ANNALES • Ser. hist, sociol. • 9 • 1999 • 1 (16)
original scientific paper UDC 376.7: 323.15 (429)
received: 1998-10-15
BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY
Colin H. WILLIAMS
Cardiff University, Department of W elsh, UK-Cardiff CF1 3XW , P .O .Box 910
A B S T R A C T
The author traces the development o f bilingual education in Wales and suggests several challenges and 
opportunities which have influenced the manner in which the Welsh people have struggled to reproduce their 
language and culture. Education is viewed in tandem with the family and the community as the principal agencies 
by which language reproduction is maintained', but unlike the other two pillars o f cultural struggle, the parameters o f 
educational life in Wales have been increasingly marked by a divergence from a British state norm. The 
development o f a national curriculum for Wales following the Educational Reform Act o f 1988 together with other 
key legislative changes such as the Welsh Language Act, 1993 and the Government o f Wales Act 1998 have 
changed the politico-legal context o f bilingualism and have insitutionalised bilingual education as a significant and 
growing element o f Welsh national life.
Key words: education, bilingualism, language minorities, Welsh people
SC U O LA  BILINGUE IN FAVORE DELLA DEM O CRAZIA  EUROPEA
SINTESI
L'autore tratta lo sviluppo della scuola bilingue nel Galles e presenta diverse sfide e occasioni che hanno 
influenzato gli sforzi dei gallesi per riprodurre la propria lingua e la propria cultura. La scuola è trattata assieme alia 
famiglia e alla comunità, quali principali ambienti in cui si conserva la riproduzione lingüistica, ma a differenza di 
queste due colonne della battaglia cultúrale, i parametri dell'educazione nel Galles sono stati sempre più 
caratterizzati da un allontanamento dalle norme britanniche. Lo sviluppo del piano nazionale d'insegnamento del 
Galles, seguito alla Legge sulla riforma scolastica del 1988, assieme ad altri fondamentali cambiamenti legislativi 
corne la Legge sulla lingua gállese del 1993 e la Legge su! governo del Galles del 1998, hanno cambiato la 
concezione politico-giuridica del bilingüismo ed hanno istituzionalizzato la scuola bilingue quale importante 
elemento della vita nazionale gállese.
Parole chiave: istruzione, bilingüismo, minoranze linguistiche, gallesi
89
ANNALES • Ser. hist, sociol. ■ 9 • 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
INTRODUCTION
Thank you for the invitation to address you and for 
the opportunity to trace the key issues which have influ­
enced the development of bilingual education in Wales 
and relate this process to the wider currents of demo­
cratic change in Europe.
Bilingual education in the service of European de­
mocracy might seem like a grandiose title to describe 
what most of you, as bilingualducators, are involved 
with day by day in teaching children of all ages the ba­
sics of the set curriculum. But I want to remind you that 
most teachers are understandably caught up in the de­
tails of what they do within the profession, together with 
the profound responsibility of looking after the interests 
of individual children, to worry about the broader con­
text of policy.
Education has always been the prime socialisation 
agency of any advanced modernising state. Accordingly 
I want to argue that teachers are in the front line of bi­
lingualism and multiculturalism in Europe. Curriculum 
and language planners may set the agenda, but it is the 
teaching profession which is best armed to deal with the 
particular needs of children and students within an in­
creasingly multi-cultural Europe. In consequence, the 
teaching profession constitutes the core of any society's 
information network. But from time to time, it can be 
unfairly burdened should popular expectations presume 
that it, above all other elements, is primarily responsibly 
for the transmission of a national language and culture.
Educational planners used to dealing with curricu­
lum development and teacher training issues are now 
also becoming more sensitive to the issues of personality 
development and the social psychological adjustment of 
children with mixed identities. Thus children experienc­
ing bilingual education ask many of the standard ques­
tions which intrigue all humans, such as:-
Who am I? Where do I come from? How do I fit in to 
this complex and often mystifying world? Will I be val­
ued when 1 leave school? How do I relate to other peo­
ple who are not quite the same as me?
But in addition, having been exposed to primary in­
struction in two or more languages, pupils at bilingual 
schools also engage in self-reflective assessment of the 
relationship between their individual bilingualism and 
the collective social identity. Sometimes it is very obvi­
ous that significant groups within multicultural societies 
do not accept or respect the particular sort of education 
which bilingual or multilingual children are receiving. 
Tension and social conflict often accompany initial at­
tempts at establishing a bilingual educational system, 
especially if the rationale of the reform is to compensate 
for the previously discriminatory experience suffered by 
a particular section of society within the state educa­
tional system.
The right to be taught in one's mother tongue or
preferred language is born in struggle and is realised in 
personal human freedom. Bilingual education is not a 
natural government planned service. It is nearly always 
a struggle between a beleaguered minority and a he­
gemonic state majority. However, there are many ways 
of wresting minority rights from states. We can struggle 
with a smile or we can struggle with a fist, but it is still a 
struggle. So we must always remain cool calm and 
hopefully sophisticated in justifying who we are and 
what we do as professionals interested in bilingualism 
and bilingual education.
So the first thing I want to do is to remind you that 
you are part of a much larger reality, namely the con­
struction of a Europe which is more democratic, more 
sensitive to the needs of its constituent cultures and less 
enamoured with the demands of the putative nation­
state and its historical agenda of subjugating dissenting 
minorities.
Some of you will be most interested in the way in 
which bilingual education delivers a better quality of 
education within specific subjects, such as nurturing 
Slovene and Italian language skills, History or Geogra­
phy. Others will add to this interest the way in which 
the bilingual skills so assiduously learned within the 
class-room can be honoured and realised within the 
wider society. Yet others will be interested in the Slo­
vene diaspora population and how bilingual education 
can assist them to keep in touch with their national cul­
ture and contribute in a very positive way to their group 
maintenance and identity.
For analysts, such as myself, it is the whole of this 
complex picture which is of interest. We ask fundamen­
tal questions such as what is bilingual education for? 
How does it relate to the socio-economic needs of the 
nation and of the international system? We then seek to 
ascertain which policy options are most likely to be 
supportive of a more tolerant, plural society. For the de­
gree, quality and social impact of bilingual education in 
some societies can be a very sensitive temperature gauge 
of democracy. Let me illustrate by reference to Wales.
EDUCATION AND THE STRUCTURE OF BILINGUAL 
WALES
When bilingual schools were established in Wales in 
the fifties and sixties they were criticised for being lin­
guistically and educationally divisive and they were ac­
cused of serving the interests of a small cultural na­
tionalist elite. Such has been their success and the con­
sequent socio-political changes within the attitudes of 
many sections of society, that bilingual schools are now 
championed as models of good education and as essen­
tial elements in preparing citizens to participate fully 
within a bilingual democracy and a multicultural Euro­
pean Union. I believe that the Welsh experience offers 
many lessons as to how a committed, non-violent
90
ANNALES • Ser. hist, sociol. • 9 • 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
popular pressure for social reform can transform centu­
ries of state-inspired discrimination and lead to a more 
plural, democratic and positive bases for civil society.
Welsh society is characterised by a unilingual ma­
jority and a bilingual minority whose bilingualism adds 
an extra frissance to life which at times is enrapturing 
and at others downright uncomfortable. Historically the 
hearth, the farm and the chapel have been the tradi­
tional domains which sustained a Welsh-medium net­
work of agencies of language reproduction. This 
reached its zenith in the late nineteenth century when 
the institutions of Welsh nonconformity created a paral­
lel 'social totality1 which enabled Wales to cope with 
the huge upheaval of industrialisation and urbanisation, 
without necessarily losing its language or culture.
At the beginning of the twentieth century (1901) 929, 
824 were recorded by the census as able to speak 
Welsh and 1,577.141 able to speak English. The Welsh 
population was comprised of 928,222 monoglot English 
residents, 280,905 monoglot Welsh residents and 
648,919 bilingual Welsh-speakers who, also spoke Eng­
lish, because English was becoming pervasive and ad­
vantageous in most spheres of life. The vast majority of 
these bilingual residents were descendants of unilingual 
Welsh speakers. However, bilingualism was not just a 
one way process of Welsh speakers acquiring English as 
their second language then as now. Some were the chil­
dren of non-Welsh migrants who had been attracted to 
the resource-rich coalfields, and had been socialised 
within the multilingual melange of the industrial cruci­
ble and had learned Welsh en passant Others were the 
children of slate quarry workers in North Wales, part of 
an immigration flow from the Midlands and North of 
England who integrated into Gwynedd's predominantly
Welsh - medium socio-cultural environment.
Successive inter-censal decline has been the marked 
feature of census evidence on Welsh-speaking ever 
since 1911, when 977,400 persons were returned as 
able to speak Welsh, 190,300 of whom were monoglots. 
Since this peak of language intensity the number of 
Welsh-speakers has consistently declined, to reach its 
present low of 590,800 hardly any of whom are adult 
monoglots. In proportional terms this represents a de­
cline from 43.5% of the total Welsh population in 1911 
to only 18.7% in 1991, a loss of 24.8%. Various expla­
nations have been advanced both for the absolute and 
relative rate of decline, with particular attention having 
been focused on the inter-war period when stigmatisa­
tion, a collapse in confidence and depression-induced 
population out-migration encouraged widespread lan­
guage shift. The period 1921 -1939 was the crisis turning 
point for Welsh. A generation was denied the opportu­
nity to learn Welsh reflecting both parental rejection of 
the language and an unresponsive school system. This 
powerful combination of forces reflected the apogee of 
British imperial values and attitudes, which deemed that 
Welsh was irrelevant in a modernising world order. 
Since the early 1950s the rate of decline has been more 
moderate, reflecting the already emasculated levels of 
Welsh fluency.
The most significant element of language decline has 
been the collapse of the monolingual population. The 
fact that all Welsh speakers are bilingual does change 
the social-psychological context of language production 
and reproduction, for unlike many other examples of 
diglossic societies, individual and societal bilingualism 
in Wales does not vary tremendously.
Tab. 1: Proportion o f population speaking Welsh, by county, 1921-1991. 
Tab. 1: Delež valižansko govoreče populacije po okrajih 1921-1991.
Percentage o f all persons 
speaking Welsh
Percentage o f all persons 
speaking Welsh only
1921 1931 1951 1961 1971 1981 1991 1921 1931 1951 1961 1971 1981
Wales 37.1 36.8 28.9 26.0 20.8 18.9 18.7 6.3 4.0 1.7 1.0 1.3 0.8
Counties
Clwyd 41.7 41.3 30.2 27.3 21.4 18.7 18.2 5.8 3.4 1.3 0.8 1.4 0.8
Dyfed 67.8 69.1 63.3 60.1 52.5 46.3 43.7 15.3 9.6 4.1 2.4 2.4 1.6
Gwent 5.0 4.7 2.8 2.9 1.9 2.5 2.4 0.2 0.1 0.1 0.2 0.1 0.1
Gwynedd 78.7 82.5 74.2 71.4 64.7 61.2 61.0 28.1 22.1 9.1 5.2 4.9 2.6
Mid
Glamorgan
38.4 37.1 22.8 18.5 10.5 8.4 8.5 2.3 0.8 0.3 0.4 0.8 0.5
Powys 35.1 34.6 29.6 27.8 23.7 20.2 20.2 6.1 3.9 1.6 0.9 1.0 0.9
S. Glamorgan 6.3 6.1 4.7 5.2 5.0 5.8 6.5 0.2 0.1 0.1 0.1 0.4 0.2
W. Glamorgan 41.3 40.5 31.6 27.5 20.3 16.4 15.0 3.6 1.3 0.5 0.5 1.0 0.8
Source / Vir: Census 1981 W elsh Language in W ales, table 4, p. 50.
