The Warwick Debates
Anthony D. Smith's opening statement
Nations and their pasts
May I first thank the Vice-Chancellor of Warwick University, and Professor
Edward Mortimer, for their kind invitation to me to take part in this unique
series of debates about nationalism. It is certainly a welcome idea to get
away from the usual lecture format, and I hope that you will forgive me, if,
with the agreement of Professor Ernest Gellner, I detain you for some
twenty minutes with a rather telescoped opening statement.
The present occasion affords me an opportunity to express my great
appreciation of the achievement and the inspiration provided to me, and to
all of us, by Ernest Gellner. It was his work that first caught my imagination
in 1964, when I was groping for a method of studying phenomena that had
already for some time absorbed me and that were, after the wave of
decolonisation in Africa and Asia, very much in the air. Not only did Ernest
encourage and guide my thesis on theories of nationalism with much
patience and care; he taught me some fundamental lessons about nations
and nationalism, which served me for a guide thereafter.
The first is that nationalism is elusive, even protean, in its manifestations;
and so we have to try to classify the rich variety of movements and
ideologies if we are to make any progress in understanding so variegated a
phenomenon. Second, he taught me to appreciate the underlying sociological reality of
nationalism and its creation, the nation. Against all those who would tell us
that the nation exists only in the imagination and that it can be
deconstructed away, Ernest has always insisted that nations and nationalism
are real and powerful sociological phenomena, even if their reality is quite
different from the tale told about them by nationalists themselves.
And third, he convinced me that nations, as well as nationalism, are
modern phenomena, in the sense that the basic features of the modern
world require nations and nationalisms. You could not have one without
the other. This is obvious in the case of nationalism, the ideological
movement, which clearly did not exist before the eighteenth century. But it
is also true of nations in general. That is to say, even if a few nations could
be found before the advent of modernity, most nations are modern in the
sense of being relatively recent in time - and necessarily modern.
And yet - and this is why, in the spirit of this evening, I shall admit that
debates are not just invented - there is at this point a certain difference
between Ernest and myself.
Insofar as he is a wholehearted 'modernist', Ernest would claim that the
nation is not only relatively recent; it is also the product of specifically
modern conditions - those of early industrialism or its anticipations, social
mobility, the need for mass literacy, public education and the like. It is the
modern transition from spontaneous, non-literate 'low' cultures to highly
cultivated, literate and specialised 'high' cultures that engenders nationalism
and nations (Gellner 1964; 1983).
Now, it is not that I find this account wrong, only that it tells half the story.
There is another half, and other ways of looking at this protean phenomenon.
I shall try to tell this other half and consider some of these other ways.
I think most of us would agree that nationalism is today one of the most
powerful forces in the world, and that the national state has been for a
century at least, and continues to be, the cornerstone of international
politics. Nationalism provides the sole legitimation of states the world over,
including the many polyethnic and federal ones. It is also the most
widespread and popular ideology and movement, and it comes as no
surprise that many of the world's most intractable conflicts - in India and
the Middle East, the Caucasus and the Horn of Africa, the Balkans and
southern Africa - are either ethno-national conflicts or possess a strong
Of course, it is easy to exaggerate the influence of nationalism, and to
inflate the terms, nation and nationalism, to cover every aspect of a state's
social, cultural and political policy, and every dimension of inter-state
relations. The first thing, therefore, is to define our concepts.
By 'nationalism' I shall mean an ideological movement for the attainment and
maintenance of autonomy, unity and identity of a human population, some of
whose members conceive it to constitute an actual or potential 'nation'.
A 'nation' in turn I shall define as a named human population sharing an
historic territory, common myths and memories, a mass, public culture, a
single economy and common rights and duties for all members.
This definition suggests that the concept of the nation refers to a
particular kind of social and cultural community, a territorial community of
shared history and culture. This is the assumption of nationalists themselves,
for whom the world is composed of unique historic culture-communities, to
which their citizens owe a primary loyalty and which are the sole source of
political power and inner freedom (Smith 1991, chs. 1, 4).
It is important to distinguish the concept of the nation from that of the
state. The state is a legal and political concept; states can be defined as
autonomous, public institutions of coercion and extraction within a
recognised territory. States are not communities (Tivey 1980).
We should also recall that the systems of states that arose, first in Europe
and then in other parts of the world, often preceded the rise of nationalism,
as well as many of today's nations, though not necessarily many of their
core ethnic groups. This lack of temporal and spatial fit between state and
nation is one of the main causes of many of today's national conflicts (Tilly
1975, Introduction and Conclusion).
Modernist and 'post-modernist' theories
For most people, nations, especially their own nations, appear to be
perennial and immemorial. They cannot easily imagine a world without
nations, nor are they happy with the idea that their nation is a recent
creation, or even a construct of elites. Indeed, an older generation of
scholars, often under the unconscious influence of nationalism, tended to
seek and find 'nations' everywhere, in all ages and continents (Walek-
Czernecki 1929; Tipton 1973).
