systems: Continuity and change, London: Sage, pp. 307–40.

Suggested citation
Freire, A. (2007), ‘Minority representation in Portuguese democracy’, Portuguese
Journal of Social Science 6 (3): 193–211, doi: 10.1386/pjss.6.3.193/1

Contributor details
André Freire is an assistant professor at ISCTE in Lisbon and is also a senior
researcher at CIES-ISCTE. Contact: André Freire, ISCTE, Avenida das Forças
Armadas, 1649-026 Lisboa, Portugal.
E-mail: andre.freire@iscte.pt

Minority representation in Portuguese democracy 211

PJSS-6-3-05-Freire 3/14/08 2:36 AM Page 212
PJSS-6-3-06-Castillo 3/14/08 2:37 AM Page 213

Portuguese Journal of Social Science Volume 6 Number 3. © Intellect Ltd 2007.
Review. English language. doi: 10.1386/pjss.6.3.213/5

Who governs southern Europe? Regime change and

ministerial recruitment, 1850–2000, Almeida, P.T. de,
Pinto, A.C. and Bermeo, N. (2003)
London and Portland, OR: Frank Cass, 242 pp.,
ISBN 0714682772 (pbk), £75
Francisco Javier Luque Castillo, University of Grenada, Faculty of Political
Science and Sociology

The interest of political scientists in research elites declined at the begin-

ning of the 1960s, whether as a consequence of the criticism that had for
decades seen them labelled as ‘elitist’ for serving (allegedly) as a back-door
justification for a determined conception of democracy, or whether for
being considered analytically insufficient (by their static nature). This dis-
interest began declining from the beginning of the following decade when,
for the first time authors such as Putnam and Dogan raised questions
about the role of the bureaucratic elites in the political process (Putnam
1973; Dogan 1975). Their indignation continued during the following
years (Aberbach, Putnam and Rockman 1981), culminating with a resur-
gence of these studies in the intellectual field at the end of the 1980s.
Since then, the waves of democratisation in southern and eastern Europe
and Latin America have proved attractive to political scientists interested
in the theme (e.g. Higley and Gunther 1992; Eyal, Szelényi and Townsley
1998). Simultaneously, the investigation of members of Western European
governments was consolidated (Blondel and Thiébault 1991). The book is
the result of two lines of research, with its objective being the study of min-
isterial elites in the most southern countries of Europe from the beginning
of the institutionalisation of the liberal state to the present day.
Southern European ministers have been the subject of a previous study,
albeit over a more restricted chronological period and with the nation-
state as the geographical reference space. Works by authors such as Lewis
(1978), Jerez (1982), Dogan (1975) and Koutsokiz (1982) have already
anticipated, at least partially, some of the conclusions concerning the
nature of the governing elite in this part of the old world. By this, however,
I do not mean the present authors have limited themselves to a simple
compilation, interpretation and synthesis – which in itself would be
worthy of eulogy. In addition to this, they have made an extraordinary
effort to produce empirical data on the ministerial elites that have not pre-
viously been studied systematically. It is important here to note that in this
attempt to cast light on the eras that remain in the shadow of understanding,

