NATIONAL IDENTITY
ANTHONY D. SMITH
PENGUIN BOOKS
PENGUIN BOOKS
Published by the Penguin Group
Penguin Books Ltd, 27 Wrights Lane, London W8 5TZ, England
Penguin Books USA Inc., 375 Hudson Street, New York, New York 10014, USA
Penguin Books Australia Ltd, Ringwood, Victoria, Australia
Penguin Books Canada Ltd, 10 Alcorn Avenue, Toronto, Ontario, Canada M4V 3B2
Penguin Books (NZ) Ltd, 182-190 Wairau Road, Auckland 10, New Zealand
Penguin Books Ltd, Registered Offices: Harmondsworlh, Middlesex, England
First published 1991
5 7 9 10 8 6 4
Copyright © Anthony D. Smith, 1991
AH rights reserved
The moral right of the author has been asserted
Printed in England by Clays Ltd, St Ives pic
Set in 10/12 pt Monophoto Bembo
Except in the United States of America, this book is sold subject
to the condition that it shall not, by way of trade or otherwise, be lent,
re-sold, hired out, or otherwise circulated without the publisher's
prior consent in any form of binding or cover other than that in
which it is published and without a similar condition including thif
condition being imposed on the subsequent purchaser
CHAPTER 2
The Ethnic Basis of National Identity
The origins of what we have termed national identity are as complex
as its nature. I am not saying simply that the origins of each nation
are in many ways unique and that there is great variation in the
starting-points, trajectories, rates and timings of modern nations.
The very question 'what are the origins of nations?' needs to be
broken down into several further questions, such as: who is the
nation? Why and how is the nation? When and where is the
nation?
In fact we can conveniently use these questions to seek a general
explanation of the origins and development of modern nations in
three parts.
1. Who is the nation? What are the ethnic bases and models of
modern nations? Why did these particular nations emerge?
2. Why and how does the nation emerge? That is, what are the
general causes and mechanisms that set in motion the processes
of nation-formation from varying ethnic ties and
memories?
3. When and where did the nation arise? What were the specific
ideas, groups and locations that predisposed the formation
of individual nations at particular times and places?
Through answers to these questions, albeit of a general and necessarily
incomplete nature, we may hope to shed some light on the
vexed problem of national origins and development.
ETHNIE AND ETHNO-GENESIS
If myths like that of Oedipus can be seen as widely believed tales
told in dramatic form, referring to past events but serving present
purposes and/or future goals, then the nation stands at the centre of
one of the most popular and ubiquitous myths of modern times:
that of nationalism. Central to this myth is the idea that nations
exist from time immemorial, and that nationalists must reawaken
19
THE ETHNIC BASIS OF NATIONAL IDENTITY
them from a long slumber to take their place in a world of nations.
The hold of the nation lies, as we shall see, partly in the promise
of the nationalist salvation drama itself. But this power is often
immeasurably increased by the living presence of traditions embodying
memories, symbols, myths and values from much earlier epochs
in the life of a population, community or area. So it is these
pre-modern ethnic identities and traditions that we must first
explore.'
The concept of 'ethnicity' has received a good deal of attention in
recent years. For some it has a 'primordial' quality. It exists in
nature, outside time. It is one of the 'givens' of human existence
(this is a view that has received some backing recently from sociobiology,
where it is regarded as an extension of processes of genetic
selection and inclusive fitness). At the other extreme ethnicity is
seen as 'situational'. Belonging to an ethnic group is a matter of
attitudes, perceptions and sentiments that are necessarily fleeting and
mutable, varying with the particular situation of the subject. As the
individual's situation changes, so will the group identification; or at
least, the many identities and discourses to which the individual
adheres will vary in importance for that individual in successive
periods and different situations. This makes it possible for ethnicity
to be used 'instrumental'' to further individual or collective interests,
particularly of competing élites who need to mobilize large
followings to support their goals in the struggle for power. In this
struggle ethnicity becomes a useful tool.2
Between these two extremes lie those approaches that stress the
historical and symbolic—cultural attributes of ethnic identity. This is
the perspective adopted here. An ethnic group is a type of cultural
collectivity, one that emphasizes the role of myths of descent and
historical memories, and that is recognized by one or more cultural
differences like religion, customs, language or institutions. Such
collectivities are doubly 'historical' in the sense that not only are
historical memories essential to their continuance but each such
ethnic group is the product of specific historical forces and is therefore
subject to historical change and dissolution.
At this point it is useful to distinguish between ethnic categories
and ethnic communities. The former are human populations whom at
least some outsiders consider to constitute a separate cultural and
20
THE ETHNIC BASIS OF NATIONAL IDENTITY
historical grouping. But the populations so designated may at the
time have little self-awareness, only a dim consciousness that they
form a separate collectivity. Thus Turks in Anatolia before 1900
were largely unaware of a separate 'Turkish' identity — separate,
that is, from the dominant Ottoman or the overarching Islamic
identities - and besides, local identities of kin, village or region were
often more important. The same can be said for the Slovak inhabitants
of the Carpathian valleys before 1850, despite their common
dialects and religion. In both cases a myth of common origins,
shared historical memories, a sense of solidarity or an association
with a designated homeland were largely absent.3
An ethnic community, on the other hand, can be distinguished
by just these attributes, even if they are firmly held and clearly
enunciated by only small segments of the designated population and
even if some of these attributes are more intense and salient than
others at a given period. We may list six main attributes of ethnic
community (or ethnie, to use the French term):
1. a collective proper name
2. a myth of common ancestry
3. shared historical memories
4. one or more differentiating elements of common culture
5. an association with a specific 'homeland'
6. a sense of solidarity for significant sectors of the popula-
tion.4
The more a given population possesses or shares these attributes
(and the more of these attributes that it possesses or shares), the
more closely does it approximate the ideal type of an ethnic community
or ethnie. Where this syndrome of elements is present we
are clearly in the presence of a community of historical culture with
a sense of common identity. Such a community must be sharply
differentiated from a race in the sense of a social group that is held to
possess unique hereditary biological traits that allegedly determine
the mental attributes of the group.5
In practice, ethnies are often
confused with races, not only in this social sense but even in the
physical, anthropological sense of subspecies of Homo sapiens such as
Mongoloid, Negroid, Australoid, Caucasian and the like. Such a
confusion is the product of the widespread influence of racist ideologies
and discourses, with their purportedly 'scientific' notions of
21
THE ETHNIC BASIS OF NATIONAL IDENTITY
racial struggle, social organisms and eugenics. In the hundred years
from 1850 to 1945 such notions were applied to the purely cultural
and historical differences of ethnies, both inside Europe and in colonial
Africa and Asia, with results that are all too well known.6
But a glance at the above list of ethnic attributes reveals not only
their largely cultural and historical content, but also (with the exception
of number 4) their strongly subjective components. Most important,
it is myths of common ancestry, not any fact of ancestry
(which is usually difficult to ascertain), that are crucial. It is fictive
descent and putative ancestry that matters for the sense of ethnic
identification. Indeed, Horowitz has likened ethnic groups to
'super-families' of fictive descent because members view their ethnie
as composed of interrelated families, forming one huge 'family'
linked by mythical ties of filiation and ancestry. Such a linkage
between family and nation reappears in nationalist mythologies and
testifies to the continuing centrality of this attribute of ethnicity.