91
ANNALES • Ser. hist, sociol. • 9 • 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
Changes between 1981 and 1991 were fairly mini­
mal and in that respect, the bilingual population appears 
fairly stable, predictable and likely to grow. If we disag­
gregate the data by age groups the 1991 census reveals 
that there have been significant increases in the 3-15 
age group (Table 2) a consolidation of the 16-44 age 
group and an expected decline in the two older age 
groups. These trends are likely to be maintained in suc­
cessive decades which suggest that the demographic 
future of Welsh is brighter than at any other time in re­
cent history. Much will depend upon how social agen­
cies and the National Assembly implement current so­
cio-linguistics reforms and how strong will be the func­
tional motivation of future generations to exercise their 
language choice and social power.
Tab. 2 : Welsh-speakers, 3-15 years.
Tab. 2 : Valitansko govoreci, 3-15 let.
Thus towards the end of the century bilingualism 
represented more of a social choice for c. 590.800 in­
dividuals, who switch language by domain, by inter­
locutor and by whim as the opportunity allows. How­
ever, the chief problem facing those who wish to use the 
Welsh language as their primary means of communica­
tion is to optimise the conditions wherein a genuine free 
choice of languages may be exercised in all domains of 
social life.
Although the census is the single most comprehen­
sive data set available it does not probe very deeply into 
the social context or use of bilingualism. Thus we need 
to supplement its findings by non-census based social 
survey analysis best represented by the Welsh Office 
'Welsh Social Survey' (1993). This contains details of 
19,056 households interviewed between September and 
December 1992. The survey revealed that Welsh speak­
ers represent 21.5 per cent of the total population (Table
3). If we disaggregate this ability factor we find that the 
highest incidence is in the youngest age range, 3-15, 
with 32.4 per cent of the population fluent in Welsh. 
The proportion drops dramatically in the age range of 
16-29, at 17.8 per cent, and falls further to 16.7 per cent 
for the next age range 30-44. The figures rise to 18.7 per 
cent for the age range 45-64 and reaches 24.2 per cent 
for the 65 and over range. Clearly this bodes well for the 
future, but in- and out-migration, marriage patterns and 
a host of other reasons preclude any firm prediction that 
this youngest cohort will necessarily maintain such rea­
sonable levels of fluency into adulthood. We need to 
know far more about first and second language patterns 
and in this respect the survey has anticipated this need 
by identifying 55.7 per cent of Welsh speakers who 
considered it to be their mother tongue. They represent 
12 per cent of the national population.
The balance between first and second language 
speakers is a delicate issue. Very often one hears about 
the need to encourage language reproduction within 
predominantly Welsh speaking families and communi­
ties. However, language production through the educa­
tion system rather than language reproduction through 
community socialisation seems to characterise the 
younger elements of the population. This is revealed 
clearly in Table 3 where the school factor is very evi­
dent for only 27 per cent of the total Welsh speakers in 
the 3-15 age range considered Welsh to be their mother 
tongue. Presumably the majority consider their Welsh 
fluency to be a school acquired skill which does not 
invalidate any of their abilities to use the language, but 
neither does it assume that for such children Welsh is 
the first, instinctive language of daily life. Interesting 
trends are revealed in Table 3 for each successive age 
cohort recorded higher proportions of mother-tongue 
speakers, reaching a peak of 79.3 per cent for the 65 
and over group. The orthodox explanation would be 
that the older age groups learned Welsh at home within 
the family and for some linguists this is a significant 
feature for their use of Welsh is likely to be natural, 
richer, more idiomatic and colloquial than the rather 
formal, English-influenced style and patterning of 
younger Welsh speakers. This raises difficult questions 
of interpretation, for it might be that in terms of vocabu­
lary and domain confidence the quality of Welsh spo­
ken by the youngest group night in general be superior 
to that of the eldest group, even if it is less idiomatic. It 
may also be suggested that for the younger age groups 
their language loyalty/affiliation may not prove to be as 
resolute in the future, if Welsh represents for them a 
predominantly second language: a useful means of 
communication rather than an automatic first choice 
language of expression.
Percentage o f age 
group able to 
speak Welsh
Numbers able to 
speak Welsh % Change
County 1981 1991 1981 1991 1981-1991
Clw yd 18.6 27.9 13796 18167 31.7
Dyfed 40.3 47.7 23163 25811 11.4
Gw ent 2.3 4.8 1921 3490 81.1
Gw ynedd 69.3 77.6 28785 27889 -3.1
Mid G lam . 8.6 16.1 8906 14604 64.0
Powys 16.7 30.0 3284 5463 66.4
S. G lam . 7.4 11.9 5152 7690 49.3
W . G lam . 9.3 15.0 6064 8719 43.8
Source / V ir: Aitchison J. and Carter, H., 1993.
92
ANNALES • Ser. hist, sociol. • 9 • 1999 -1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
Tab. 3 : Welsh by age, 1992 (thousands).
Tab. 3 : Valižani po starosti', 1992 (v tisočih).
Age
category
Sample
size
Popu­
lation 
base 
(aged 3 
and  
over)
No. o f  
Welsh- 
spe- 
akers
Welsh- 
spe- 
akers as 
% o f  
popu­
lation
Mother- 
tongue 
spe­
akers as 
% o f all 
Welsh- 
spe­
akers
Welsh- 
spe- 
akers as 
% o f  
popu­
lation, 
1991 
Census
3-15 5094 486.2 157.4 32.4 27.0 24.3
16-29 4809 517.0 91.8 17.8 48.9 15.9
30-44 5741 585.2 97.5 16.7 60.8 14.8
45-65 6674 664.4 123.7 18.7 70.7 17.4
65+ 5335 498.0 120.4 24.2 79.3 22.6
Total 27653 2750.7 590.8 21.5 55.7 18.7
Source / V ir: W elsh Social Survey, 1992, W elsh office, HM SO, 
Cardiff.
Further evidence on self-assessed language ability is 
provided by the Social Survey which indicates that 
368,000 (13.4 per cent) are fluent in Welsh. A further 
94,900 (3.5 per cent) described themselves as able to 
speak quite a lot of Welsh, and 467,300 (17.0 per cent) 
described themselves as speaking only a small amount 
of Welsh. Thus 930,200 (33.9 per cent ) were able to 
speak a little Welsh and 462,900 (16.9 per cen t) were 
capable of speaking a considerable amount of Welsh. 
These figures are far higher than the normally cited 
Welsh-speaking population of c. 590,800 people and 
should prove useful as a rough guide to the potential 
Welsh-speaking mass for government services or con­
sumer/audience affairs.
Of those who claimed to be fluent, 80.5 per cent 
came from families where both parents spoke Welsh, 
7.2 per cent from where the mother was fluent, 4.6 per
cent from where the father was fluent and 7.7 per cent 
from families where neither parent was fluent.
Welsh speakers were asked to describe one state­
ment which best represented their current use of Welsh 
(Table 4). Interesting county variations are revealed with 
Gwynedd and Dyfed, as might be expected, recording 
the highest usage of Welsh at 79 per cent and 71.1 per 
cent respectively. Lower proportions are recorded for 
Powys and Clwyd at 51.5 per cent and 40.9 per cent, 
while West Glamorgan and the amalgamated category 
of the three counties of the south-east record 32.8 per 
cent and 33.1 per cent respectively. Interestingly whilst 
only 6.9 per cent of fluent Welsh speakers in West Gla­
morgan would claim that they rarely use the language as 
many as 15.3 per cent in the industrial south-east found 
little reason /opportunity to use Welsh (Table 4).
THREE PILLARS OF WELSH: FAMILY, COMMUNITY 
AND SCHOOL
The family
The data suggests that social context, family lan­
guage transmission and exposure to formal bilingual 
education are key factors in language reproduction. 
However, both the community and the family seem to 
be less powerful as agents of language reproduction 
than was true in the past. This implies that bilingual 
education, for both children and adults, will be a more 
significant agency of language production and repro­
duction than hitherto. Thus we need to specify with 
more precision just what are the relationships between 
these three elements and then move on to a considera­
tion of how formal language planning can translate the 
undoubted potential inherent within bilingual education 
into the reality of employing either Welsh or English as 
languages of choice in the widest possible range of so­
cial domains.
Tab. 4: Welsh-speakers1 curent use o f Welsh, 1992 (figures are percentages). 
Tab. 4 : Uporaba valižanskega jezika danes, 1992 (v odstotkih).
Welsh ability Wales Clwyd Dyfed Gwynedd Powys West Glam. Gwent, Mid 
& South 
Glam.
Rarely speak 
Welsh
4.5 8.9 1.9 1.3 5.7 6.9 15.3
Occasionally 12.0 21.7 7.4 5.5 17.9 26.8 21.8
Half & half 21.2 28.5 19.6 14.3 25.3 33.6 29.7
Most or all 
the time
62.3 40.9 71.1 79.0 51.1 32.8 33.1
S o u rce /Vir: W elsh Social Survey, 1992, W elsh Office, HM SO Cardiff.
ANNALES • Ser. hist, sociol. • 9 • 1999 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
Tab. 5 : Language attributes and household type. 
Tab. 5 : Jezikovne lastnosti in vrsta gospodinjstva.
Household types Percentage of 
households
Type 1 All households
(i) Households without Welsh speakers 73.6
(ii) Households with Welsh speakers 26.4
Type 2 Households with Welsh speakers
(i) Households wholly Welsh-speaking 53.6
(ii) Households partly Welsh-speaking 46.4
Type 3 Household composition and Welsh speech
a Households wholly Welsh-speaking
(i) With children 10.9
(ii) Without children 42.7
b Households partly Welsh-speaking
CO With Welsh-speaking children 18.9
(ii) With non-Welsh-speaking children 5.8
(iii) With no children 21.7
Type 4 Household size, composition and Welsh speech
a Households wholly Welsh-speaking
(i) With children 10.9
(ii) Single-person households 21.3
(iii) Without children 21.5
b Households partly Welsh-speaking
(i) Households with Welsh-speaking children-single speaker 6.2
(ii) Households with Welsh-speaking children-more than one Welsh speaker 12.7
c Households partly Welsh-speaking but with non-Welsh-speaking children
(i) Single Welsh speaker 4.9
(ii) More than one Welsh speaker 0.9
d Households partly Welsh-speaking without children
(i) Single Welsh speaker 18.6
(ii) More than one Welsh speaker 3.1
Note: in the above table each type is nested, ie it is a successive division of the previous type. Thus Type 2 is a subdivision of Type 1 (ii); 
Type 3 is a subdivision of Type 2 (i); Type 4 is a further subdivision of Type 3. Each type (1 to 4) therefore sums to 100.
Source / V ir: Aitchison and Carter (1997), p. 362.
Recent analysis of family/household composition 
patterns by Aitchison and Carter (1997) reveal signifi­
cant and possibly damaging trends. Their analysis of 
SARs data shows that an extremely high proportion of 
Welsh speakers are linguistically isolated within their 
home environments. Table 5 describes a nested 
hierarchy of four types of households based on the 
language ability of household members. A basic di­
stinction was drawn by Aitchison and Carter between 
Type 1 (i) - those households that have at least one 
Welsh speaker (defined as 'Welsh speaking house­
holds'), and Type 1 (ii) - those that have no Welsh-spe- 
akers. Twenty two per cent of all households belong to 
the first of these two types, but over half (51 per cent) of 
the households contain only one Welsh speaker within
them and many of these are elderly persons living alone. 