PJSS 6 (3) 213–219 © Intellect Ltd 2007 213

PJSS-6-3-06-Castillo 3/14/08 2:37 AM Page 214

1 Given that the some are more successful than others – depending on the availability of
periodisation information, particularly concerning the more distant periods. Nevertheless,
established by the
authors is different each chapter of this book is a small jewel of political science, an ambitious
for each country, we and useful sketch of the leadership minorities that, in four southern
adopt the criteria of
the type of regime for European countries, were the protagonists during 150 years of regime
comparing the change, industrialisation and social and cultural transformation.
ministerial elite.
The book consists of five chapters, four of which are dedicated to
Portugal, Spain, Italy and Greece respectively, with the fifth being a com-
parative analysis in the form of a conclusion by Nancy Bermeo, who also
co-wrote the preface with Pedro Tavares de Almeida and António Costa
Pinto (this latter two also wrote the chapter on Portuguese ministers).
The editors begin by explaining the importance of ministers in the con-
temporary political arena and describing the purpose of this compilation,
viz, to examine the composition of the southern European ministerial
elites and the rules for their recruitment during the past 150 years, in
order to evaluate the impact of the different regime types and modes of
transition, to scrutinise the similarities and differences between the various
countries and to identify the dominant tendencies and variations over
time. In these first pages the editors also describe the orientation transmit-
ted to the authors through analysis of the ministers’ social profile and
their political cursus honorum, in order to determine the attributes, quali-
ties and types of experience that during the different periods placed the
individuals in the advantageous position from which they could become
members of the ministerial elite.
In the application of the first methodological demand, that is, in the
periodisation of the main regime changes, it is possible to clearly observe
the first similarity between the countries being studied. The seven regime
changes that took place in Greece and Spain, and the five that were identi-
fied in Portugal and Italy, suggest similar national historical trajectories in
terms of political stability – or instability. However, the exercise of the com-
parison becomes even more pertinent when it is established that, through-
out the period being studied, the four states experienced monarchism,
republicanism, authoritarian fascism and democracy almost contempora-
neously. Having arrived at this point, the first question arises: does a model
of political organisation produce a specific type of ministerial elite, inde-
pendent of location? In other words, does the political nature of a regime
determine the recruitment standard and the profile of its ministers?
Answering this question calls for a transversal comparison that demands
a flexible analysis of the nature of the different type of regime.1 That is to
say, that in referring to the monarchical ministerial elite it considers every
Spanish government between 1874 and 1931, even though the Franco
regime was – at least nominally – a monarchy, and not forgetting that the
present head of the Spanish state is a monarch, while in the case of
Greece, it is only concerned with the executives nominated prior to the
installation of Metaxas’s dictatorship (1936–41). The authors also assume
that Italian monarchy ended in 1924, the year in which the last competitive

214 PJSS 6 (3) Review © Intellect Ltd 2007

PJSS-6-3-06-Castillo 3/14/08 2:37 AM Page 215

elections took place. Portugal is the only country in which the regimes fol- 2 In chapter two Linz,
lowed each other in an almost linear sequence from monarchy to republic Jerez and Corzo divide
the first Bourbon
(1910) to dictatorship (1926) to parliamentary democracy (1974). Restoration into three
The average age of the ministers at the date of first appointment ranged periods: 1874–1902,
1902–23 and
between 47 in Portugal and 55.3 in Spain (between 1902 and 1923) during 1925–31 (civil
the monarchical periods.2 While it was common for the capital cities to be directorate/
over-represented within government in the four countries during that
era – a situation that persists to the present – it is also worth noting the
disproportionate number of ministers who came from other regions –
Piedmont in the case of Italy and the Peleponnese and Sterea Hellas in
Greece. The hegemony of these territories in the set of ministers can be
explained by the role their elites played at key moments in the process of
nation building. Nevertheless, over time the proportion of ministers from the
Italian north-west gradually declined (from 47.1 per cent in 1861–76, to
33 per cent in 1913–22), thereby obtaining greater correlation with the
area’s demography (as a proportion of the total population) and the percent-
age of ministers who were integrated into the government. The same trend is
not be seen with the ancient Greek territories, from where at least half of the
members of government were always recruited (with the exception of the
period 1910–36, when they represented 39 per cent of the total).
The level of academic attainment of ministers during this period was
very high. Only rarely did the percentage of those without either a univer-
sity education or military instruction surpass 10 per cent. In the same
fashion, civilians outnumbered members of the armed forces, even though
these latter had a significant presence in governments (apart from Italy
between 1913 and 1922, at least one-fifth of all southern European min-
isters were members of the armed forces). Examining the professions of the
civil ministers we note the preponderance of jurists, who shared their pro-
tagonism with civil servants, university professors and, to a lesser extent,
with writers and journalists in Italy, Portugal and Spain, respectively. In
the Greek case, lawyers represented a much lower proportion (between
6.1 and 13.5 per cent) due to the introduction of the ‘full-time politician’
category that Soritopoulos and Bourikos defined as ‘those who enter poli-
tics immediately after finishing their studies . . . without having exercised
any other occupation’, supporting themselves with family or party
resources. In any case, the percentage of Greek ministers with juridical
qualifications (38.7 per cent during the period 1843–78 and 52 per cent
between 1878 and 1910) is the lowest of the four countries being studied.
The structure of opportunities offered to career politicians in the monar-
chical regimes of southern Europe was marked by clientelism and by the
excessive governmentalisation of the political game. In Spain and Portugal,
for example, caciques and local notables collaborated in the manipulation of
elections in order to form a parliament that was favourable to the govern-
ment in power. Ministers were then selected from amongst these parlia-
mentarians, taking into account such criteria as personal loyalty (to the
head of government) and party loyal (by belonging to the group that

