Chapter 3
Conceptualising equal opportunities
A cartoon shows a man behind a desk in a jungle clearing addressing a line-up of various creatures cat. monkey, elephant, sea-lion, snake, frog, bird and a goldfish in a bowl on a pedestal. 'To ensure a fair selection you all get the same test' he says 'You must all climb that tree."
(DGXXII (1995) Alpha Toolkit: Make (hand's for today which will he .solutionsfor tomorrow, Brussels: IXiXXII European Commission p. 25)
The concept of equal opportunities is riddled with problematic constructs. Policies aimed at achieving EO are. by the same token, deep!) fraught and tend to be limited in their effectiveness Rooted in ideologies about the nature of a "just society", competing and at times contradictory underlying assumptions are rarely thoroughly questioned or examined by organisations committed to being 'equal opportunities employers*. Added to this complexity is the tension in the relationship between HO principles and the workings of a laissez lain- labour market. Some commentators argue that policies designed to achieve EO are an expensive luxury and at odds with the operation of an efficient economy. Others maintain that to have the distribution of positions within hierarchies (and indeed the structuring of organisations themselves) influenced b\ ascriptive (inherited) characteristics such as gender and race rather than by merit is economically inefficient and supremelv wasteful of human resources. Moreover, it is argued, in the context of the global economy, it is precisely, the efficient use of human resources which provides a competitive edge.
I his chapter scrutinises some of these issues and outlines a range of approaches towards I () Beginning with liberal philosophical thinking about equal treatment and equality of access, it moves on to examine concerns about equality of outcome, and positive action and positive discrimination. linaliv it looks at more recent ideas
Conceptualising equal opportunities 27
about deconstructing the underlying orientations of organisations, such as their androcentricity (or 'male-centredness*) and Eurocentrism. and the way they operate to the advantage of stakeholders, reproducing elites who share common characteristics in dimensions of gender, class and race. This leads to a consideration of those policies which seek to "mainstream" EO which are based on the 'politics of difference*. They seek to transform organisations and create a culture of diversity in which people of a much broader range of characteristics and backgrounds may contribute and llourish. This potentially has the advantage not simply of widening the gender base of organisations but also of tackling other forms of disadvantage and discrimination, on the grounds of race, ethnic origin, nationality, disability, sexual orientation and regional, linguistic and cultural difference. There are. however, competing understandings of the mainstreaniing approach, some of which would almost certainly have deleterious effects for EO.
Much has been written on how language affects the social construction of male and female subjectivity (Violi 1992; Weedon 1987) and how co-existing discourses have different statuses in relation to each other (Foucault 1980). In discussions of EO, a set of key concepts recurs, most of which have a variety of meanings and usages. This is at the root of some of the difficulties in devising effective policies in education, training and employment. These terms include equality, equity, fairness, sameness, difference, just, justice, disadvantage, and discrimination. There are many helpful, detailed discussions, both of meanings of these terms and of their implications for effective delivery of EO policies (see for example Brah 1991; Brine 1995a; Cockburn 1989, 1991; Jewson and Mason 1986; Maynard 1994; Phillips 1987; Webb and Liff 1988; Young 1990). The purpose of this chapter is to draw attention to the main conceptual approaches and to offer working definitions for interpretations of accounts which follow.
The discourse of EO gives rise to many interpretations, not simply by different actors sharing the same language but in transnational debates. Such debates at the European level illustrate the variation in the extent to which the languages of the Member Stales share (or rather, fail to share) a common set of concepts and the words to describe them. In some languages, for example Flemish and Swedish, there is no satisfactory translation for the word 'gender". This can give rise to confusion in transnational discussions where a distinction needs to be made between biological (sex) and
28  Conceptualising equal opportunities
Conceptualising equal opportunities 29
socially constructed (gender) differences between men and women. In Spanish, 'positive action3 is translated as 'positive discrimination', despite the distinct conceptual difference in meaning in English and other languages (see below). Employers seeking to keep within the law on equal treatment when drafting job advertisements face particular challenges if the language they use is a gendered one. Indeed, the Equal Opportunities Commission (EOC) has found it necessary to publish guidelines for employers advertising in Welsh, given that the language has no female versions of some job titles (Awbery 1997). The gendering of concepts more generally creates difficulties for the analysis of EO in education, training and the labour market. This issue is taken up in Chapter 5 with regard to the English language, where concepts such as skill and unemployment are shown to be gendered in their social construction. This has a series of implications for the analysis, understanding and development of policy. Mapping terms and concepts concerned with EO at the European level is clearly a vital task in the mainstreaming agenda.1
After discussion of three broad approaches to EO (equal treatment, positive action and discrimination, and politics of difference), the chapter offers corresponding models of EO approaches in training policy, characterised as •tinkering', 'tailoring' and 'transforming*. Tinkering can be roughly equated to liberal approaches to equal treatment (resting on the notion of 'sameness'), while tailoring seeks to integrate women into organisations and cultures structured around the needs of men. by making special provision for the ways in which they are 'different' from men. However, transforming acknowledges the differences among men as well as those among women, and. by recasting mainstream provision, seeks to accommodate both gender and other dimensions of discrimination and disadvantage, for example those based on ethnicity and disability.
This conceptualisation of models of EO forms the theoretical framework of critiques of EO policies in relation to training strategies at the European level for the rest of the book. The empirical chapters which follow seek to illustrate the limitation of equal treatment and positive action approaches in training policy and argue the need for a transformative approach.