This does not bode well for the future.
A second distinction is that between households 
which are wholly or partly Welsh-speaking, Types 2(i) 
and 2(ii). Just over half of Welsh speaking households 
are wholly Welsh speaking (54 per cent) but they repre­
sent only 14 percent of all households in Wales. Wholly 
Welsh speaking households can be further sub-divided 
into those with and those without children (aged 3-17 
years) - Types 3a(i) and 3a(ii). Aitchison and Carter aver 
that such a pattern is disconcerting for Welsh as the data 
show that a very high proportion of such households 
have no children within them; furthermore almost half 
are single person households (Type 4(ii)). Similarly of 
partly Welsh speaking households, nearly two thirds (64
94
ANNALES • Ser. hist, sociol. • 9 • 1999 -1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
percent) have just a single Welsh speaker (Types 4a(i), 
4b(i) and 4c(i), the majority of whom are in households 
which have no children. Encouragingly 41 percent of 
the households that are partly Welsh speaking have one 
or more children who are able to speak Welsh. More 
sobering is the realisation that some 70 per cent of the 
Welsh speaking households have no Welsh speaking 
children within them. Welsh households are in the main 
linguistically fractured and structurally diverse in com­
position.
The community
Clearly by 1991 the monolingual reservoir had dis­
appeared and the family as a pillar of Welsh cultural 
reproduction was under strain, There is current concern 
that a second pillar, the communities of the northern 
and western heartland will also fragment irretrievably. 
The dominant theme in the socio-linguistics of Welsh 
has been the collapse of the territorial frame of the north 
and west. Fragmentation and decline is all too pressing 
a description of many Welsh-speaking comminuties. If 
this cultural resource base atrophies what then for the 
production and reproduction of the identity transmitted 
by the Welsh language?
The key issue is whether or not a virulent Welsh 
culture can survive without its own autochthonous heart­
land as a resource-base. The relationship between an 
autochthonous language group and its territorial base is 
often described as being natural or primordial. However, 
in Brittany, Euskadi, Friuli and Wales, there is a concern 
that the core area which sustained these distinctive 
language communities is atrophying, with dire 
consequences for the cultural reproduction and the 
maintenance of group identity. Changes in the New 
World Order and the globalisation of economic organi­
sation threaten to increase the pressure for uniformity 
and render small-scale language communities even more 
vulnerable and marginal to the interests of global culture. 
Clearly one would be naive to deny the rapid advance­
ments made in technologically-induced cultural chage. 
The whole premise of an integrated Europe depends in 
large part on the technical ability to realise a European 
Union without internal borders and trade barriers which 
is influenced by two processes.
The first, centrifugal, process is the decentralist 
challenge of the 'ethnic revival' which has characterised 
the past generation. The ethnic intelligentsia in many of 
the lesser-used language regions of Europe have stressed 
the organic authenticity of language. Their focus on the 
inviolability of the ethnic homeland has given a literal 
interpretation to the search for roots in the soil, com­
munity and landscape of one's own people. In Wales 
concepts derived from 'cydymdreiddiad tir ac iaith' (the 
interpenetration of land and language) have been given 
practical shape in planning policies aimed to bolster
indigenous Welsh-speaking communities through envi­
ronmental improvement and rural-economic diversifica­
tion Oones, J. R. 1966). The language movement has re­
discovered its 'ecological' heritage, and has re-packaged 
what were deemed to be rural community issues in the 
twenties and thirties as issues of 'cultural species' sur­
vival and as a local response to globalisation (Williams, 
1994). The gift of Wales, so to speak, to the diversity of 
the world is the preservation and conservation of its 
unique cultural heritage.
The second, centripetal, process is the internationali­
sation of language, described by Mackey (1986) as the 
'definitive liberation of language from its traditional 
bounds of time and space ... when language is no longer 
inevitably attached to spatial boundaries as it was in the 
past, when its speakers had to be limited to one or a few 
areas of the globe.' Telecommunication changes and 
mass migrations have empowered world languages such 
as English and French to perform critical inter-commu­
nicative roles which are historically predicated upon the 
economic power of Western capitalism. But they also 
derive immense power from the extension of digital 
technology, Mini-Satellite T .V ., and interactive comput­
ing systems to the farthest reaches of this globe. Tech­
nology further empowers such languages as essential 
means of communication and endow them with a cu­
mulative relative advantage vis a vis all other languages.
The key question then becomes whether smaller lan­
guages such as Welsh and Breton can benefit from the 
same liberation from time and space? Whether they too 
can be technologically empowered so as to compensate 
for the loss of territorial dominance? Further, we need to 
know whether there is a relationship between decentral­
ist localism and globalisation? If there is, how do both 
processes at either end of the continuum mediate what 
happens in mainstream society? Is the former process a 
primordial reaction to cope with the new threats and 
demands of the latter? Put in its most fundamental form 
does the increased internationalisation of English inevi­
tably threaten the ability of Welsh to compete and sur­
vive well into the next century?
English has been the official state language since 
mediaeval times and Welsh lost any formal role as a 
language of government and state affairs following the 
Acts of Union of 1536 and 1542. However, although 
there had not been a statutory obligation to recognise 
Welsh as an official language within Wales since the 
16th century, agencies of the state and of the established 
Church did employ Welsh in a wide range of domains 
in the intervening period because they were dealing 
with a predominantly monolingual population. How­
ever, it was not until the passing of the Welsh Courts Act 
of 1942 that the provisions prohibiting the use of Welsh 
by the Acts of Union were rescinded. Further legal rec­
ognition was given in the Welsh Language Act of 1967 
which offered an initial and inadequate definition of
ANNALES • Ser. hist, sociol. - 9 -1999 -1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
equal validity of English and Welsh in Wales. This was 
related mainly to the greater provision of Welsh in the 
Courts of Law and in legal proceedings generally, but 
did not extend into the wider sphere of public admini­
stration and formal bilingual rights.
During the sixties and seventies a number of statu­
tory and non-statutory bodies called for greater state 
support for the language. One initial response by the 
Welsh Office was the establishment in 1977 of the 
short-lived Cyngor yr laith Gymraeg (The Welsh Lan­
guage Council). It also led to some limited financial 
support for Welsh language activities, both in the public 
arena and in education via specific provisions in two 
government acts passed during the latter part of the 
1970s. Under section 26 of The Development of Rural 
Wales Act 1976 , the Welsh Office provide support for 
Welsh language social activities, most importantly the 
work of Mudiad Ysgolion Meithrin (The Welsh Nursery 
Schools Movement, founded in 1971), Eisteddfod 
Genedlaethol Cymru (The National Eisteddfod of 
Wales), Cyngor Llyfrau Cymraeg (The Welsh Books 
Council) and via Menter a Busnes (Business and Enter­
prise), concerned with the promotion of a new spirit of 
enterprise in rural Welsh speaking Communities.
Bilingual education provision: the school system
With the secularisation of society and the break­
down of the relative homogeneity of rural communities 
an alternative, urban, formal set of domains has been 
constructed mainly in the south and east. The principal 
instrument of language production and reproduction in 
such environments is the Welsh medium school system 
which has done more than any other agency to promote 
the language and introduce it, and its related culture, to 
hitherto non-Welsh speaking families. Welsh-medium 
education across the curriculum now includes Maths 
and Science, Technology and Computing in addition to 
the conventional Humanities and Social Science sub­
jects. The nation-wide network of primary, secondary 
and tertiary Welsh-medium institutions, are actively so­
cialising a younger generation into participating in a 
fully functional bilingual society.
A relevant curriculum for Wales?
For the majority of the modern period Welsh educa­
tion was planned in tandem with educational provision 
for England, but over the last thirty years a separate and 
distinct Welsh system has evolved. The main currents of 
educational reform are summarised in Figure 1 which 
charts both those features which were held in common 
between England and Wales and those which contrib­
uted to a separate Welsh character. The process may be 
divided into three periods, 1944-1964 which saw the 
evolution of two types of curriculum, 1964-1988 which
saw a swing towards a teacher's informed curriculum, 
and the post 1988 period where both London and Cardiff 
sought to develop separate national curricula for England 
and Wales respectively. This latter period was not 
without its difficulties for any attempt to develop an 
independent educational body which would meet the 
indigenous needs of the Welsh electorate had to be 
tempered by state-wide currents of educational thought 
and policy, which in practice were unsympathetic to too 
many divergences between English and Welsh policy.
Let us illustrate such tensions, which incidentally 
face most minority bilingual educational sectors, by ref­
erence to the experience of the limited-life experiment 
of the Committee for the Development of Welsh-Me­
dium Education (known in Welsh as 'Pwyllgor Datblygu 
Addysg Gymraeg1 PDAG).
PDAG was established in 1987 and given a five-fold 
task, namely:
1) to provide a forum to discuss Welsh-language 
education policies;
2) to identify needs for development, priorities 
among those needs and means of meeting them;
3) to co-ordinate developments in Welsh-language 
education in order to ensure the optimum use of avail­
able resources;
4) to provide information and publicity;
5) to designate areas of research and development. 
(PDAG, 1988; Williams, 1994).
This was the first ever bilingual educational planning 
body in the UK. PDAG was created by the authority of 
the Welsh Office, but it specified that it should function 
within the purview of the Welsh Joint Education Com­
mittee (CBAG, Cyd-Bwyllgor Addysg Gymraeg) as a Lo­
cal Authority directed, but Welsh Office and County 
Council funded agency. Its brief was to chart the antici­
pated needs of Welsh-medium education in the statutory 
sector, but it quickly enlarged this brief by tackling both 
the nursery and post-school provision, envisaging itself 
as a body concerned with all aspects of such education 
from the cradle to the grave. As an innovative and pio­
neering body, it helped shape the priorities in the me­
dium term, collated valuable experience and informa­
tion, distributed government finances to support re­
source and learning development and generally acted as 
a mouth-piece for bilingual educational issues in Wales. 
As such, despite its small staff and budget, it was an ef­
fective forerunner of what may become a more compre­
hensive national educational institute.
In the period 1987-1992 PDAG, performed a signifi­
cant planning role in the field of bilingual education, 
focussing on strategic questions, the devising of research 
programmes to answer acute needs and the encourage­
ment of bilingual educators at all levels within the pro­
fession. However, in 1992 as a result of a major gov­
ernment re-think on the reform of the educational sys­
tem PDAG's future looked bleak. The debate centred
96
ANNALES • Ser. hist, sociol. • 9 • 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
England / Wales Wales
1944-64 Two types o f curriculum
Butler Education Act 1944
Sec Schools Exam Council 1946 147 grammar/24 technical/127 secondary modern
1948
Ysgol Gymraeg Llanelli
CBAG. Welsh Joint Education Committee
GCE O Level (16+) 1950
3.2% 1952 Ysgol Gymraeg Y Barri
1954 WYEC textbook scheme
1956 Ysgol Gian Clwyd
Finance Act 1958
(extending LEA powers) 1960 35% grammar/35% modern/12% comprehensive
Education Act 1962 Ysgol Uwchradd Rhydfelen
1963-86 The teacher's curriculum
School Council 1964 Secretary of State for Wales
Circular 10/65; CSE exam 1966 The Welsh committee of the Schools Council
5% GDP 1968 The Gittins Report
Plowden Report 
Black Papers 1970 Education transfered to Welsh Office
1972 Mudiad Ysgolion Meithrin (Nursery)
APU 1974 Local Government Reform
Callaghan's Great Debate 1976
Curriculum 11-16 (HMI) 1978
Framework for the School 1980 97% Comprehensive Schools
Curriculum Welsh grants from section 21
(DES) 1984 Development Committee of the schools curriculum
SCDC; SEC 
GCSE exam 1986 PDAG
1988-99 The centre rules-which centre?