Without such descent myths it is difficult to see ethnies surviving for
any length of time. The sense of'whence we came' is central to the
definition of'who we are'.7
What I have termed 'shared historical memories' may also take
the form of myth. Indeed, for many pre-modern peoples the line
between myth and history was often blurred or even non-existent.
Even today that line is not as clear-cut as some would like it to be;
the controversy over the historicity of Homer and the Trojan War
is a case in point. So are the tales of Stauffacher and the Oath of the
Riitli, and of William Tell and Gessler, which have entered the
'historical consciousness' of every Swiss. It is not only that widely
believed dramatic tales of the past serving present or future purposes
grow up readily around kernels of well-attested events: in addition,
myths of political foundation, liberation, migration and election
take some historical event as their starting-point for subsequent
interpretation and elaboration. The conversion of Vladimir of Kiev
to Christianity (in AD 988) or the founding of Rome (in 753 BC?)
may be treated as historical events, but their significance resides in
the legends of foundation with which they are associated. It is these
associations that confer on them a social purpose as sources of
political cohesion.8
Similarly, attachments to specific stretches of territory, and to
22
THE ETHNIC BASIS OF NATIONAL IDENTITY
certain places within them, have a mythical and subjective quality.
It is the attachments and associations, rather than residence in or
possession of the land that matters for ethnic identification. It is
where we belong. It is also often a sacred land, the land of our
forefathers, our lawgivers, our kings and sages, poets and priests,
which makes this our homeland. We belong to it, as much as it
belongs to us. Besides, the sacred centres of the homeland draw the
members of the ethnie to it, or inspire them from afar, even when
their exile is prolonged. Hence, an ethnie may persist, even when
long divorced from its homeland, through an intense nostalgia and
spiritual attachment. This is very much the fate of diaspora communities
like the Jews and Armenians.9
It is only when we come to the varying elements of a common
culture that differentiate one population from another that more
objective attributes enter the picture. Language, religion, customs
and pigmentation are often taken to describe objective 'cultural
markers' or differentiae that persist independently of the will of
individuals, and even appear to constrain them. Yet it is the significance
with which colour or religion is endowed by large numbers
of individuals (and organizations) that matters more for ethnic
identification even than their durability and independent existence,
as the growing political significance of language and colour over the
last two or three centuries demonstrates. It is only when such
markers are endowed with diacritical significance that these cultural
attributes come to be seen as objective, at least as far as ethnic
boundaries are concerned.10
All of this suggests that the ethnie is anything but primordial,
despite the claims and rhetoric of nationalist ideologies and discourses.
As the subjective significance of each of these attributes
waxes and wanes for the members of a community, so docs the
cohesion and self-awareness of that community's membership. As
these several attributes come together and become more intense and
salient, so does the sense of ethnic identity and, with it, of ethnic
community. Conversely, as each of these attributes is attenuated and
declines, so does the overall sense of ethnicity, and hence the ethnie
itself would dissolve or be absorbed."
How docs an ethnie form? We can give only some very tentative
answers. Where such processes are visible in the historical record
23
THE ETHNIC BASIS OF NATIONAL IDENTITY
they suggest certain patterns of ethnic formation. Empirically, these
arc of two main kinds: coalescence and division. On the one hand,
wc can trace ethnic formation through the coming together of
separate units, and this in turn can be broken down into processes of
amalgamation of separate units, such as city-states, and of absorption
of one unit by another, as in the assimilation or regions or 'tribes'.
On the other hand, ethnies may be subdivided through fission, as
with sectarian schism, or through what Horowitz calls 'proliferation',
when a part of the ethnic community leaves it to form a new
group, as in the case of Bangladesh.12
The frequency of such processes suggests the shifting nature of
ethnic boundaries and the malleability, within certain limits, of their
members' cultural identity. It also reveals the 'concentric' nature of
ethnic, and more generally collective cultural, affiliations. That is to
say, individuals may feel loyalty not only to their families, villages,
castes, cities, regions and religious communities, as well as to class
and gender identifications; they may also feel allegiances to different
ethnic communities at different levels of identification simultaneously.