Today, however, most scholars would regard the idea of nations existing
perennially through antiquity and the middle ages as simply 'retrospective
nationalism'. For most post-war scholars, nations and nationalisms are
fairly recent phenomena, arising immediately before, during or in the wake
of the French Revolution. They also tend to see nations and nationalisms as
products of modernisation and features of modernity. Many of these modernist'
theories are, at root, materialist. In some cases, the materialism
is explicit. Tom Nairn, for example, regards nationalism as the product of,
and response to, the 'uneven development' of capitalism. In other cases, the
materialism is part and parcel of other, cultural processes of modernisation
- be it the mobile society based on a public system of mass, standardised
literary education which Ernest Gellner regards as critical, or the rise of
reading publics engendered by the spread of the technology of 'print-
capitalism', stressed by Benedict Anderson. In all these cases, nations and
nationalisms are viewed as more or less inevitable outgrowths of a modern,
industrial society, however regrettable their consequences may be (Anderson
1983; Gellner 1973; Nairn 1977, ch. 9; Smith 1988).
It is, of course, in the deconstructionist models of Benedict Anderson and
Erie Hobsbawm that the question of the real or imagined status of the
nation has been most sharply posed. In Hobsbawm's approach, the nation
is seen, in large part, as a set of 'invented traditions' comprising national
symbols, mythology and suitably tailored history. In Anderson's model, the
nation is seen as an 'imagined political community', one that is imagined as
both finite and sovereign. I do not think that either would regard the nation
as a wholly imaginary category; at the same time, they wish to debunk
nationalist views of the nation as somehow 'primordial' and 'perennial'.
(Anderson 1983, ch. 3; Hobsbawm and Ranger 1983, ch. 1).
This seems to me quite proper, provided that, in designating the nation
as an imagined community or tradition, we do not gainsay its reality or
consider it a fabrication. There is nothing contradictory about saying that
something is both imagined and real: the Parthenon, Chartres and the
Sistine ceiling are no less real and tangible for all the imagination of their
creators and spectators down the ages. But, if nations are not fabricated,
are they cultural artefacts created in the same way as artistic monuments? I
shall argue that, although we can often discern elements of deliberate
planning and human creativity in their formation, nations and nationalisms
are also the products of preexisting traditions and heritages which have
coalesced over the generations.(1)
Let me return to the far more acceptable 'modernist' theories, those that
emphasise the sociological reality of nations, once they have been formed.
There are certain problems associated with these theories. The first is their
generality. Though they make out a convincing case for explaining
'nationalism-in-general', they are often pitched at such a level of abstraction
that they cannot be easily applied to specific areas or cases. They appear to
cover everything, and yet, when we look to them to illuminate actual
historical instances, they so often invoke exceptional circumstances - like
the religious factor, or colour, or a history of ethnic antagonism (Gellner
1983, ch. 6).(2)
Second, their materialism is often quite misleading. Nationalism can
emerge in all kinds of socioeconomic milieux - in rich Quebec and poor
Eritrea, in areas of decline as well as improvement, in pre-industrial as well
as industrial conditions. Nor is it easy to explain the content and intensity
of particular nationalisms through the workings of global capitalism or the
dynamics of relative deprivation (Connor 1984; Smith 1981, ch. 2).
But the third problem is the most crucial, since it stems from their
commitment to modernism, the idea that nations and nationalisms are the
product of modernisation. What this systematically overlooks is the
persistence of ethnic ties and cultural sentiments in many parts of the world,
and their continuing significance for large numbers of people. Erie
Hobsbawm, indeed, goes so far as to deny any connection between the
popular 'Proto-national' communities that he analyses and subsequent
political nationalisms (Hobsbawm 1990, ch. 2).
This is exactly where I disagree. Modern political nationalisms cannot be
understood without reference to these earlier ethnic ties and memories, and,
in some cases, to pre-modern ethnic identities and communities. I do not wish
to assert that every modern nation must be founded on some antecedent
ethnic ties, let alone a definite ethnic community; but many such nations have
been and are based on these ties, including the first nations in the West -
France, England, Castile, Holland, Sweden - and they acted as models and
Pioneers of the idea of the 'nation' for others. And when we dig deeper, we
shall find an ethnic component in many national communities since - whether
the nation was formed slowly or was the outcome of a more concerted project
of 'nation-building' (see Armstrong 1982; Smith 1986, 1994).
I believe that this kind of approach which we may term 'ethno-symbolic',
is more helpful for understanding the growth of nations, the rise of ethno-
nationalisms and the conflicts to which they give rise.