National Curriculum; NCC, SEAC 1988 The Curriculum Council of Wales
LMS; direct grant schools 1990 History/Welsh Committees
3.3% GDP 1992 Geography/Arts/Music
SCAA 1994
6 Direct Grant Schools
Schools Examination and Assessment Authority
Constitutional Reform 1997 Government Wales Act
Devolution in UK 1999 National Assembly for Wales
Welsh core subject in all schools
Source / V ir: PD A G  (1993) W illiam s (1994) with recent revisions and additions.
Fig. 1: A relevant curriculum for Wales?
SI. 1: Ustrezen učni načrt za Wales?
upon two radically alternative conceptions of the need 
for such an agency. Defenders of the fledgling system, 
such as the Welsh Joint Education Committee, wished to 
see itself designated as the Curriculum and Assessment 
Authority in Wales thereby avoiding any fragmentation 
of PDAG's responsibilities which offered an holistic 
framework from nursery level education to adult educa­
tion. The WJEC claimed that it had a history of demo­
cratic accountability, an unparalleled record of success­
ful examination and curriculum assessment and a na­
tional network of experienced professional staff working 
in schools and colleges throughout the country. The 
status quo would thus provide a firm platform for ex­
pansion of PDAG's role and responsibilities within a
97
ANNALES • Ser. hist, sociol. - 9 - 1999 - 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
proven educational system.
In complete contrast the government's White Paper 
on Education duly 28th, 1992) proposed the establish­
ment of two non-departmental public bodies, answer- 
able to Ministers, and more in keeping with its philoso­
phy of centralised direction and control of the statutory 
education, 5 - 1 6 ,  sector, (see Figure 1). The Schools 
Funding Council for Wales (SFCW) and the Curriculum 
and Assessment Authority for Wales (CAAW) would thus 
become "the major authority in Wales on Welsh lan­
guage education" and accordingly on December 18th, 
1992 the Conservative Government's Welsh Office 
Minister, Sir Wyn Roberts announced the transfer of 
PDAG's responsibilities to CAAW, effective from late 
1994. Within CAAW a new Welsh Department would 
oversee the operation of the national Curriculum and 
concentrate its efforts on the statutory education sector.
Initially this development was viewed as an attack 
on the fairly radical and autonomous stance on bilingual 
education which PDAG had adopted. It also meant that 
very many outside specialists who had been drafted in 
to assist in the thinking and operation of PDAG would 
now be excluded from influencing government depart­
mental thinking and this was interpreted as an increase 
in bureaucratic unaccountability and a decrease in 
communal involvement in the bilingual educational 
process.
The government's response was in part to argue that 
bilingual educational needs would be satisfied by the 
greater institutionalisation of the Welsh language 
through the passing of the Welsh Language Act, 1993 
and its attendant strengthened Welsh Language Board.
In many respects, this new institutionalisation of 
Welsh consequent to the recent reforms promises to in­
fluence attitudes and behaviour patterns in a more con­
structive manner. The past decade has seen a more ac­
commodating reaction on behalf of unilingual English 
speakers and this bodes well for the future extension of 
services which seek to convince them that they also 
may benefit from sharing a bi-cultural community.
Flowever, the bilingual character of Wales is far 
more than the public, formal acknowledgement of lan­
guage rights, public services and educational opportu­
nities.
The school's role in promoting a bilingual society 
has increased following the reforms of the 1988 Educa­
tion Act which insisted that Welsh be a core subject in 
the National Curriculum. It is now possible to teach a 
wide range of subjects including Maths and Science, 
Design and Computing through the medium of Welsh. 
After 1994 , all save a few opt out schools in the secon­
dary sector, were obliged to teach Welsh to the pupils in 
the lower forms. This, in turn, has exposed a far greater 
number of Welsh youth to the language and culture of
their homeland and requires a significant investment in 
teachers and resources to be successful. In Higher Edu­
cation circles also there is a wide range of vocational 
and non-vocational courses available to full and part- 
time students, but again it must be emphasised that in 
such developments the numbers involved within par­
ticular courses are small. Even so the trend and direction 
of change is significant for it extends both the domain 
use and practical utility of bilingualism in society.
The structure of Welsh-medium education 1980-1999
During the period 1990-1997 the proportion and 
number of pupils in formally designated Welsh medium 
primary and secondary schools has grown very slowly. It 
is within the dominant, conventional English-medium 
sector where the greatest impact of educational reforms 
may be charted and analysed by reference to the follow­
ing suite of tables. Table 6, for example, charts the 
trends in maintained primary schools teaching through 
the medium of Welsh, 1990-1997. Of the 1,718 schools 
in 1990, designated Welsh-medium schools where 
Welsh was the sole or main medium of instruction ac­
counted for 25.9% (445) of the total. The number in this 
category had risen slightly to 26.7%, (449) schools by 
1997 and will witness a gradual growth over the next 
decade.
However, it is in the non-conventional Welsh-me­
dium sector that significant changes have been re­
corded. In 1990 50.7% (870) of schools had classes 
where Welsh was taught as a second language only. By 
1997 this proportion had risen to 67.6% (1,136). Most of 
the increase was due to the curriculum impact of the 
1988 Education Act and the social effects of the Welsh 
Language Act, 1993 which in effect abolished Category 
D schools (Table 6). Whilst in 1990 14.2% (244) of pri­
mary schools were not obliged to teach any Welsh, by 
1997 only one school was exempted from this statutory 
requirement.
An alternative method of measuring the impact of 
the reformed curriculum is to analyse changes in the 
numbers of children able to speak Welsh as a direct re­
sult of being exposed to the school influence in addition 
to any home or parental fluency in Welsh. Table 7 sug­
gests that over the period 1990-1997, about the same 
number of children (aged 5-11) (c. 14, 500) speak Welsh 
at home. But throughout this period there is a steady in­
crease in the number of children who can speak Welsh 
fluently but who do not speak it at home, from 15,181 
to 21,221. Again, there is a doubling of the numbers 
who can speak Welsh but are not completely fluent, 
from 30,753 in 1990 to 67,666 in 1997, and therefore a 
corresponding drop in the numbers who cannot speak 
Welsh at all from 155,796 to 124,682.
98
ANNALES • Ser. hist, sociol. - 9 - 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
Tab. 6: Maintained primary schools teaching through the medium o f Welsh (a). 
Tab. 6: Pouk v osnovni šoli s koncesijo v valižanščini kot učnem jeziku.
Number o f schools
1990/91 1993/94 1994/95 (b) 1995/96 1996/97
A Schools having classes where:
W elsh is the sole or main medium of instruction 
Percentage of schools
445
25.9
460
27.1
465 453 
27.5 26.8
455
27.1
449
26.7
B Schools having classes where:
W elsh is used as a medium of teaching for part of the
curriculum
Percentage of schools
116
6.8
108 120 
6.4 7.1
106
6.3
95
5.7
(B) Schools having classes of first and second language 
pupils where some of the teaching is through the medium  
of W elsh
Percentage of schools
36
2.1
(C) Schools having classes of second language pupils 
where some of the teaching is through the medium of 
W elsh
Percentage of schools
122
7.1
C Schools having classes where:
W elsh is taught as a second language only 
Percentage of schools
870
50.7
1,068
62.9
1,091 1,091 
64.5 64.5
1,109
66.0
1,136
67.6
D Schools where:
No W elsh is taught 
Percentage of schools
244
14.2
54
3.2
27 27  
1.6 1.6
11
0.7
1
0.1
Total schools 1,718 1,698 1,691 1,691 1,681 1,681
(a) At January each year. Prior to 1993/94 at September each year, includes grant maintained schools. The mode of instruction in 
primary schools varies w idely according to linguistic background and a school may have classes in more than one category. 
However, each school appears once only in this table, under an appropriate heading. Figures from 1991/92 onwards are not directly 
comparable with previous years because of a change classification, therefore, figures are shown separately for previous years.
(b) The method of classification changed in 1994/95: the first column for that year indicates the schools under the old classification with 
the second showing schools under the new classification. For details of W elsh classification for LEA and grant maintained primary 
schools see Glossary (Appendix III) (1998).
Tab. 7 : Maintained primary school pupils, aged 5 years and over, by ability to speak Welsh (a).
Tab. 7 : Govorna sposobnost v valižanskem jeziku pri učencih osnovnih šol s koncesijo> starih 5 ali več let.
1990/91 1993/94 1994/95 1995/96 1996/97
Number of pupils who:
Speak W elsh at home
Do not speak W elsh at home but who can speak it with fluency 
Speak W elsh but not fluently 
Cannot speak W elsh at all 
Total
14,827
15,181
30,753
155,796
216,377
14,328
18,647
64,631
124,197
221,803
14,632
19,608
60,907
128,837
223,984
14,343
18,300
63,356
131,003
227,002
14,656
21,221
67,666
124,682
228,225
Percentage of pupils w ho:
Speak W elsh at home
Do not speak W elsh at home but who can speak it with fluency 
Speak W elsh but not fluently 
Cannot speak W elsh at all 
Total
6.9
7.0 
14.1
72.0
100.0
6.5
8.4
29.1
56.0
100.0
6.5 
8.8 
27.2
57.5 
100.0
6.3
8.1
27.9
57.7
100.0
6.4
9.3
29.6
54.6 
100.0
(a) At January each year. Prior to 1993/94 at September each year. Includes grant maintained schools. These figures are derived from 
assessments made by headteachers.
Alternative evidence of the same structural change pupil distribution (Table 9). Note that there has been a
may be presented in relation to the organisation of general increase in the range and number of classes
teaching through the medium of Welsh in maintained taught through the medium of Welsh and a correspond-
primary schools, by class distribution (Table 8) and by ing absolute drop of four fifths in those classes wherein
99
ANNALES • Ser. hist, socio!. • 9 • 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
Tab. 8 : Organisation o f teaching through the medium o f Welsh in maintained primary schools, by type o f class (a). 
Tab. 8 : Organizacija pouka v valižanščini kot učnem jeziku v osnovnih šolah s koncesijo glede na tip razreda.
Num ber o f classes
1990/91 1993/94 1994/95 1995/96 1996/97
Classes where W elsh Is the sole or main medium of instruction 1,736 2,065 2,147 2,170 2,192
Classes w here W elsh is used as a teaching medium for part of the 
curriculum
503 411 421 309
Classes of first and second language pupils where some of the teaching is 
through the medium of W elsh 188
Classes of second language pupils where some of the teaching is through 
the medium of W elsh 764
Classes w here W elsh is taugh as a second language 5,804 7,318 7,591 7,876 8,262
Classes being taught no W elsh 2,455 1,280 1,032 804 550
All classes 10,947 11,166 11,181 11,271 11,313
(a) At January each year. Prior to 1993/94 at September each year. Includes grant maintained schools. Classes as taught (i.e. as organised 
for teaching). Figures from 1991/92 are not directly comparable with previous years because of a change in classification, therefore, 
figures are shown separately for previous years.
Tab. 9: Pupils by use o f Welsh in class, in maintained primary schools (a).
Tab. 9: Učenci osnovnih šol s koncesijo, ki uporabljajo valižanski jezik  v razredu.