An example of this in the ancient world would be the
sentiment of ancient Greeks as members of a polis, or the 'sub-ethnie'
(Dorians, Ionians, Aeolians, Boeotians, etc. — really ethnic identities
in their own right) and of the Hellenic cultural ethnie.11
In the
modern world the various clans, languages and ancestral 'sub-ethnies'
of the Malays or Yoruba furnish examples of the concentric circles
of ethnic identity and allegiance. Of course, at any one time one or
other of these concentric circles of allegiance may be to the fore for
political, economic or demographic reasons; but this serves only to
reinforce 'instrumentalist' arguments against the primordial nature
of ethnic communities and to highlight the importance of boundary
changes.1
'*
At the same time this is the only part of the story. We must not
overstate the mutability of ethnic boundaries or the fluidity of their
cultural contents. To do so would deprive us of the means of
accounting for the recurrence of ethnic ties and communities (let
alone their original crystallizations) and their demonstrable durability
over and above boundary and cultural changes in particular
instances. It would dissolve the possibility of constituting identities
that were more than successive fleeting moments in the perceptions,
24
THE ETHNIC BASIS OF NATIONAL IDENTITY
attitudes and sentiments of identifying individuals. Worse, we would
be unable to account for any collectivity, any group formation,
from the myriad moments of individual sentiment, perception and
memory. But the fact remains that, as with other social phenomena
of collective identity like class, gender and territory, ethnicity exhibits
both constancy and flux side by side, depending on the
purposes and distance of the observer from the collective phenomenon
in question. The durability of some ethnies, despite changes in
their demographic composition and some of their cultural distinctiveness
and social boundaries, must be set against the more
instrumentalist or phenomenological accounts that fail to consider
the importance of antecedent cultural affinities that set periodic
limits to the redefinitions of ethnic identities.'5
Any realistic account of ethnic identity and ethno-genesis must,
therefore, eschew the polar extremes of the primordialist-instrumentalist
debate and its concerns with, on the one hand, fixity of
cultural patterns in nature and, on the other, 'strategic' manipulability
of ethnic sentiments and continuous cultural malleability.
Instead we need to reconstitute the notion of collective cultural
identity itself in historical, subjective and symbolic terms. Collective
cultural identity refers not to a uniformity of elements over
generations but to a sense of continuity on the part of successive
generations of a given cultural unit of population, to shared
memories of earlier events and periods in the history of that unit
and to notions entertained by each generation about the collective
destiny of that unit and its culture. Changes in cultural identities
therefore refer to the degree to which traumatic developments disturb
the basic patterning of the cultural elements that make up the
sense of continuity, shared memories and notions of collective destiny
of given cultural units of population. The question is how far
such developments disrupt or alter the fundamental patterns of
myth, symbol, memory and value that bind successive generations
of members together while demarcating them from 'outsiders' and
around which congeal the lines of cultural differentiation that serve
as 'cultural markers' of boundary regulation.16
We may illustrate these points by considering briefly some cases
of disruptive culture change that nevertheless renewed, rather than
destroyed, the sense of common ethnicity and its identity as we
25
THE ETHNIC BASIS OF NATIONAL IDENTITY
defined it above. Typical events that generate profound changes in
the cultural contents of such identity include war and conquest,
exile and enslavement, the influx of immigrants and religious
conversion. The Persians, at least from the Sassanid period, were
subjected to conquest by Arabs, Turks and others, were gradually
converted to Islam and experienced more than one influx of immigrants.
Yet, despite all the changes of collective cultural identity
consequent on these processes, a Persian sense of distinctive ethnic
identity persisted, and at times received a new lease of life, notably in
the renaissance of the New Persian linguistic and literary revival of
the tenth and eleventh centuries.'7
The Armenians too experienced
traumatic events that had profound consequences for the cultural
contents of their ethnic identity. They were the first constituted
kingdom and people to convert to Christianity, were fought over
by Sassanids and Byzantines, were defeated, excluded and partly
exiled, received considerable influxes of immigrants and were finally
subjected to mass deportation and genocide in part of their homeland.
Yet, despite changes in location, economic activities, social
organization and parts of their culture over the centuries, a sense of
common Armenian identity has remained throughout their diaspora,
and the forms of their antecedent culture, notably in the sphere ot
religion and language/script, have ensured a subjective attachment
to their cultural identity and separation from their surroundings.'8
These examples suggest the further observation that a combination
of often adverse external factors and a rich inner or 'ethno'history
may help to crystallize and perpetuate ethnic identities. If
the origins of cultural differentiation itself are lost in the last states
of prehistory, we may at least attempt to isolate those recurrent
forces that appear to coalesce the sense of ethnic identification and
ensure its persistence over long periods.
Of these, state-making, military mobilization and organized religion
appear to be crucial. Long ago Weber commented on the
importance of political action for ethnic formation and persistence,
arguing 'It is primarily the political community, no matter how
artificially organized, that inspires the belief in common ethnicity.'19
It is possible to exaggerate the role of state-making in ethnic crystallization
(one thinks of the failure of Burgundy, and the qualified
success of Prussia); yet, clearly, the foundation of a unified polity, as
26
THE ETHNIC BASIS OF NATIONAL IDENTITY
in ancient Egypt, Israel, Rome, Sassanid Persia, Japan and China,
not to mention France, Spain and England, played a major role in
the development of a sense of ethnic community and, ultimately, of
cohesive nations.20
Warfare is, if anything, even more important. Not only does
'war make the state (and the state makes war)', as Tilly declared; it
fashions ethnic communities not only from the contestants but even
from third parties across whose territories such wars arc often conducted.
The case of ancient Israel is only the most striking, caught as
it was between the great powers of the ancient Near East, Assyria
and Egypt. Armenians, Swiss, Czechs, Kurds and Sikhs afford other
instances of strategically located communities whose sense of
common ethnicity, even when it did not originate from these events,
was crystallized time and again by the impact of protracted warfare
between foreign powers in which they were caught up. As for the
contestants themselves, we need note only the frequency with which
ethnies are antagonistically paired: French and English, Greeks and
Persians, Byzantines and Sassanids, Egyptians and Assyrians, Khmers
and Vietnamese, Arabs and Israelis . . . While it would be an exaggeration
to deduce the sense of common ethnicity from the fear of
the 'outsider' and paired antagonisms, there is no denying the central
role of warfare, not, as Simmel suggested, as a crucible of ethnic
cohesion (war may fracture that cohesion, as it did in the Great War
in some European countries) but as a mobilizer of ethnic sentiments
and national consciousness, a centralizing force in the life of the
community and a provider of myths and memories for future generations.