To begin with, it is an approach or perspective, not a theory. I doubt
whether we are in a position yet to offer a theory of so protean and many-
sided a set of phenomena as ethnies, nations and nationalisms, except at a
very general level.(3) Second, this kind of approach may help to explain which populations are likely to give rise to a nationalist movement under certain conditions, and
what the content of their nationalism is likely to be - though there is much
work to be done here. An exploration of earlier ethnic configurations will, I
suggest, help us to explain the major issues and concerns of a subsequent
nationalism in a given population and provide us with clues about the likely
growth of a nation and its nationalism. Modern Greece provides an
example. Its dual heritage of Byzantine imperial Orthodoxy and classical
democratic antiquity shaped the patterns and contents of rival - Greek
nationalisms in the nineteenth century and beyond - and suggests some
reasons for expecting the rise of a powerful nationalism among the Greeks
rather than, say, the neighbouring Vlachs (Campbell and Sherrard 1968,
ch. 1; Kitromilides 1989).
Third, the approach that I recommend emphasises the important role of
memories, values, myths and symbols. Nationalism very often involves the
pursuit of 'symbolic' goals - education in a language, having your own
language TV channel, the preservation of ancient sacred sites like the
mosque at Ayodhya or the Wailing Wall area, the right to worship in one's
own way, have one's own courts, schools and press, wear particular
costume, and so on - goals which often bring protest and bloodshed, based
as they are on popular memories, symbols and myths. Materialist,
rationalist and modernist theories tend to have little to say about these
issues, especially the vital component of collective memories (see Connor
1993; Horowitz 1985, ch. 2; Hutchinson 1987; Kapferer 1988; cf. Tonkin,
McDonald and Chapman 1989).
Fourth, an ethno-symbolic approach can help us to understand why
nationalism so often has such a widespread popular appeal. The intelligen-
tsia may 'invite the masses into history' and politicise them and their
cultures. But why do 'the people' respond? Not simply because of promises
of material benefits. Their vernacular culture is now valued and turned into
the basis of a new mass, public culture of the nation. So nationalism often
involves the vernacular mobilisation of the masses (Nairn 1977, ch. 2; Smith
This is why the ethnic form of nationalism has become such a powerful
force today. Unlike the civic, territorial nationalism of the French Revolu-
tion and the West, which sees the nation as a territorial association of
citizens living under the same laws and sharing a mass, public culture,
ethnic nationalism regards the nation as a community of genealogical
descent, vernacular culture, native history and popular mobilisation. The
civic kind of nationalism is a nationalism of order and control, and it suits
the existing national states and their dominant ethnies. But it has little to
offer the many submerged ethnic minorities incorporated into the older
einpires and their successor states. So they and their intelligentsias turn to
ethnic nationalism, and try to reconstruct their community as an ethnic
nation. Theirs is the politics of cultural revolt. Revolt not only against alien
rulers, but against 'the fathers', the passive older generations, guardians of
ancestral traditions and notables of a traditional order. To achieve their
cultural revolution, they must thrust their ethnic communities into the
political arena and turn them into political nations (see Kedourie 1971,
Introduction; and Smith 1995, ch. 4).
Here is the deeper, inner source of so many ethnic and national conflicts
today. The clash of rival nationalisms, ethnic and civic, is at the heart of the
conflicts in the Middle East, India, the Caucasus and Balkans. We can also
find it in more muted, but no less persistent, form in the West: in Quebec
and Euzkadi, Scotland and Catalonia, Flanders and Corsica, wherever
members of marginalised, threatened or aspiring ethnic communities seek to
restore their heritage, language and culture.
What follows from this analysis? First, that in a world of political and
cultural pluralism where states and ethnies operate with rival conceptions of
the nation and its boundaries, ethno-national conflict is endemic. Second,
that nations and nationalisms are a political necessity in a world of
competing and unequal states requiring popular legitimation and mobilisa-
tion (Smith 1995, ch. 6). Third, that because so many people feel their
nation performs important social and political functions, it is going to take
more than a Maastricht Treaty to wean them away from these deeply felt
national allegiances.(4) And finally, because so many nations are historically
embedded in pre-modern ethnic ties, memories and heritages, we are
unlikely to witness in our lifetime the transcendence of the nation and the
supersession of nationalism, of which so many utopians have dreamt!
1 These arguments about the role of nationalist 'agency' versus modern or pre-modern, structures' can be found in Breuilly (1993) and Smith (1991).
2 Gellner (1994), in distinguishing between time-zones in the development of nationalism in different parts of Europe, does implicitly introduce contingent historical elements to supplement his general theory.
3 Geliner is perhaps the only scholar to ofrer a full and explicit theory; but Nairn (1977, ch. 2), Breuilly (1993) and Mann (1995) offer partial theories of aspects and/or types of nationalism.
4 On the question on European integration and national identity, see Smith (1992) and Schlesinger (1992).