N um ber o f  pupils
1990/91 1993/94 1994/95 1995/96 1996/97
Pupils in classes where W elsh is the sole or main medium of instruction 38.404 46,950 49,382 50,327 50,392
Pupils in classes where W elsh is used as a medium of teaching for part of 
the curriculum 12,062 10,303 10,906 7,649
First and second language pupils in classes where some of the teaching is 
through the medium of W elsh 3,780
Second language pupils in classes where some of the teaching is throuh 
the medium of W elsh 17,308
Pupils in classes where W elsh is taught as a second language 151,037 192,839 201,148 209,103 220,136
Pupils in classes being taught no W elsh 62,245 33,843 28,20 21,870 14,553
All pupils 272,774 285,694 289,036 292,206 292,730
(a) At January each year. Prior to 1993/94 at September each year. Includes grant maintained schools. Classes as taught (i.e. as organised 
for teaching). Figures from 1991/92 are not directly comparable with previous years because of a change in classification, therefore, 
figures are shown separately for previous years.
Welsh is taught, down from 2,455 in 1990 to only 550 
in 1997. Similarly there has been a significant growth in 
the number of pupils in classes where Welsh was the 
sole or main medium of instruction from 38, 404 in 
1990 to 50, 392 in 1997. Conversely there has been a 
sharp decline in the numbers of pupils in classes being 
taught no Welsh from 62, 245 in 1990 to 14, 553 in 
1997.
At the secondary level a similar picture obtains but 
we may trace the trend over a longer time period 1980- 
1997. Table 10 reveals that there has been a more 
structured and linguistically differentiated pattern of 
school type. The composite school category, where 
Welsh was taught as both a first and second language 
has shrunk, while there has been a corresponding in­
crease in the number of schools where Welsh is taught 
as a first language only, from the five pioneering schools
which existed in 1980 to the 18 such schools by 1997. 
A larger number of secondary schools are now classified 
as equipped to teach Welsh as a second language from 
116 in 1980 to 161 in 1997. As a consequence the final 
category of schools (N=35) where no Welsh was taught 
in 1980 has been eliminated from the classification by 
1997. Most of these 35 schools were either state-funded 
Catholic schools or secondary schools located within 
long-Anglicised areas of Wales, mostly in the border 
counties abutting England.
There is a clear relationship between progression 
through the secondary school system and lessening ex­
posure to Welsh, as revealed by Table 11. As Welsh is 
part of the core curriculum for all students in year 
groups 7, 8, and 9 it is not surprising that the numbers 
taught Welsh remain constant. However, once an ele­
ment of option choice for formal school-leaving exami­
100
ANNALES • Ser. hist, sociol. • 9•1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
nations in introduced in year 10 (aged c. 15+) there is a 
drastic falling away from c. 85.2% of all pupils to c. 
34.7%. An even more sharp decline takes place in the 
final years of schooling, during years 12/13 where only 
5.5 % and 5.0% of pupils are taught Welsh as a first or 
second language respectively.
That the position of Welsh medium education has 
been strengthened is a significant development in its 
own right. However, there has been a corresponding 
shift in the proportion of all pupils now exposed to the 
indigenous language, literature and culture of Wales. 
Some of these go on to be very active participants 
within Welsh-medium networks and public life. The vast 
majority do not, but having had such a long term expo­
sure to formal education through Welsh, the previous 
generations' suspicion and tension which surrounded 
the use of Welsh as a language of the minority has 
largely disappeared. The general effect of this is recor­
ded in much more positive attitudes towards bilingua­
lism and the construction of a bilingual society per se. 
However, beneath this positive trend there remains for 
many a grumbling doubt as to the real worth of bi­
lingualism, for it is argued that once pupils have left the 
artificial confines of the school classroom there is little
economic and instrumental justification for maintaining 
fluency in Welsh. Time will tell as to whether this 
judgement can still be made after the current period of 
institutional bilingualism has had its full impact. Outside 
the educational system many other pillars of Welsh 
culture are entering a more dynamic phase which also 
leads to increased language reproduction.
OTHER PILLARS OF THE WELSH-MEDIUM CULTURAL 
SYSTEM
Broadcasting is one acid test of the social use and 
adaptability of Welsh culture. Radio paved the way with 
a limited range of Welsh medium transmissions in the 
fifties and sixties, largely devoted to religious, children or 
daily life issues. In 1977-79 as a result of the devel­
opment of VHF wavebands, an English-medium Radio 
Wales and a predominantly Welsh-medium Radio Cym­
ru service was launched. The latter provides some 127 
hours per week, ninety of which are in Welsh, and are of 
a high quality. Radio Cymru might be said to act like a 
national community network for many, as it encourages 
audience participation to a greater extent than do its far 
more diverse and hence specialist English counterparts.
Tab. 10: Maintained secondary schools teaching Welsh (a).
Tab. 10: Srednje šole s koncesijo, kjer poučujejo valižanski jezik.
Num ber o f schools
1980/81 1990/91 1993/94 1994/95 1995/96 1996/97
Schools where W elsh is taught as both a first and second 
language 82 68 53 48 50 50
Percentage of schools 34.5 29.6 23.3 21.1 21.9 21.8
Schools where W elsh is taught as a first language only 5 11 13 18 17 18
Percentage of schools 2.1 4.8 5.7 7.9 7.5 7.9
Schools where W elsh is taught as a second language only 116 129 153 158 159 161
Percentage of schools 48.7 56.1 67.4 69.6 69.7 70.3
Schools where no W elsh is taught 35 22 8 3 2 0
Percentage of schools 14.7 9.5 3.5 1.3 0.9 0.0
Total 238 230 227 227 228 229
(a) At January each year. Prior to 1993/94 at September each year. Includes grant maintained schols.
Tab. 11: Pupils taught Welsh in maintained secondary schools, by year group and type o f course, 1996/97 (a). 
Tab. 11: Dijaki, ki se učijo jezik  v valižanskih srednjih šolah s koncesijo glede na letnik in vrsto programa, 1996/97.
Num ber o f pupils
Year group:
7 8 9 10 11 12/13 Total
Tought W elsh as a first language: 4,870 4,502 4,387 4,298 4,191 1,171 23,419
Percentage of pupils 13.3 12.8 12.5 12.2 12.1 5.0 11.7
Taught W elsh as a second language: 31,715 30,014 29,923 12,190 11,066 1,293 116,201
Percentage of pupils 86.4 85.1 85.2 34.7 32.0 5.5 58.0
Not taught W elsh at all 101 746 822 18,609 19,361 21,029 60,668
Percentage of pupils 0.3 2.1 2.3 53.0 55.9 89.5 30.3
Total 36,686 35,262 35,132 35,097 34,618 23,493 200,288
(a) At January 1997. Includes grant maintained schools.
101
ANNALES • Ser. hist, sociol. • 9 • 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
The greatest boost to the popular and technical use 
of Welsh in this post-modern, international era was of 
course the inauguration of the television service, Sianel 
Pedwar Cymru on the 1st of November, 1982. It had 
been preceded by some thirty years of intermittent and 
gradually expanding television output in Welsh by the 
BBC, TWW , Teledu Cymru and Harlech TV which had 
demonstrated the potential for a sustained independent 
channel to serve the needs of a bilingual audience. The 
absence of such a channel had clear implications, not 
only for language reproduction, but also for sustained 
dissatisfaction on behalf of the unilingual English- 
speaking majority in Wales. Up to 1982 some ten per­
cent of programmes were transmitted in Welsh, with the 
effect that those who preferred not to watch Welsh me­
dium output, had their television sets tuned permanently 
to English transmitters in the Mendips, Shrewsbury, 
Kidsgrove and Chester. In consequence one had the 
strange anomaly that very many households in Wales 
received their daily diet of regional news and accompa­
nying programmes from a neighbouring region across 
the border. This also limited the appeal and impact of 
English-medium programmes produced in Wales, and of 
course, diminished the potential revenue derived from 
commercial advertising. Both sides of the 'linguistic di­
vide1 were thus profoundly unhappy with the situation. 
However, identifying the problem is one thing, acting to 
redress it is quite another, especially when there are 
huge political and financial implications. Nevertheless 
at a number of key turns in the seventies, it became ob­
vious that there was a growing support in favour of a 
fourth channel being devoted in whole or in part to 
Welsh medium services. The 1974 Crawford Committee 
endorsed this view as did the Conservative manifesto 
pledge of 1979. However, within a few months of taking 
office the new administration withdrew their committ­
ment, preferring to improve the existing broadcasting 
arrangements. This policy change engendered the larg­
est mass protests witnessed in post-war Wales, with a 
plethora of social movements, political parties and non- 
aligned interest groups campaigning in tandem to force 
the government to honour its pledge. The focus of this 
campaign was the May 5th, 1980 decision of Gwynfor 
Evans, a former leader of Plaid Cymru, to fast to death 
unless the government announced the creation of S4C. 
To the great relief of many, on 1 7th of September 1980 
the government reversed its decision and Wales has 
subsequently enjoyed a popular and varied Welsh-me­
dium broadcasting service. Some thirty out of 145 hours 
per week are transmitted in Welsh, mainly at peak time. 
The. programmes reach a relatively high percentage of 
their target audience. S4C is a commissioning rather 
than a production body, and in consequence has 
spawned a network of independent film makers, anima­
tors, creative designers, writers etc. who can turn their 
original Welsh language programmes into English or
'foreign' languages for sale in the international media 
market place.Cardiff ranks second to London as a me- 
dia-production centre in the UK with all the technical, 
economic and post-production implications of a grow­
ing infra-structure in such a specialist industry.
Three issues dominate the current broadcasting de­
bate in Wales viz., (1) financial self-sufficiency versus 
subsidy; (2) the relaxation of boundaries inside pro­
grammes between the use of Welsh and English; and (3) 
the multicultural nature of S4C which transmits Euro­
pean soccer and sport, repackaged documentaries, 
soaps, quizzes and a host of other material all dubbed 
into Welsh. In part this is to attract new viewers to the 
channel and in part it reflects S4C's participation in the 
European Broadcasting Union and commercial market­
ing of international television material.
At a more voluntary level there is a very active net­
work of Eisteddfodau (competitive cultural festivals) 
which nurture school-based and community-based per­
formances of Welsh plays, or plays in translation, of 
musical items, poetry, craft work, art and design and 
scientific projects. This network starts at the local level 
and the successful competitors progress through inter­
vening stages to reach either the National Eisteddfod or 
the Urdd National Eisteddfod. During the present century 
it was the Eisteddfod system which acted as a vehicle for 
national culture, setting both the themes and the stan­
dards of popular representation of Welshness. More re­
cently the Urdd (The Welsh League of Youth) has re-in­
terpreted traditional Welsh mass culture by adding go- 
karting, tenpin bowling, discos, and surfing 'in Welsh'. 
An additional voluntaristic element is the adult learning 
of Welsh through Ulpan and related schemes which are 
geographically widespread and well subscribed. These in 
turn often feed Welsh clubs and social centres which 
may have sport, folk dancing or music as their focus but 
offer a wider entrée into the indigenous culture. Their 
children, and those of in-migrants, may attend Language 
Centres for English-only speakers, designed to speed up 
their integration into the local community. However, as 
in most unbalanced bilingual countries there are severe 
difficulties in reconciling the rights and obligations of 
indigenous citizens with those of incomers, many of 
whom are antagonistic or hostile to the requirement that 
their children attend a bilingual school. Thus Wales ex­
periences grass-roots parents movements in favour of 
extending bilingual education, and a lesser number, but 
often well organised, who oppose such an extension on 
the grounds that it limits their natural rights as British 
citizens.