It is perhaps this last function that enters most deeply into the
constitution of ethnic identity.21
As for organized religion, its role is both spiritual and social. The
myth of common ethnic origins is often intertwined with creation
myths - such as that of Deucalion and Pyrrha in Hesiod's Theogony
and that of Noah in the Bible — or at least presupposes them. Very
often the heroes of the ethnic community are also those of religious
lore and tradition, albeit treated as 'servants of God' rather than
ethnic founders or leaders, as was the case with Moses, Zoroaster,
Muhammad, St Gregory, St Patrick and many others. The liturgy
and rites of the Church or community of the faithful supply the
texts, prayers, chants, feasts, ceremonies and customs, sometimes
27
THE ETHNIC BASIS OF NATIONAL IDENTITY
even the scripts, of distinctive ethnic communities, setting them
apart from neighbours. And over all this heritage of cultural difference
stand the 'guardians of the tradition', the priests, scribes and
bards who record, preserve and transmit the fund of ethnic myths,
memories, symbols and values encased in sacred traditions commanding
the veneration of the populace through temple and church,
monastery and school, into every town and village within the realm
of the culture—community.22
State-making, protracted warfare and organized religion, though
they figure prominently in the historical record of ethnic crystallization
and persistence, may also operate to break up, or cut across,
ethnic identifications. This happened when empires like those of
Assyria and Achaemenid Persia created the conditions for a sustained
intermingling of ethnic categories and communities in an Aramaicspeaking
and syncretistic civilization, and when prolonged wars and
rivalries put an end to ethnic states and communities like the Carthaginians
and Normans (in Normandy). Ethnic identity also developed
when religious movements burst across ethnic frontiers and
founded great supra-territorial organizations, Buddhist, Catholic or
Orthodox, or conversely, through schism, divided the members of
ethnic communities such as the Swiss or Irish. Yet, for all these
cases, we may find many more that confirm the close links between
ethnic crystallization and the antecedent role of states, warfare and
organized religion.
ETHNIC CHANGE, DISSOLUTION AND SURVIVAL
The importance of these and other factors can also be seen when we
turn to the closely related questions of how ethnies change in character,
dissolve or survive.
Let me start with ethnic change and with a well-known example,
that of the Greeks. Modern Greeks are taught that they are the heirs
and descendants not merely of Greek Byzantium, but also of the
ancient Greeks and their classical Hellenic civilization. In both cases
(and there have in fact been two, rival, myths of descent at work
since the early nineteenth century), 'descent' was seen in largely
demographic terms; or rather, cultural affinity with Byzantium and
ancient Greece (notably Athens) was predicated on demographic
28
THE ETHNIC BASIS OF NATIONAL IDENTITY
continuity. Unfortunately for the classicist Hellenic myth, the
demographic evidence is at best tenuous, at worst non-existent. As
Jacob Fallmereyer demonstrated long ago, Greek demographic continuity
was brutally interrupted in the late sixth to eighth centuries
AD by massive influxes of Avar, Slav and, later, Albanian immigrants.
The evidence from the period suggests that the immigrants
succeeded in occupying most of central Greece and the Péloponnèse
(Morea), pushing the original Greek-speaking and Hellenic inhabitants
(themselves already intermingled with earlier Macedonian,
Roman and other migrants) to the coastal areas and the islands of
the Aegean. This shifted the centre of a truly Hellenic civilization to
the cast, to the Aegean, the Ionian littoral of Asia Minor and to
Constantinople. It also meant that modern Greeks could hardly
count as being of ancient Greek descent, even if this could never be
ruled out.23
There is a sense in which the preceding discussion is both relevant
to a sense of Greek identity, now and earlier, and irrelevant. It is
relevant in so far as Greeks, now and earlier, felt that their 'Greekncss'
was a product of their descent from the ancient Greeks (or Byzantine
Greeks), and that such filiation made them feel themselves to be
members of one great 'super-family' of Greeks, shared sentiments of
continuity and membership being essential to a lively sense of identity.
It is irrelevant in that ethnies are constituted, not by lines of
physical descent, but by the sense of continuity, shared memory and
collective destiny, i.e. by lines of cultural affinity embodied in
distinctive myths, memories, symbols and values retained by a given
cultural unit of population. In that sense much has been retained,
and revived, from the extant heritage of ancient Greece. For, even
at the time of Slavic migrations, in Ionia and especially in Constantinople,
there was a growing emphasis on the Greek language,
on Greek philosophy and literature, and on classical models of
diought and scholarship. Such a 'Greek revival' was to surface again
in the tenth and fourteenth centuries, as well as subsequently, providing
a powerful impetus to the sense of cultural affinity with ancient
Greece and its classical heritage.24
This is not to deny for one moment cither the enormous cultural
changes undergone by the Greeks despite a surviving sense of
common ethnicity or the cultural influence of surrounding peoples
29
THE ETHNIC BASIS OF NATIONAL IDENTITY
and civilizations over two thousand years. At the same time in
terms of script and language, certain values, a particular environment
and its nostalgia, continuous social interactions, and a sense of religious
and cultural difference, even exclusion, a sense of Greek identity
and common sentiments of ethnicity can be said to have persisted
beneath the many social and political changes of the last two thousand
years.35
I shall return in a moment to the role of ethnic exclusion in
ensuring ethnic persistence. For the present I want to look at the
other side of the coin: ethnic dissolution. We say how ethnies can be
dissolved through fission or proliferation. But in a sense the ethnic
community remains in some form in such cases — smaller, perhaps,
or reduplicated, but none the less still 'in the field'. Can we then
speak of ethnic extinction — the disappearance of an ethnie, not just
in the form it possessed until that point but in any form?