PJSS 6 (3) Review © Intellect Ltd 2007 215

PJSS-6-3-06-Castillo 3/14/08 2:37 AM Page 216

3 Tavares de Almeida sustained the executive). Tavares de Almeida and Costa Pinto call particu-
and Pinto establish lar attention to the fact that the subversion of liberal democratic proce-
two differentiated eras
within the Portuguese dures did not affect the status of parliament as the main source of
authoritarian period: ministerial elite recruitment. And in truth, in these two countries – Spain
the Military
Dictatorship and Portugal – at least 75 per cent of ministers were parliamentary
(1926–33) and the deputies before they entered government.
New State (1933–74).
In the comparative The period between the two world wars is normally associated with
analysis of the two regime types: liberal republicanism and fascist inspired dictatorships.
authoritarian regimes
we only can account,
If in Portugal, Spain and Greece one preceded the other, in Italy there was
in the Portuguese an observable continuity between parliamentary monarchism and ultra-
case, the data right authoritarianism. Given there is only information available for the
referring to the New
State. With respect to Portuguese and Spanish republics (since data on Greek ministers during
Greece, since we do the republican period are included in the period from 1910–36), it is possi-
not have the complete
information for the ble to highlight some singularities shared by the two Iberian states during
Metaxas dictatorship, the republican period. Firstly, in both countries the advent of the republic
our analysis will only
consider the Colonels’ resulted in a visible renewal of the leading elite. For example, in the case of
Regime (1967–74). Spain, 96.6 per cent of ministers and approximately 85 per cent of deputies
were new to their positions. The changes in the composition of the politi-
cal class were simultaneously accompanied by changes in the profile of its
members. In a clear break with the historical predominance of the
national capitals as the centre of governing elite recruitment, over half
(52.1 per cent) of Portuguese ministers and 44.9 per cent of Spanish min-
isters now came from towns and small cities (Estèbe 1982). In this sense,
and as a reflection of the access to power that the rural middle classes
had during these years, there was a significant increase in the number
of medical doctors, school teachers, lawyers and registrars amongst the
ministers in both countries. It is worth noting some of the more significant
differences, such as the number of military officers in the Portuguese gov-
ernment (the military represented 44.8 per cent of all ministers – ten
times more than was the case in Spain) and the hegemony of the
Democratic Party in the majority of Portuguese cabinets, in clear contrast
to the diverse coalitions that existed throughout the life of the Spanish
Second Republic.
As noted above, the years between the two world wars also witnessed
the ascension to power in the majority of European states of heterogeneous
conservative coalitions that led to the establishment of authoritarian and
proto-fascist regimes. The most durable of these, with 48 and 38 years of
existence, were in Portugal and Spain respectively. At the other extreme
was the experience of the Greek dictatorships of 1936–41 and 1967–74,
two periods of authoritarianism separated by two decades of limited
democracy.3 In these regimes ministers were mainly nominated for the
position for the first time after they had reached 50 years of age. Among
them, the academic level was always very high and the proportion of mil-
itary officers was very significant (between 25.6 per cent in the Colonels’
Regime and 33.3 per cent in Franco’s Spain). In the cabinets of Portugal’s
New State, university professors came to represent over 30 per cent of all