Factors which affect the structure of the education 
system
Let me first pay tribute to the countless hundreds of 
dedicated individuals whose exemplary commitment to
102
ANNALES • Ser. hist, sociol. • 9 • 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
Wales and its youth have brought us to the promising 
plateau we occupy today, Despite all the accompanying 
social reforms, the education system remains the most 
fundamental agency for the promotion of bilingualism 
even if, as I have warned in the past, there is a danger 
that we over burden the education system as the saviour 
of our language.
The Education Act 1988 allows for a certain national 
differentiation of Wales from England, reflected in a 
separate national curriculum containing mandatory 
Welsh and variants of the manner in which English is 
taught among other subjects. A dualism reflecting 
greater flexibility in curriculum design and central 
National Assembly control is characteristic of the new 
order in Wales. The twin agencies of implementation 
are the School Funding Council for Wales and the 
Schools Examination and Assessment Authority. In a 
small country new systems are as good as the character 
of the personalities who will implement them, but 
structures and strategic relationships are also crucial. In 
consequence, after the opening of the National 
Assembly for Wales in May 1999, managers of the 
current structure are tentatively awaiting what new 
system may be devised by the Assembly's Education 
Committee in the new millennium.
A second uncertainty is the relationship between the 
Schools Funding Council for Wales and the Local Edu­
cation Authorities. I have suggested that it could be an 
unbalanced partnership, reflecting the emasculated role 
of local government during the Conservative admini­
strations of Prime Minister Margaret Thatcher and John 
Major (1979-1997). L.E.A.s have acted increasingly as 
intermediate agencies processing government initiatives 
in, for example, training and in dispensing formula 
funding to individual schools. But if the trend towards 
grant maintained and opted out schools had gathered 
pace, then the Common Funding Formula for grant 
maintained schools would have transformed L.E.A.s into 
dead letter post boxes, for they would reflect rather than 
initiate educational trends. In comparison with England, 
very few schools in Wales chose to opt out. So much of 
our good practice in bilingual education has its origins 
at the local level, developing county-wide infra-struc­
tures, that one might be forgiven for feeling pessimistic 
about current trends toward the central control of the 
curriculum, its funding and its strategy.
And yet to do so would be to miss-read an oppor­
tunity. Under the L.E.A. formula an element of increased 
power has been transferred to individual schools and to
local communities. Should the institutionalisation of 
Welsh continue apace, then both the educational re­
forms and the 'free market ' philosophy which drive it 
could benefit the Welsh language and culture, and that 
for a number of complex and unpredictable trends 
elaborated upon below which relate to the implications 
of passing the Welsh Language Act, 1993 with its 
strengthened Welsh Language Board and the establish­
ment of the National Assembly for Wales, 1999.
The Welsh Language Act 1993 and language planning
The current challenge is to realise a fully functional 
bi/multilingual society through creating new opportuni­
ties for language choice within the public, voluntary and 
private sector of the economy. Legislation is critical, not 
only in authorising linguistic rights, but also in establish­
ing the infra-structure wherein such rights can be exer­
cised without let or hindrance. Too often individuals 
and groups have a titular right to certain services, but 
such rights are held in abeyance because of a lack of 
commitment to honour language choice rights at the 
point of local contact.
The Welsh Language Act 1993 provided a statutory 
framework for the treatment of English and Welsh on the 
basis of equality and inaugurated a new era in Language 
Planning. Its chief policy instrument is the re-fashioned 
and strengthened Welsh Language Board, established on 
21 December 1993, as a non-departmental statutory 
organisation. It is funded by a grant from the Welsh Of­
fice, which in the year ending 31 March 1998 totalled 
£5,756,00. It has three main duties:
1. Advising organisations which are preparing lan­
guage schemes on the mechanism of operating the cen­
tral principle of the Act, that the Welsh and English lan­
guages should be treated on a basis of equality.
2. Advising those who provide services to the public 
in Wales on issues relevant to the Welsh language.
3. Advising central Government on issues relating to 
the Welsh language.
The Welsh Language Act 1993 details key steps to 
be taken by the Welsh Language Board and by public 
sector bodies in the preparation of Welsh language 
schemes. These schemes are designed to implement the 
central principle of the Act, namely to treat Welsh and 
English on the basis of equality. Since 1995 a total of 67 
language schemes have been approved including all 22 
local authorities. In 1998 notices had been issued to a 
further 59 bodies to prepare schemes.1
1 Under the spirit of the 1993 Act the Board has also developed partnerships with the 22 Unitary Authorities through Rhwydwaith 
(Network), with the W elsh Consumer Council, the W elsh Council for Voluntary Action and with a range of private sector 
organisations. During the financial year 1997-97 grants totalling £2,254,792 were distributed under the Board's main grants scheme 
to organisations as varied as the National Eisteddfod, the W elsh Books Council and Shelter Cymru (Welsh Language Board, 1998).
103
ANNALES • Ser. hist, sociol. - 9 • 1999 - 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
Aims and objectives
The Welsh Language Board's primary goal is to en­
able the language to become self-sustaining and secure 
as a medium of communication in Wales. It has set itself 
four priorities: 1) to increase the numbers of Welsh-spe- 
akers; 2) to provide more opportunities to use the lan­
guage; 3) to change the habits of language use and 
encourage people to take advantage of the opportunities 
provided, and 4) to strengthen Welsh as a community 
language.
In order to meet its first aim of increasing the num­
bers speaking Welsh it has focused its efforts on normal­
ising the use of Welsh among young people by seeking 
to ensure that the provision of Welsh-language and 
Welsh-medium education and training is planned in 
conjunction with the key players. It also seeks to ensure 
an appropriate level of provision to obtain Welsh-lan­
guage education services for young people, to formulate 
policies and effective initiatives which ensure the proper 
provision of public and voluntary services and to pro­
vide grants for initiatives which promote the use of 
Welsh among young people.
The Board's second objective is 'to agree measures 
which provide opportunities for the public to use the 
Welsh language with organisations which deal with the 
public in Wales, giving priority to those organisations 
which have contact with a significant number of Welsh- 
speakers, provide services which are likely to be in 
greatest demand through the medium of Welsh or have 
a high public profile in Wales, or are influential by vir­
tue of their status or responsibilities'.
A third objective is to change the habits of language 
use and encourage people to take advantage of the op­
portunities provided. This is done through an innovative 
marketing campaign, including attractive bilingual pub­
lic display signs, the development of a Welsh spell­
checker and on-line dictionary, a direct Welsh Link Line 
for queries regarding the Welsh language and language- 
related services, a language in the workplace portfo­
lio/file, a Plain Welsh campaign with excellent guide­
lines for writing Welsh, and other improvements to the 
infra-structure so necessary before a real language 
choice can be made by the general public.
The WLB's fourth objective is 'that Welsh-speaking 
communities be given the facilities, opportunities and 
the encouragement needed to maintain and extend the 
use of Welsh in those communities'. This aspect of lan­
guage planning is concerned with participation and 
community-level language empowerment. Because the 
conventional domains are unable to sustain the lan­
guage it has been posited that individuals are more at­
omised than ever before, thus rather than talk of Welsh­
speaking communities it is more realistic in many parts 
to talk of Welsh-speakers in the community.
An essential means of increasing the provision of
Welsh-medium services is to create a framework which 
is supportive of bilingualism; that is, the processes of 
realising the civil rights which are incorporated in the 
Language Act and in the respective charters on human 
rights. The core of this social change would be to 
deepen the fairly superficial bilingualism which cur­
rently obtains and to extend it to the working environ­
ment of institutions which offer a service to the public. If 
Welsh public services are serious about their response to 
the recent guidelines on developing bilingual policies, 
then they should provide a comprehensive bilingual 
service rather than give the impression that this is 
merely a compromise in order to satisfy the require­
ments of recent legislation. Many institutions require an 
external stimulus to quick start their reaction. Thus one 
of the principal recommendations of the Community 
Language Report (1998) was that the Welsh Language 
Board, government departments, companies and all 
types of institutions should develop the idea of linguistic 
animateurs - individuals charged with the special re­
sponsibility of promoting the use of the language. Three 
other recommendations are worthy of note. The first is 
the rapid extension of a national network of Language 
Enterprise Agencies (Mentrau laith) which would stimu­
late the use of Welsh within particular communities by 
harnessing local needs and initiatives to national strate­
gic resources and expertise, A second recommendation 
was the development of County Resource Centres which 
would provide county-wide services and training to lo­
cal authority personnel and members of the general 
public concerned with developing bilingual service de­
livery in new domains. Above this level there would be 
a National Resource Centre charged with the threefold 
function of language planning, terminological and lan­
guage standardisation and translation requirements to­
gether with the provision of bilingual software, pro for­
mas and administrative guidelines/illustrations to realise 
the bilingual potential of the local state.
The National Assembly
Much will depend on the political lead and wisdom 
shown by the newly formed National Assembly, which 
forms an integral part of the Labour government's consti­
tutional reform for the United Kingdom. Established in 
May 1999 the Assembly is committed to developing a 
substantive bilingual policy within a multicultural con­
text and will likely seek to adopt a stronger multilingual 
line in keeping with the demands of a rapidly evolving 
European Union.
Clearly there is considerable advantage to be gained 
from the recognition of the Assembly as a bilingual body 
from its inception. It is a manifest statement to the world 
that Wales is a bilingual country. While it is fair to as­
sume that the Assembly, in legal and functional terms, 
will seek to operate in a comprehensive bilingual man­
104
ANNALES • Ser. hist, sociol. • 9 • 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
ner, legislation alone cannot create a supportive envi­
ronment within which bilingual working practices will 
flourish. Rather it is the attitude of Assembly Members 
towards Welsh and promoting bilingualism which will 
be the determining factor. Establishing a strong bilingual 
infrastructure for the Assembly is an essential pre­
requisite for being able to operate effectively, but the 
guidelines need to be routinised well in advance of any 
conflict so that disserting members understand the rules 
of engagement. Leaders of the political parties recognise 
this necessity but need to convince some of their 
members that the Assembly could be the biggest boost 
yet to the promotion of Wales as a bilingual society and 
not as a society comprised of a majority and a minority 
linguistic community.
National language policy will focus on three impor­
tant aspects. First language policy in relation to educa­
tion and public administration, equal rights and the so­
cialisation of citizens within civil society. This would 
involve inter alia , issues such as interaction with the 
British state and its unwritten constitution, the European 
Convention on human rights, European Community lan­
guage policies, the development of bilingual education, 
bilingual service provision in local government, health 
and social services. Secondly, economic policies and 
regional development initiatives which seek to stabilise 
predominantly Welsh-speaking communities, to create 
employment, and to promote bilingual working oppor­
tunities. Thirdly, consideration of the interests of Welsh 
language and culture as they are impacted upon by 
town and country/structure planning and improvements 
to the transport system. In addition the pressing housing, 
property control and rural service issues highlighted by 
various bodies including Jigso and Cymdeithas yr laith 
should be addressed directly. A fundamental precept of 
the Assembly's mission should be that its policies on bi­
lingualism should be complemented by the promotion 
of positive attitudes to Welsh culture and heritage.