I think we can if we hold to the historical, cultural and symbolic
criteria of ethnic identity I have been employing. There are two
main kinds of ethnic extinction in the full sense: genocide and
ethnocide, which is sometimes - at times misleadingly — called
'cultural genocide'. In one sense genocide is a rare and probably
modern phenomenon. It includes those cases where we know that
mass death of a cultural group was premeditated and the basis of
that targeting was exclusively the existence and membership of that
cultural group. Nazi policies towards the Jews and a part of the
Gypsies were of this kind; so perhaps were European actions towards
the Tasmanian Aborigines, and the Turkish actions in Turkish Ar-
menia.26
Other policies and actions were genocidal in their consequences
rather than their intentions; such ethnic destruction occurred
when the American Whites encountered the American Indians, and
when the Spanish conquistadors encountered the Aztec and other
Indian populations of Mexico (though here disease played a larger
part). In these cases ethnic extinction was not deliberately aimed at,
yet no attempt was made to mitigate those policies whose sideeffect
was genocidal. These genocidal actions need to be distinguished
again from large-scale massacres like those by the Mongols
in the thirteenth century or in modern times by the Soviets and
Nazis of selected populations (for example, the Katyn massacre or
the reprisals of Lidice and Oradour), which are designed to break a
30
THE ETHNIC BASIS OF NATIONAL IDENTITY
spirit of resistance by terrifying the civilian population or rendering
it leaderless.27
The interesting point about genocide and genocidal actions, at
least in modern times, is how rarely they achieve their stated goals
or unintended consequences. They rarely extinguish ethnies or ethnic
categories. In fact they may do the opposite, reviving ethnic cohesion
and consciousness, or helping to crystallize it, as they did with
the Aborigines' movement or Romany Gypsy nationalism. Perhaps
there are deep-rooted facets of modernity that both encourage and
preclude successful genocide (where success is measured by total
extinction), and this may have much to do with the conditions and
diffusion of nationalism. It may have been easier to destroy an ethnie
in pre-modern times. At any rate, when at last the Romans decided
to destroy Carthage once and for all they erased the city and massacred
three quarters of its population, selling off the rest into slavery.
Though vestiges of Punic culture persisted till the time of St Augustine,
the Carthaginians as a western Phoenician ethnie and ethnic
state were extinguished.28
The same fate awaited several peoples of the ancient world, including
the Hittites, Philistines, Phoenicians (of Lebanon) and Elamites.
In each case loss of political power and independence presaged
ethnic extinction, but usually through cultural absorption and ethnic
intermingling. These are cases of ethnocide rather than genocide,
despite the drama of the political events that precipitated them.
When he destroyed Susa and eliminated the Elamite state from
politics in 636 BC Asshur-bani-pal, king of Assyria, did not set
about exterminating every Elamite (the Assyrians in fact usually
deported the élites of the peoples they conquered). Yet so massive
was the act of destruction that Elam never recovered, new peoples
settled within its borders, and, though its language persisted into the
Achaemenid Persian period, no Elamite community or state reemerged
to sustain the myths, memories, values and symbols of
Elamite religion and culture.29
The fate of Assyria itself was even more swift and dramatic.
Nineveh fell in 612 BC to a combined onslaught of Cyaxares' Medes
and Nabopolassar's Babylonians, and her last prince, Asshur-uballit,
was defeated at Harran three years later. Thereafter, we hear little of
'Assyria'. Its gods were received by Cyrus back into the pantheon at
31
THE ETHNIC BASIS OF NATIONAL IDENTITY
Babylon, but there is no further mention of state or people, and
when Xenophon's army marched through the province of Assyria
he found all her cities in ruins with the exception of Erbil. Was this
a case of genocidal actions or even genocide? 3
°
It is unlikely. The goal of Assyria's enemies was destruction of
her hated rule. That meant destroying her major cities so that there
was no chance of a revival of her political fortunes. True, Nabopolassar
talked about 'turning the hostile land into heaps and ruins',
but this did not mean exterminating every Assyrian, even if this had
been feasible. Perhaps the Assyrian élites were evicted; but, in any
case, in terms of religion and culture they were less and less differentiated
from the Babylonian civilization they sought to emulate.
Besides, the latter days of the vast Assyrian empire witnessed severe
social divisions both in the army and the countryside, and considerable
ethnic intermingling in the empire's heartlands, and use of an
Aramaic lingua franca for commercial and administrative purposes
following a large influx of Arameans. Hence the ethnic distinctiveness
of the Assyrians was severely compromised well before
the downfall of the empire, and cultural syncretism and ethnic
intermingling helped to ensure the attenuation and absorption of
the Assyrian ethnic community and its culture by the surrounding
peoples and cultures.31
As with the Phoenicians, Elamites and others, the relatively swift
disappearance of an Assyrian culture and community must be seen
as an example of ethnocide. In the ancient world at least, destruction
of a community's or state's gods and temples was seen as the means
of destroying the community itself; that seems to have been the aim
of the Persians when they destroyed the Babylonian temples in 482
BC, and perhaps of the Romans when they destroyed the Temple in
Jerusalem in AD 70.1Z
The aim in all such cases was the eradication
of the group's culture, rather than the group itself, and it differs in
its intended effects from the much slower, unplanned processes of
cultural absorption which have undermined many small ethnic
categories and communities.
History is replete with instances of unintended cultural absorption
and ethnic dissolution. Engels, surveying the ethnic map of Europe
in 1859, referred to these dying ethnic cultures and communities as
so many 'ethnographic monuments', which he hoped would soon
32
THE ETHNIC BASIS OF NATIONAL IDENTITY
disappear to make way for the large capitalist nation-state. He has,
in fact, been largely disappointed. At the same time the diminution
of many former ethnies, and the attenuation of their sentiments, as
in the cases of the Occitanic, Sorbs, Wends and many others, demonstrates
these widespread processes of gradual absorption through
incorporation and fragmentation.
But, equally, they suggest the other side of the coin - the durability
of ethnic tics, the longevity of their cultures and the persistence
of collective identities and even communities over several centuries.
If ethnic boundaries and cultural contents undergo periodic change,
how shall we account for ethnic survival potential, sometimes across
millennia?