216 PJSS 6 (3) Review © Intellect Ltd 2007

PJSS-6-3-06-Castillo 3/14/08 2:37 AM Page 217

ministers, affirming themselves as a powerful social group. In neighbouring

Spain, jurists (34.2 per cent) were the only professionals with a similar
level of ministerial representation to that of the military. In Greece, the
proportion of engineers and architects, which, following the 1967 coup,
increased by up to four times compared to the previous period – probably
as a result of the Colonels’ wish to present a technocratic image of them-
selves that could legitimate them in the eyes of the public. The formation
of governments in the Iberian dictatorships was also subject to the same
‘technocratic’ zeal that decisively marked the recruitment channels and
regulations, turning the public administration into the main centre of
extraction for the ministerial elite. However, although meritocratic criteria
may have prevailed in the ascension to government, to be affiliated in a
single party, such as Portugal’s National Union, or to be well connected
with one of Spain’s Franco regime’s families, Falangist, traditionalist or
Catholic, constituted excellent jumping-off points for personal promotion.
On the other hand, only among the ministers of the Portuguese dictator-
ship was there a significant proportion of members of the previous legisla-
tive power (30.1 per cent had been deputies in the National Assembly).
When the dictatorships came to an end, their ministerial leaders were
permanently removed. In the case of Spain, however, it is still not possible
to speak of an absolute discontinuity between regimes – since the govern-
ments of the democratic era include some of Franco’s former ministers,
including nine who were in transitional cabinets – the level of renewal has
also been very high. One by-product of this ministerial elite renewal was
such rejuvenation that the current average age of ministers in southern
Europe’s democracies ranges between 44.7 (Spain) and 48.7 (Greece).
Other aspects, such as the academic level of the democratic ministerial
class, present fewer variations with respect to the past. Thus, the percent-
age of ministers without university education (1.2 per cent in Portugal,
1.5 per cent in Spain and 2.4 per cent in Greece) suggests a consolidation
of the historical model of working-class exclusion. Even in Italy, where
9 per cent of ministers do not have university education, it is still well
below the 23 per cent that is the western European average of ministers
between 1945 and the mid-1980s who do not have a university educa-
tion. The occupational profile of the ministers also altered with the col-
lapse of southern Europe’s authoritarian regimes, producing a progressive
demilitarisation of cabinets coupled with an increase in the number of
economists and managers. It is also possible to observe a moderate reduc-
tion in the dependence of ministers on public employment in Portugal and
Spain, which could be an indication of the strengthening of the ministerial
elite’s independence in these countries (Etzioni-Halevy 1993). Moreover,
in these democracies, the path towards a ministerial career passed – in the
majority of cases – through parliament (over half of the ministers were
members of a representative chamber) or by senior administrative posi-
tions (over 45 per cent were under-secretaries of state, secretaries of state
or directors-general) – although this last characteristic does not include

PJSS 6 (3) Review © Intellect Ltd 2007 217

PJSS-6-3-06-Castillo 3/14/08 2:37 AM Page 218

the Greek case. Nevertheless, the three countries do seem to coincide in

the fact that their ministerial elite have acquired a more technical profile
relative to their predecessors over the past few years. According to Nancy
Bermeo, the democratisation associated with membership of the European
Community, the need to apply IMF stabilisation programmes (in the case
of Portugal) and development based on state intervention (in the case of
Greece) were factors that stimulated the ‘technocratisation’ of the govern-
ments of the third wave European democracies.
An attentive reader will note that the Italian ministerial elites have
escaped many of the generalisations formulated here. In fact, the ministers
of that country present characteristics that differ in many respects from
their Spanish, Greek and Portuguese peers. Traditionally linked to interest
groups (between 1946 and 1992 over half of all ministers were linked to
one or more of these groups), or with experience in local and regional pol-
itics, they illustrate contemporary Italian ministerial idiosyncrasies. It may
be that the days of these singularities were numbered, since the crisis of the
Italian political system of 1992–6 and the establishment of a majoritarian
democracy have opened the horizons for the formation of a new minister-
ial class in Italy. It is not yet known if it will be closer in its characteristics
to the countries mentioned above. At the present time, as Cotta and
Verzichelli have noted, the changes introduced in the electoral process and
government formation have consolidated the presence of ‘technocrats’ –
who made their appearance at the beginning of the 1990s – and have led
to a decline in the number of ministers with a purely party political past.
When another decade has passed, perhaps we will have a sample suffi-
ciently large to enable us to establish rules and identify tendencies.
In conclusion, there are many qualities to Who governs southern Europe?
Firstly, it is the first book to deal with the evolution of ministerial elites in each
of the four countries, adopting a macro-historical perspective that over-
comes the static vision is traditionally provided by structural-functionalism.
Secondly, the conclusions the authors reach through the investigation of
their respective national elites, provides the reader – both the specialist
and the interested layperson – with an opportunity to verify whether or
not there is a common historical identity in southern Europe.
There can be little doubt this work will prove be a starting point for
future research – whether by the scale of the work undertaken (both in the
revision of specialist literature and in the production of empirical data) or
via the debate the theses and hypotheses presented in these pages may
stimulate. In this regard, what happens with Who governs southern Europe?
is the same as what happens with modern art – one of its attractions rests
in the reflections it invites us to make.