A prime issue in the normalisation of Welsh will be 
the extent to which it can become a cross-cutting me­
dium of governance and administration and not limited 
to its o\vn Committee for the Welsh language and culture 
i.e. not become commodified and separated out as a 
'problem area'. A second issue is the degree to which 
operating a bilingual Assembly will influence the langua­
ge-choice behaviour of the public. Critics sympathetic to 
the promotion of Welsh have observed that local autho­
rities have invested heavily in statutory language sche­
mes which in reality are of little interest to all but a hand­
ful of Welsh speakers. It would be regrettable if the As­
sembly's commitment were not matched by the public's 
adoption of Welsh as a language of interaction with na­
tional government. In turn the Assembly may use its posi­
tion as an exemplar, a testing ground, an educator and a 
significant actor to influence behaviour in this regard.
A third issue is the supply of specialists to operate
the Assembly. Critics have warned that just as the 
development of a bilingual media (c. 1985-1997) drew 
mainly on the talented professionals of Welsh-medium 
high schools (and to a lesser extent chapels, as hap­
pened in an earlier period), so the fresh opportunities 
afforded by the Assembly and its associated domains wil 
pose a further threat to the staffing levels of the educa­
tion system. This is a major challenge to the university 
sector which as a matter of urgency should provide 
training courses and bilingual specialist diplomas in 
matters related to a range of functions which fall under 
the remit of the Assembly.
The Assembly has the potential to be a major fillip to 
the fortunes of Welsh but it should not be viewed either 
as the 'saviour of the language' or the sole agency for 
language promotion, other voices should be encour­
aged. Currently moves are afoot to establish a new 
Welsh language movement, something akin to the style 
and purpose of Friends of the Earth, which would be a 
national and voluntary network of representatives of 
Welsh-medium agencies and individual members.
The Assembly's recruiting policy and training pro­
gramme could also impact on the public sector, local 
government and especially bilingual educational provi­
sion. Currently there is an acknowledged shortage of 
competent accredited translators, experienced language 
tutors, and skilled bilingual administrators and technical 
specialists. The training infra-structure for a bilingual 
workforce is woefully inadequate. Consequently special 
attention should be paid to how the Government's 
training agencies, such as the TECs, are resourcing or 
failing to resource the required training programmes for 
an increasingly sophisticated, bilingual economy. The 
skills gap in the workplace needs to be addressed ur­
gently if the relationship between the Assembly and the 
rest of the public sector is to operate harmoniously. The 
Local Government Management Board suggest that the 
Welsh Assembly and the TECs should acknowledge this 
skills shortage and establish industry-recognised accred­
ited courses for training tutors. This would necessitate a 
partnership between the further education colleges, the 
universities and the WJEC together with the TECs to de­
velop accredited intense and part-time courses. Go­
vernment agencies should recognise Welsh language 
courses as a necessary skills-based training course for 
those searching for employment. In recognising the ne­
cessity for such skills agencies should pay both course 
and examination fees and lobby to receive government 
assistance for upgrading the skills level of its employ­
ment force as a consequence of the changes set in train 
by the Welsh Language Act, 1993.
The necessity to produce bilingual legislation will 
also have a direct impact on the development of a 
Welsh-language legal community to match that of the 
media community as co-equality of language use be­
comes common in many domains. It is imperative that
105
ANNALES • Ser. hist, sociol. • 9 • 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
the university system and professional training of legal 
specialist take due regard of this trend and attend to the 
very real employment and training needs of the profes­
sion. In addition strong positive messages on the societal 
value of bilingualism will not go unheeded within the 
private sector, even though at present it is not covered 
by the Welsh Language Act. It is likely that the Assembly 
will seek to strengthen the 1993 Act so as to include 
within its remit the private and voluntary sectors.
The social impact of the National Assembly:
Key questions
1) What effect will the arrangements for the bilingual 
servicing of the National Assembly have on the legitimi- 
sation of bilingualism as a societal norm?
2) How will the experiences generated within the 
National Assembly impact on the bilingual character of 
educational and public administrative services, together 
with the local government and legal system?
Two issues of direct importance for education will be 
the wider exposure of all children in Wales to bilingual 
instruction and the introduction of new software and 
terminological data bases which will enable students to 
shift back and forth between information generated in 
English and that generated in Welsh.
A second consideration is the creation of new bilin­
gual opportunities for specialist use of Welsh and Eng­
lish within the democratic institutions and in related 
socio-economic agencies. How will this impact on the 
education system and a whole host of other activities? 
At present, due to the lack of systematic research we 
know virtually nothing as to how aspects of the educa­
tion system will influence these issues. Neither do we 
know very much in detail about the formative influences 
on conceptions of national identity in Wales nor of the 
appropriate use of Welsh or English within selected 
specialist domains. Much of public policy is based on 
very broad assumptions and generalisations about the 
relevance of bilingual education in the service of 
society, but very few research projects have sought to 
document the bilingual educational experience vis a vis 
a monolingual English educational experience.
Two recent exceptions have produced limited infor­
mation relating to the performance of bilingual schools 
in the industrialised south east of Wales. The first is a 
study by Beilin and Higgs and Farrell (1998) which at­
tempted to investigate the language shift process by a 
socio-spatial analysis of census and non-census data wit­
hin selected communities. The study is typical of social 
and educational research in that it is localised, concer­
ned more with a cross-cutting snapshot to reveal a pat­
tern rather than the myriad processes involved and it is 
fairly dependent upon rather blunt aggregate data gat­
hered for purposes other than analysing language shift.
A second, more focused study commissioned by the
Institute of Welsh Affairs was undertaken by Reynolds, 
Beilin and ab leuan (1998). This project sought to assess 
whether or not the Welsh-medium school sector in the 
Rhondda Valleys was characterised by several structural 
advantages which enabled their pupils to out-perform 
pupils of very similar socio-economic backgrounds who 
attended English-medium or Church in Wales schools. 
The results suggested that bilingual children in Welsh 
medium education were distinctive because i): - their bi­
lingualism generates an interaction effect whereby utilis­
ing the two main languages together boosts cognitive 
performance and intellectual confidence: ii) their Welsh 
identity gives them security in a rapidly differentiating 
and increasingly heterogeneous world; and iii) they are 
conscious of their distinctiveness in a way which helps 
to boost social confidence and self-esteem.
As the trend toward sending children from non-Welsh 
speaking homes to Welsh-medium schools is increasing 
there is a real possibility that within the next generation 
bilingual citizens in Wales will be advantaged and 
monolingual citizens will feel at a relative disadvantage. 
In other words, it is the exact opposite of what has hith­
erto been the norm. Because language-related consid­
erations have always been a feature of political mobili­
sation in Wales, it is logical to assume that in future there 
will be increased agitation and mobilization of the needs 
of the monoglot English inhabitants in terms of identity 
formation, educational opportunities and access to cer­
tain types of employment and representation. There is a 
growing tension between those who claim that the 
Welsh language has been rendered politically neutral 
and therefore acceptable, and others who argue that this 
is a smoke screen, despite rapid advances in the status of 
Welsh, for the struggle between Welsh and English re­
mains at a more subtle level and needs investigating.
In more general terms the contemporary relationship 
between education and nationality - or identity-forma- 
tion is still unclear. Many academic studies assert that 
Welsh-medium education predisposes children to de­
velop socio-political attitudes which are favourable to 
nationalism and assume that this transfers into support 
for Plaid Cymru at a political level because of that 
party's championing of Welsh cultural and linguistic 
issues. But in truth such assumptions are not backed up 
by consistent social scientific analysis.
Thus the significance of several recent educational 
research projects in terms of the national picture cannot 
be measured as they are not representative samples. The 
absence of national, time-series educational data of this 
type makes both inferential reasoning and educational 
planning very difficult and dangerous at times.
Post-devolutionary bilingual policy
A pressing major research interest is post-devolution 
comparative work on bilingual policy and language
106
ANNALES • Ser. hist, sociol. • 9 • 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
equality issues within the UK and Ireland. Future policy 
could be directed toward instigating research which 
sought to:
1) contribute both a theoretical and a practical ele­
ment to Language Planning and Language Policy in the 
UK and Ireland.
2) assess the character, quality and success of the 
institutional language policies of the newly-established 
political assemblies in Scotland, Wales and Northern 
Ireland.
3) investigate the complex nature of bilingual edu­
cational and administrative systems in Wales together 
with regional specific systems in Scotland and Northern 
Ireland.
4) assess the role of cross-border arrangements for 
the increased recognition of Irish on the island of Ire­
land, together with Northern Irish, Gaelic and Ulster 
Scots links with Scotland. An obvious, if hitherto un­
known quantity here is the role of the Irish-British 
Council in terms of co-operation, and political bargain­
ing at the UK and European level, together with bilin­
gual education, civil rights and group equality issues in 
Northern Ireland. As a new political forum its role and 
impact should influence the deliberations of European 
Committee of the Regions and several other interna­
tional bodies.
5) investigate to what degree the institutionalisation 
of Celtic languages vis â vs the established dominance 
of English can be a model for the relationship of other 
lesser-used languages world-wide in their relationship 
with English. Potentially this issue is of global signifi­
cance if one can transfer several of the lessons to be 
learned from the survival of the Celtic languages to 
multilingual contexts as varied as contemporary India, 
and much of Sub-Saharan Africa, let alone the evolving 
European political system.
6) gauge the degree to which the information tech­
nology and media opportunities developed in connec­
tion with the National Assembly of Wales and the 
Northern Irish Assembly are capable of sustaining a 
wider range of bilingual practices in public life. It is 
noteworthy that the development of Sianel 4 Cymru, 
and to a lesser extent the Gaelic-medium television 
service, has created a self-confident and pluralist bilin­
gual workforce which sustains a wide range of media 
activities. It is possible that both Assemblies will have a 
similar impact in relation to the information society as it 
relates to matters of public administration, education, 
legal affairs and the voluntary sector.
7) analyse the economic demand for a skilled bilin­
gual workforce in several sectors of the economy; de­
termine to what extent bilingual working practices in 
Wales offer a model for subsequent parallel develop­
ments within a range of multilingual contexts within the 
English regions e.g. either in respect of several European 
languages or selected non-European languages such as
Arabic, Urdu, Hindi or variants of Chinese languages of 
wider communication.
THE WELSH CASE IN EUROPEAN CONTEXT
The technological trends underpinning European in­
tegration suggest four outcomes in relation to the inter­
action of language, ethnic identity, territory and the state 
system. First, mass communication technology has rein­
forced the dominance of English. This has led French, 
German and Spanish, to re-negotiate their positions 
within the educational, legal and commercial domains 
of an enlarged Europe. Fears of Anglicisation ushering in 
greater North American and Japanese influence has 
stimulated the following propositions ; 1) that the EU 
educational system should encourage all students to 
acquire two foreign languages; 2) that a first foreign lan­
guage should be obligatory; 3) that English should be 
taught as a second foreign language, but never as a first;
4) that less information and cultural loss would occur if 
the principle of multilingualism could be instituted in 
most supra-national public and governmental affairs.
In an earlier paper in this journal I discussed the 
implications for autochthonous language groups, such 
as the Basques, Bretons, Irish and Welsh, who might be 
further marginalized in an increasingly complex and 
competitive social order (Williams, 1998. p. 270). I ar­
gued that their only hope lay in establishing regional 
bilingualism as the dominant pattern in education, pub­
lic administration and the law. Ethnic mobilisation is so 
often a surrogate for issues such as political struggle, 
economic deprivation and psychological adjustment 
that the salience of identity is likely to increase as West­
ern Europe avers a more open, pluralistic society. Our 
major challenge is interpreting the disjuncture between 
formal political units and the social behaviour of in­
creasingly autonomous and individualistic citizens.