Again, it is useful to consider a well-known example. Jews trace
their ancestry to Abraham, their liberation to the Exodus, their
founding charter to Mount Sinai, and their golden age to (variously)
the Davidic and Solomonic kingdom or the era of the sages
in the late Second Temple period and after. These arc all myths in
the sense outlined above, and they retain their religious potency
today. But their potency is not only religious. They remain, even
for secular Jews, charters of their ethnic identity. Here, too, as with
the Greeks and Armenians, the Irish and Ethiopians, there is a felt
filiation, as well as a cultural affinity, with a remote past in which a
community was formed, a community that despite all the changes it
has undergone, is still in some sense recognized as the 'same' community.
To what is this sense of continuity, of shared memory and
of collective destiny owed?
The simple answer, that peoples survive in some form because they
are rooted in their homelands and enjoy a large measure of independent
statehood, will clearly not do in the Jewish case. The Jews have
been exiled from both for nearly two thousand years. Not that either
is unimportant to the Jewish sense of identity; but both figure more as
symbol than as living memory. Certainly this is true of statehood, the
Hasmonean being the last truly independent Jewish state — unless we
include the kingdom of the Khazars. The land of Israel was at times
more than a symbol of messianic restoration; groups of Jews made
their way there from time to time and founded synagogues. Yet here
too the yearning for Zion was often more spiritual than actual, a
vision of perfection in a restored land and city.33
33
THE ETHNIC BASIS OF NATIONAL IDENTITY
Another common view, which this time is directed specifically to
diaspora peoples, is that their survival depends on their ability to
find a distinct economic niche in host societies, usually as middlemen
or artisans, between military and agrarian elites and the peasant
masses. That Jews, Greeks and Armenians, like Lebanese and Chinese
traders, found such niches in medieval European and early modern
societies is not in question and neither is the role of such occupational
niches in reinforcing residential patterns and cultural segregation
where these already exist. What is at issue is the method by which
the category 'occupational niche' is separated from the nexus of
conditions that make up typical diasporas and assigned a prior casual
weight in ensuring ethnic survival and status. Rather, as Armstrong
has argued, archetypal diasporas that stem from religious and cultural
differences must be seen as a totality of interrelated aspects and
dimensions in which occupational segregation and middleman status
serves to reinforce and articulate, but not necessarily to ensure,
ethnic difference and survival. Certainly, in Moorish Spain Jews
held every kind of occupational position, but their ethnic survival
was bound up with more fundamental religious and cultural distinctions
from their neighbours.34
A more basic consideration stems from the earlier emphasis on
organized religion. In the case of diaspora communities, as of
sects-turned-ethnies like the Druse, Samaritans, Maronites and Sikhs,
religious rituals, liturgy and hierarchies have played a powerful
conserving role, ensuring a high degree of formal continuity between
generations and from community to community. Add to this the separating
power of sacred languages and scripts, texts and calendars, and
the apparent mystery of millennial diaspora survival appears soluble.
But there are difficulties here, too. For one thing this says nothing
about the shape, size or location of the surviving community. The
Samaritans, for example, were till quite recently heading for ethnic
extinction, because after centuries of decimation endogamy could
no longer replenish their numbers. In the case of the Beta Israel (or
Falasha) of northern Ethiopia the attrition of their numbers in war,
and the isolation of their craftsman community, might have spelt
absorption had it not been for a wider Jewish ethnic self-renewal
and the rise of Zionism and the state of Israel.35
This thesis also says nothing about the vitality of the community.
34
THE ETHNIC BASIS OF NATIONAL IDENTITY
Religion may become petrified and antiquarian, as did the Assyrian
state religion; in that case, as we saw, it contributed nothing to the
chances of ethnic survival. The same inner decay can be found in
later Roman religion, as in the Pharaonic religion of Ptolemaic
Egypt. In neither case could we hang an argument for ethnic survival,
let alone ethnic vitality, on any movement within the traditional
religion.3fi
Religion, then, may preserve a sense of common ethnicity as if in
a chrysalis, at least for a period, as was the case with Greek Orthodoxy
for the self-governing Greek Orthodox millet under Ottoman
rule. But unless new movements and currents stir the spirit
within the religious framework, its very conservatism may deaden
the ethnie or it may become a shell for an attenuated identity.37
Clearly, organized religion by itself is not enough. What then are
the characteristic mechanisms of ethnic self-renewal? I would single
out four such mechanisms:
1. Religious reform Having accepted the importance of organized
religion for ethnic survival potential, we need to consider the role
of movements of religious reform in stimulating ethnic self-renewal.
In the case of the Jews there are a number of instances. These range
from the Prophetic and Deuteronomic movements in eighth-century
and seventh-century BC Judah to Ezra's reforms in the mid-fifth
century BC, the rise of Pharisaism and Mishnaic rabbinism in the
second century AD, right up to the Chassidic and neo-Orthodox
movements of the eighteenth and nineteenth centuries. In each case
religious reform was intertwined with ethnic self-renewal; the community's
mode of renewal was religiously inspired.38
Conversely, failure of religious reform or petrified conservatism
may turn the modes of ethnic self-renewal elsewhere. This occurred
among the Greeks at the beginning of the nineteenth century. The
Greek Orthodox hierarchy in Constantinople became increasingly
remote from middle-class and popular aspirations, including those
of the lower clergy who supplied the revolt in the Morea with
some of its leaders. Here Greek aspirations found increasingly secular
ideological discourses for their goals.39
2. Cultural borrowing In the wider field of culture ethnic survival
finds sustenance not from isolation but from selective borrowing
35
THE ETHNIC BASIS OF NATIONAL IDENTITY
and controlled culture contact. Here again we can find an example
from Jewish history. The stimulus of Hellenistic culture, from the
time of Alexander on, provoked a lively encounter between Greek
and Jewish thought that, though it had fierce political repercussions,
strengthened through enrichment the whole field of Jewish culture
and identity.40
There are many other examples of the ways in
which external cultural stimuli and contacts have renewed the sense
of ethnic identity through selective cultural appropriation;
nineteenth-century Japan, Russia and Egypt afford well-known cases.