Putnam, R. (1973), ‘The political attitudes of senior civil servants in Western
Europe’, British Journal of Political Science 3: 257–90.

218 PJSS 6 (3) Review © Intellect Ltd 2007

PJSS-6-3-06-Castillo 3/14/08 2:37 AM Page 219

Dogan, M. (ed.) (1975), The mandarins of Western Europe: The political role of top civil
servants, New York, NY: John Wiley.
Aberbach, J., Putnam, R. and Rockman, B. (1981), Bureaucrats and politicians in
Western democracies, Cambridge, MA: Harvard University Press.
Higley, J. and Gunther, R. (eds.) (1992), Elites and democratic consolidation in Latin
America and Southern Europe, Cambridge, Cambridge University Press.
Eyal, G., Szelényi, I. and Townsley, E. (1998), Making capitalism without capitalists:
The new ruling elites in Eastern Europe, London: Verso.
Blondel, J. and Thiébault, J.-L. (eds.) (1991), The profession of government minister in
Western Europe, London: Macmillan.
Estèbe, J. (1982), Les ministres de la république, 1871–1914, Paris: Presses de la
Fondation Nationale des Sciences Politiques.
Etzioni-Halevy, E. (1993), The elite connection: Problems and potential of Western
democracy, Cambridge: Polity Press.

PJSS 6 (3) Review © Intellect Ltd 2007 219

PJSS-6-3-06-Castillo 3/14/08 2:37 AM Page 220
PJSS-6-3-07-Index 3/14/08 2:38 AM Page 221

Index – Volume 6

Almeida, M.A.P., Memory and trauma of the Portuguese agrarian reform: A case study,
pp. 63–76.
Burawoy, M. (2007), Open the social sciences: To whom and for what?, pp. 137–146.
Cerezales, D., ‘Fascist lackeys’? Dealing with the police’s past during Portugal’s transition
to democracy (1974–1980), pp. 155–169.
Eaton, M., From Porto to Portadown: Portuguese workers in Northern Ireland’s labour
market, pp. 171–191.
Freire, A., Minority representation in Portuguese democracy, pp. 193–211.
Gschwend, T., Institutional incentives for strategic voting and party system change in
Portugal, pp. 15–31.
Hespanha, A.M., Form and content in early modern legal books: Bridging material bibli-
ography with history of legal thought, pp. 33–59.
Pinto, J., ‘Open the social sciences: To whom and for what?’, by Michael Burawoy,
pp. 147–154.
Southern, P., German border incursions into Portuguese Angola prior to the First World
War, pp. 3–14.
Teixeira, A.A.C., How has the Portuguese innovation capability evolved? Estimating a
time series of the stock of technological knowledge (1960–2001), pp. 77–95.
Torres, A., Mendes, R., and Lapa, T., Families in Europe, pp. 97–133.

PJSS 6 (3) Index © Intellect Ltd 2007 221

PJSS-6-3-07-Index 3/14/08 2:38 AM Page 222
Volume Six Number Three
ISSN 1476-413X

Portuguese Journal of Social Science | Volume Six Number Three

Portuguese Journal of Social Science
Volume 6 Number 3 – 2007 6.3
137–146 Open the social sciences: To whom and for what?
Michael Burawoy

147–154 ‘Open the social sciences: To whom and for what?’, by Michael Burawoy
José Madureira Pinto
155–169 ‘Fascist lackeys’? Dealing with the police’s past during Portugal’s transition
to democracy (1974–1980)

Journal of
Diego Palacios Cerezales
171–191 From Porto to Portadown: Portuguese workers in Northern Ireland’s
labour market
Martin Eaton

Social Science
193–211 Minority representation in Portuguese democracy
André Freire

213–219 Francisco Javier Luque Castillo

221 Index – Volume 6

The Portuguese Journal of Social Science is published by intellect in partnership with:

intellect Journals | Media & Culture

ISSN 1476-413X

9 771476 413007 www.intellectbooks.com