Logically, if globalization and interdependence can 
enhance the productive capacity of majority, 'nation- 
state' interests, they can also be harnessed to develop 
the interests of lesser-used language groups. The EU has 
harmonised state and community policies so as to 
strengthen its majority language regimes. But the wider 
question of the relative standing of official languages 
makes political representatives wary of further compli­
cating administrative politics by addressing the needs of 
roughly fifty million citizens who have a mother tongue 
which is not the main official language of the state 
which they inhabit. Recent expansion of the EU has in­
creased the difficulties in translating multi-cultural 
communication and guaranteeing access to information 
and hence power for all groups. The real geolinguistic 
challenge is to safeguard the interests of all the non-state 
language groups, especially those most threatened with 
imminent extinction.
Some language groups face a more positive future if
107
ANNALES • Ser. hist, sociol. • 9 • 1999 -1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
they can use this international spur to democratic par­
ticipation in public affairs, by encouraging representa­
tion within decision-making bodies at all levels from the 
local, to the supra-state level. However, reliance on this 
form of supra-structural definition of rights could pro­
vide a false dawn of optimism, if it is not also accom­
panied by a parallel sub-structural reform of many as­
pects of life in multi-cultural societies. It is essential that 
experiments in language planning regimes such as those 
currently undertaken in Euskadi and Wales be fully 
honoured and funded by the local state. Because these 
are 'true Abstand11 minority languages (Trudgill, 1993), 
what happens to the struggling minority in these two 
countries is a predictor of the linguistic fortunes of other 
groups, including those such as Norwegian and Danish 
who are bolstered by their own state, but nevertheless 
face the same external competition from languages of 
wider communication. This is why bodies such as the 
European Parliament, the European Cultural Foundation 
and the Council of Europe are engaged in a fundamental 
political assessment of 'which languages for Europe', 
asking to what extent the international agencies can 
sustain all the competing language interests within a 
representative, but effective, European political system.
CONCLUSION
The search for an appropriate place for both English 
and Welsh in a bilingual Wales will continue to evolve, 
as will the attitudes towards Welsh in general. The inter­
play between the forces promoting the spread of English 
and those perpetuating the resistance of Welsh in the 
past seven centuries captures the power, ambiguity, 
duality and opportunity inherent in any such relation­
ship between neighbours. But the portents are more en­
couraging than ever before that such a search will be 
continued in a non-violent, constitutional mode, em­
phasising the gradualist rather than the radical or reac­
tionary approaches available.
A presistent problem is framing policy on the basis of 
inadequate data and specialist research. Hitherto there 
has been an over-reliance on manipulating the decen­
nial census source which asks four self-report questions 
on Welsh language abilities. Occasionally such data are 
supplemented by very localised language attitudes sur­
veys but the samples are generally too small and poorly 
conceived to be of any analytical or policy-related use. 
The only exceptions are the'Welsh Office, Welsh Social 
Surveys', the Educational Surveys (Welsh Office, 1993) 
and the Welsh Language Board community surveys of 
the use of Welsh (Williams and Evas, 1998).
Despite these comprehensive surveys the overall 
picture is fairly bleak in terms of national data sets. 
There is an acute need for sophisticated, behavioural 
rather than merely attitudinal data-sets on many aspects 
of democracy, devolution and bilingualism. This is an
acute need as so many new policies will be enacted in 
the future on the basis of very little research and almost 
no tradition of time-series analysis of issues generated 
by seeking to operate a bilingual education system 
within a plural society.
Nevertheless, the public place of Welsh is beginning 
to be institutionalised, even if this necessitates greater 
dependence upon the local state as the source of both 
its legitimacy and finance. Having fought to obtain a 
semblance of equality for both languages in the post-war 
period, it is likely that the next generation's concerns 
will have far more to do with questions of economic 
development, environmental protection and conserva­
tion and holistic planning. This is particularly true for 
the Heartland region, for despite its relative decline as 
the source for initiative and social change, it is still the 
base and bedrock of a relatively autonomous Welsh 
culture. We may be attracted by the rapid development 
of bilingual networks in the burgeoning urban districts of 
the anglicised lowlands, but we neglect the northern 
and western base of Welsh culture at our peril, for its 
experience has taught that only when the language is a 
fully functional element in an integrated society will it 
prosper and please. No amount of formal language 
planning, however desperately needed, can substitute 
for the popular will to reproduce a language and cul­
ture. In that respect bilingual education in Wales faces a 
more optimistic future as the first century of the new 
millennium dawns. Holistic interpretations of the role of 
bilingualism within a Welsh and European context will 
inform the decisions of the National Assembly who will 
be able to advance, colour and scrutinise the direction 
of educational and social policy.
The reconstituted Welsh Language Board looks set to 
become the most critical government agency yet in the 
social history of Welsh. Such status language planning 
allows for a measure of purposive rather than reactive 
thought and policy formulation. It also presages a new 
era of fresh initiatives and holistic interpretation of lan­
guage in society which, though small-scale, represent 
grounded local involvement in socio-linguistic issues. 
The emphasis on planning requires integrated action by 
a number of agencies and represents a more holistic ap­
proach than hitherto has been the norm. Additionally, 
rather than be exclusively concerned with the needs and 
interests of Welsh speakers, we are beginning to address 
the concerns and involvement of second-language 
learners and non-Welsh speakers in a more systematic 
fashion, thereby extending the bicultural nature of soci­
ety from both ends of the spectrum.
In consequence, the Welsh language has entered a 
new phase of legitimacy, best evidenced by the working 
out of the Welsh Language Act (1993). Doubtless there 
will be calls for a revised Act to cover the private and 
voluntary sector, yet support for Welsh can no longer be 
interpreted as essentially a symbol of resistance to An-
108
ANNALES • Ser. hist, sociol. • 9 • 1999 -1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
glicisation, for it is itself deeply imbued in the process of 
state socialisation. The language has become a con­
tested instrument both of reform and of governance, of 
opposition and of authority. Welsh is increasingly incor­
porated into the machinery of government, of justice, of 
public administration and of civic control. In that one 
respect the promotion of bilingualism in Wales repre­
sents one small step for the realisation of popular de­
mocracy in Europe. My fervent prayer is that many other 
minority communities throughout this troubled conti­
nent may take heart from this example and seek to take 
hold of their own destiny by realising their dreams in 
and through a language struggle.
ACKNOWLEDGMENTS
This paper was given as the plenary address to the 
conference on "Language Policy and Bilingual Educa­
tion for Minorities in Europe", hosted by the Faculty of 
Education, University of Ljubljana and the ZRS, Koper, 
15th of May 1998. I am grateful to Dr. Lucija Cok for her 
gracious invitation and kind hospitality.
Tables 6-10 are reproduced with permission from 
'Statistics of Education and Training in Wales: Schools 
1998', Government Statistical Section, The Welsh Of­
fice, Cardiff.
DVOJEZIČN O ŠO LSTVO  V KORIST EVROPSKI DEMOKRACIJI
Colin H. WILLIAMS 
Univerza v Cardiffu, O ddelek za velški jezik, UK- Cardiff CF  1 3XW , P .O .Box 910
POVZETEK
Avtor raziskuje razvoj dvojezičnega Šolstva v Walesu in predstavi številne izzive ter priložnosti, ki so vplivali na 
prizadevanja Valižanov, da bi reproducirali svoj jezik  in kulturo. Tako šolstvo kot družina in skupnost so 
obravnavani kot glavna področja, kjer se ohranja jezikovna reprodukcija, vendar za razliko od drugih dveh stebrov 
kulturnega boja so bili parametri vzgoje v Walesu vedno bolj zaznamovani z  oddaljevanjem od britanskih norm. 
Razvoj nacionalnega učnega načrta za Wales, ki je  sledil Zakonu o šolski reformi leta 1988, je  skupaj z  drugimi 
ključnimi zakonskimi spremembami, kot sta Zakon o valižanskem jeziku leta 1993 in Zakon o vladi v Walesu leta 
1998, spremenil politično-zakonsko pojmovanje dvojezičnosti in institucionaliziral dvojezično šolstvo kot po­
memben element valižanskega nacionalnega življenja.
Avtor posreduje predloge o šolski politiki in razvojni politiki skupnosti glede na delovanje Državnega zbora za 
Wales, ki je  bil ustanovljen maja leta 1999. Članek se osredotoči na oceno vpliva, ki ga ima institucionalizacija dvo­
jezičnega šolstva na socialne spremembe in enakost skupin v Walesu. Posebna pozornost se posveča politiki 
primerjalnosti jezikov in podani so predlogi za britansko in irsko ustavno partnerstvo po pooblastitvi. Ta nastajajoča 
in potencialno radikalna zveza se smatra kot pomembna preizkušnja za jezikovne pravice in demokratične spre­
membe v okviru preoblikovanega evropskega političnega sistema.
Ključne besede: izobraževanje, dvojezičnost, jezikovne manjšine, Valižani
REFERENCES
Aitchisor, J., Carter, H. (1997): The Welsh language 
and Hausehold Structure, Planet, 362.
Beilin, W., Higgs, G., Farrell, S. (1997): Halting or Re­
versing Language Shift: A Social and Spatial Analysis of 
South East Wales, Final Report for ESRC grant no. 
R000236330.
Higgs, G., Beilin, W., Farrell, S., White, S. (1997):
Educational Attainment and Social Disadvantage. Con- 
textualising School League Tables. Regional Studies, 31 
(8), 779-793.
Jones, J. R. (1966): Prydeindod. Llyfrau'r Dryw. Llandy- 
bi'e.
109
ANNALES • Ser. hist, sociol. • 9 • 1999 • 1 (16)
Colin H. WILLIAMS: BILINGUAL EDUCATION IN THE SERVICE OF EUROPEAN DEMOCRACY, 89-110
Mackey, W. (1988): An introduction to geolinguistics. 
In: Williams, C. H. (ed.): Language in Geographic Con­
text. Clevedon, Avon, Multilingual Matters, 20-46. 
Morris Jones, R., Ghuman Singh, P. A. (eds) (1995): 
Bilingualism, Education and Identity. Cardiff, University 
of Wales Press.
PDAG (1988): Interim Report of the Secondary Working 
Committee, PDAG. Cardiff, unpublished (mimeo).
PDAG (1992): Response to the White Paper on Educa­
tion, Diversity and Choice: A New Framework for 
Schools, PDAG. Cardiff, unpublished (mimeo).
Reynolds, D., Beilin, W ., ab leuan, R. (1998): A Com­
petitive Edge. Why Welsh Medium Schools Perform 
Better. Cardiff, The Institute of Welsh Affairs.
Trudgill, P. (1993): The Ausbau Sociolinguistics of Mi­
nority Languages in Greece. Mimeo, Lucerne University.
Welsh Office (1993) - Welsh Social Survey. Cardiff, The 
Statistical Section, The Welsh Office.
Welsh Office (1998) - Statistics of Education and Train­
ing in Wales: Schools, 1998. Cardiff, The Statistical Sec­
tion, The Welsh Office.
Williams, C. H. (1994): Called Unto Liberty. Clevedon, 
Avon, Multilingual Matters.
Williams, C. H. (1998): Operating through two lan­
guages. In: Osmond, J. (ed.): The National Assembly 
Agenda. Cardiff, The Institute of Welsh Affairs, 101-116. 
Williams, C. H. (1999): A language policy for the Na­
tional Assembly of Wales. In: ECTARC (ed.): Culture, 
Nation and Region in Europe. Llangollen, ECTARC. 
Williams, C  H., Evas, J. (1998): The Community Lan­
guage Project. Cardiff, The Welsh Language Board/ 
Cardiff University. 220 p.
110