3. Popular participation Socially, too, we can discern modes of ethnic
self-renewal in the movements of social strata and classes. Of these,
the most relevant are popular movements for greater participation
in the cultural or political hierarchy. The great socio-religious popular
movement of the Mazdakites in fifth-century Sassanid Persia
renewed the severely damaged fabric of Sassanid Persian and Zoroastrian
community at the same time as it undermined the foundations
of the Sassanid state. This in turn provoked a repressive, but
also ethnically regenerative, movement under Chosroes I in the sixth
century, which included the codification of the basis of the Book of
Kings, a return to Iranian mythology and ritual, and a national revival
in literature, protocol, learning and the arts.41
The popular movements
in Judaism, from the Mosaic era to the Chassidim just mentioned,
also served to renew a demotic ethnie through enthusiastic
popular participation and missionary zeal. The same is true of various
popular movements in Islam, including its foundation and the movements
of Sunni or Shi'ite purification and messianism to this day,
such as Wahhabism, Mahdism and the Shi'ite revolution in Iran.42
4. Myths of ethnic election In many ways myths of ethnic chosenness
go to the heart of the modes of ethnic self-renewal and hence
survival. What we notice, first of all, is that ethnies that, for all their
cthnocentrism towards others, lacked such myths (or failed to instil
them in the general population) tended to be absorbed by other
communities after losing their independence. This may of course be
an argument from silence. Generally speaking, it is ethnies with
religious myths of ethnic election that possess the specialist classes
whose position and outlook are so heavily bound up with the success
and influence of election myths — and it is they who are often our
36
THE ETHNIC BASIS OF NATIONAL IDENTITY
only literary witnesses. Nevertheless, when we consider the fate of
many ethnies that possessed such classes but boasted no such myth
of ethnic election (as opposed to royal election), then, as the cases of
Assyria, Phoenicia and the Philistines reveal, it is clear that their
chances of ethnic survival were considerably diminished.
This, of course, merely puts the onus of explanation back on to
the conditions which foster and sustain myths of ethnic election.
Yet such a method short-circuits the process of ethnic survival
through exclusive election. For what the myth of election promises
is a conditional salvation. This is vital for grasping its role in survival
potential. Its locus classicus is found in the book of Exodus: 'Now
therefore, if ye will obey my voice indeed, and keep my covenant,
then shall ye be a peculiar treasure unto me from all the peoples;
for all the earth is mine; and ye shall be unto me a kingdom of
priests, and an holy nation.' 43
To see oneself as potentially 'an holy
nation' is to link chosenness indissolubly with collective sanctification.
Salvation is accessible only through redemption, which in turn
requires a return to former ways and beliefs, which are the means of
sanctification. Hence the recurrent note of 'return' in many ethnoreligious
traditions that inspire movements of both religious reform
and cultural restoration. Given the ineluctable subjectivity of ethnic
identification, this moral summons to re-sanctify the potential elect
provides a powerful mechanism for ethnic self-renewal and hence
long-term survival. This is certainly one key to the problem of
Jewish survival in the face of adversity, but we can also trace its
revitalizing effects among other peoples - Amharic Ethiopians, Armenians,
Greeks converted to Orthodoxy, Orthodox Russians,
Druse, Sikhs, as well as various European ethnies like the Poles,
Germans, French, English, Castilians, Irish, Scots and Welsh, to
name a few. So widespread a phenomenon clearly bears more
thorough-going investigation.44
'ETHNIC CORES' AND THE FORMATION OF NATIONS
Religious reform, cultural borrowing, popular participation and
myths of ethnic election: these are some of the mechanisms that,
along with location, autonomy, polyglot and trading skills and
organized religion, help to ensure the survival of certain ethnic
37
THE ETHNIC BASIS OF NATIONAL IDENTITY
communities across the centuries despite many changes in their
social composition and cultural contents. These cases again bring us
up sharply against the central paradox of ethnicity: the coexistence
of flux and durability, of an ever-changing individual and cultural
expression within distinct social and cultural parameters. The latter
take the form of a heritage and traditions received from one generation
to another, but in slightly or considerably changed form, which
set limits to the community's outlook and cultural contents. A
certain tradition of images, cults, customs, rites and artefacts, as well
as certain events, heroes, landscapes and values, come to form a
distinctive repository of ethnic culture, to be drawn upon selectively
by successive generations of the community.
How do such traditions influence subsequent generations? In premodern
communities it is the priests, scribes and bards, often organized
into guilds and castes, who recount, re-enact and codify
traditions. Often as the only literate strata, and being necessary for
intercession with divine forces, priests, scribes and bards achieve
considerable influence and prestige in many communities. Organized
in their brotherhoods and temples and churches, they form a network
of socialization in the major towns and much of the surrounding
countryside - depending upon their degree of organization and
mental monopoly in the community's territory. Indeed, in many
ancient and medieval empires priesthoods and their temple and
scribal infrastructure formed indispensable partners in government
and/or rival centres of power to the Court and bureaucracy, especially
in ancient Egypt and Sassanid Persia.45
Even in diaspora communities we find the priests, rabbis and
doctors of law, organized along more or less centralized lines, forming
an encompassing network of tribunals and counsel, and endowing
far-flung enclaves with religious, legal and cultural unity in the
face of an often hostile environment. Especially among Jews and
Armenians, as Armstrong has demonstrated, this highly evolved
network of religious officials and institutions was able to ensure the
subjective unity and survival of the community and its historical
and religious traditions.46
It is through such unifying and embracing mechanisms that what
we may term 'ethnic cores' are gradually built up. These are fairly
cohesive and self-consciously distinctive ethnies which form the
38
THE ETHNIC BASIS OF NATIONAL IDENTITY
kernel and basis of states and kingdoms such as the barbarian regna
of the early medieval era. Among the kingdoms of the Franks,
Lombards, Saxons, Scots and Visigoths the sense of a community of
customs and common descent played a vital role, despite the fact
that many of their inhabitants did not belong to the dominant
ethnic community. Nevertheless, in popular perception, such regna
were seen as increasingly communal and possessed of a unifying
cultural basis.47
By the later medieval period these subjectively
unified communities of culture formed the core around which large
and powerful states erected their administrative, judicial, fiscal and
military apparatus, and proceeded to annex adjacent territories and
their culturally different populations. Under Edward I, for example,
the English (Anglo-Norman) state expanded into Wales, destroying
the Welsh kingdoms and bringing most Welshmen into the realm
as a peripheral cultural community under the domination of the
English state. Something similar happened in France under Louis
VIII to the pays d'oc, notably the County of Toulouse, at the time of
the Albigensian Crusade.48
Locating such ethnic cores tells us a good deal about the subsequent
shape and character of nations - if (and when) such nations
emerge. It helps us to answer in large part the question: who is the
nation? and to some extent: where is the nation? That is to say, a
state's ethnic core often shapes the character and boundaries of the
nation; for it is very often on the basis of such a core that states
coalesce to form nations. Though most latter-day nations are, in
fact, polyethnic, or rather most nation-states are polyethnic, many
have been formed in the first place around a dominant ethnie, which
annexed or attracted other ethnies or ethnic fragments into the state
to which it gave a name and a cultural charter. For, since ethnies are
by definition associated with a given territory, not infrequently a
chosen people with a particular sacred land, the presumed boundaries
of the nation are largely determined by the myths and memories of
the dominant ethnie, which include the foundation charter, the
myth of the golden age and the associated territorial claims, or
ethnic title-deeds. Hence the many conflicts, even today, for
sundered parts of the ethnic homeland, — in Armenia, in Kosovo, in
Israel and Palestine, in the Ogaden, and elsewhere.
Both the close relationship and the differences between the
39
THE ETHNIC BASIS OF NATIONAL IDENTITY
concepts of ethnie and nation and their historical referents may also
be seen by recalling our definition of the nation. A nation, it was
argued, is a named human population sharing an historic territory, common
myths and historical memories, a mass, public culture, a common economy
and common legal rights and duties for all members. By definition the
nation is a community of common myths and memories, as is an
ethnic. It is also a territorial community. But whereas in the case of
ethnies the link with a territory may be only historical and symbolic,
in the case of the nation it is physical and actual: nations possess
territories. In other words nations always require ethnic 'elements'.
These may, of course, be reworked; they often are. But nations are
inconceivable without some common myths and memories of a
territorial home.
This suggests a certain circularity in the argument that nations are
formed on the basis of ethnic cores. There is, indeed, considerable
historical and conceptual overlap between ethnies and nations. Nevertheless,
we are dealing with different concepts and historical formations.
Fthnic communities do not have several of the attributes of
the nation. They need not be resident in 'their' territorial homeland.
Their culture may not be public or common to all the members.
They need not, and often do not, exhibit a common division of
labour or economic unity. Nor need they have common legal codes
with common rights and duties for all. As we shall see, these attributes
of nations are products of particular social and historical conditions
working upon antecedent ethnic cores and ethnic minorities.
On the other side of the picture we should note the possibility of
forming nations without immediate antecedent ethnie. In several
states nations are being formed through an attempt to coalesce the
cultures of successive waves of (mainly European) immigrants — in
America, Argentina and Australia. In other cases states were formed
out of the provinces of empires which had imposed a common
language and religion, notably in Latin America. Here, too, creole
élites began a process of nation-formation in the absence of a distinctive
ethnie. In fact, as nation-formation proceeded it was found
necessary to fashion a distinctively Mexican, Chilean, Bolivian, etc.
culture, and to emphasize the specific characteristics - in terms of
separate symbols, values, memories, etc. — of each would-be
nation/9
40
THE ETHNIC BASIS OF NATIONAL IDENTITY
The dilemma is even sharper in sub-Saharan Africa, whose states
were created, if not deliberately across ethnies, at least with little
reference to them. Here the colonial states had to foster a purely
territorial patriotism, a sense of political loyalty to the newly created
states and their embryonic political communities. In the independent
states born of these territorial communities several ethnies, ethnic
fragments and ethnic categories were drawn together by political
regulation and social boundaries that had come to include previously
unrelated groups in the post-colonial political system, and had
brought them, even against their will, into a new struggle for scarce
resources and political power. In these circumstances the ruling
élites, who may often have been recruited from a dominant ethnie
or coalition of ethnic groupings, were tempted to fashion a new
political mythology and symbolic order not only to legitimate their
often authoritarian regimes, but also to head off threats of endemic
ethnic conflict and even movements of secession. In these cases the
state is utilized to fashion the 'civil religion' whose myths, memories,
symbols and the like will provide the functional equivalent of a
missing or defective dominant ethnie. So the project of nationformation
in sub-Saharan Africa suggests the creation of the components
of a new ethnic identity and consciousness that will subsume,
by drawing together, some of the loyalties and cultures of the
existing ethnies. At least that has been the national 'project' of many
Africian and Asian élites.50
This means that the relationship of modern nations to any ethnic
core is problematic and uncertain. Why then should we seek the
origins of the nation in pre-modern ethnic ties when not every
modern nation can point back to an ethnic base? There are, I think,
three reasons why we should do so.
The first is that, historically, the first nations were, as we shall sec,
formed on the basis of pre-modern ethnic cores; and, being powerful
and culturally influential, they provided models for subsequent cases
of the formation of nations in many parts of the globe.
The second reason is that the ethnic model of the nation became
increasingly popular and widespread not only for the foregoing reason,
but also because it sat so easily on the pre-modern 'demotic' kind of
community that had survived into the modern era in so many parts of
the world. In other words the ethnic model was sociologically fertile.
41
THE ETHNIC BASIS OF NATIONAL IDENTITY
And third, even where a nation-to-be could boast no ethnic
antecedents of importance and where any ethnic tics were shadowy
or fabricated, the need to forge out of whatever cultural components
were available a coherent mythology and symbolism of a community
of history and culture became everywhere paramount as a
condition of national survival and unity. Without some ethnic
lineage the nation-to-be could fall apart. These three factors in the
formation of nations provide the point of departure for our analysis
in the next two chapters.