EQUAL TREATMENT
The concept of equality of opportunity constructed as equal treatment or equal access for women arguably traces its roots to Mary Wollstonecraft (1967 [1792]) and the liberal feminist tradition. She extended the notion of citizens' rights, debated by philosophers such as Jean Jacques Rousseau at the time of the French Revolution, to include those of women (although she focused exclusively on middle class women: like many Enlightenment writers of the time, her notion of equal rights did not extend across class boundaries). Fundamentally the concept of equality implies that no individual should have fewer human rights or opportunities than any other. It is a concept that few but the most extreme now have difficulty in supporting, at least in principle.
However, if the intention is to challenge the significance of certain ascribed characteristics in determining the allocation of positions, then the main problem with the equal treatment approach to EO is simply that it is ineffective. This is unsurprising. It will inevitably reproduce inequalities that exist in the broader context. The systems and structures which perpetuate unequal power relations between men and women (and between people from different classes or races) as groups in wider society inevitably impact upon organisations seeking to offer 'equal treatment' to men and women.
The weakness of the equal treatment approach lies in a lack of analysis of the relationship between public and private spheres. In effect, public life is ring-fenced as if it were unaffected by inequalities in the arena of private life. As Humm (1989: 63) has argued:
even with the granting of equality to women in public life, women's domestic labour will always be unequal to that of men. More subtly, the hidden patriarchal agendas of public institutions can subvert the apparent equality of legal rights.
There are clear connections articulated here between patterns of inequality in the private sphere and those which then result in the public sphere. Granting equal access to men and women will only benefit certain women: those whose cultural capital, experiences, family circumstances and share of domestic responsibilities are similar to those of men as a group.
The equal treatment model is rooted, then, in a narrow distributive conception of justice, and focuses the debate upon the allocation of positions within a hierarchy which is given. It ignores
30  Conceptualising equal opportunities
Conceptualising equal opportunities 31
the impact of patriarchy in the home and its interaction with capital to produce gendered organisations which systematically disadvantage women. It discounts the impact of other forms of unequal power relations, for example those which accrue as a result of class or racial oppression and discrimination. Systematic domination of some groups by others is reflected in organisational institutions, systems and structures. The equal treatment approach simply removes the more obvious structural barriers to individuals' access to systems which themselves are shaped by those patterns of domination and oppression.
The ineffectiveness of equal treatment as a model of EO policy is illustrated by the highly gendered patterns of segregation operating within the professions in the UK to which women have technically been granted 'equal access'. In the nineteenth and early twentieth centuries, women were legally prohibited from becoming qualified in the major professions such as law and accountancy. Legislation banning the exclusion of women from the professions was finally passed in the 1919 Sex Disqualification (Removal) Act. While women began to enter these professions, it remained legal to pay them different rates and to discriminate against them on the grounds of gender. Women civil servants were obliged to resign on marriage in Northern Ireland up until 1972. The Sex Discrimination Act and Equal Pay Act passed in the 1970s and brought into effect in 1975 made it illegal to discriminate on the grounds of sex in education, training, employment and pay. The EOC was set up at that time to monitor the law and suggest amendments, to foster the development of EO for men and women, and to back legal cases of discrimination brought by individuals and groups.2
In the 1980s and 1990s, women have been entering training for the professions in increasingly substantial numbers, aided by this legislation and by the transparency and qualifications-based specificity of entry requirements. As Crompton (1987) argues, it is more difficult to discriminate directly against the entry of members of a specific group into training for professions which operate a qualifications lever for entry. By contrast, it is noticeable that women have not made the same inroads into senior management positions in the UK, particularly in the private sector (Hansard Society Commission on Women at the Top 1990; McRae 1996) where routes of entry are not as transparent and are less linked to the acquisition of specific qualifications, and where networks are far more significant in the allocation of posts (Coe 1992).
Women now constitute about half the undergraduates on courses for the professions in the UK. However, gaining equal access to courses has not resulted in the disappearance of gender as an organising principle in the allocation of positions within professions (see Atkinson and Delamont 1990 for a general account of female marginality in the learned occupations; see also Allen (1988) on medicine; Fogarty el al. (1981) on the civil service, architecture and industry; Spencer and Podmore (1987) on law, medicine, health service administration and engineering; and West and Lyon (1995) on academia). Women are clustered in certain fields within the professions. In medicine, for example, they are found in general practice and community medicine, while in law, they are likely to focus on family and divorce law or work in the public rather than the private sector which tends to offer more flexible work conditions and return to work arrangements, if not better salaries. In academic life, women are concentrated in contract research posts. Even the 3.5 per cent (1989/90) of professors in the UK who are women earn considerably less than their male counterparts after holding constant age and years of service (Association of University Teachers 1992; West and Lyon 1995).
This shift from exclusion from certain spheres of public life to segregation within it (Walby 1990) represents the fall-out of the interaction between patriarchy and capitalism. The interests of both systems co-determine certain prescribed and limited terms on which women may enter and progress within the professions. Few are found in senior positions in the professional associations which shape the agenda and determine the nature of the hurdles for admission and progression. Overwhelming evidence suggests that there have not been major shifts in the domestic division of labour despite the increase in female economic activity rates (Brannen and Moss 1991; Morris 1990; Pahl 1984; Warde and Hetherington 1993; Witherspoon 1988). The group which is particularly heavily burdened comprises those women with part-time jobs, who tend to have young children and are either single parents (Dex 1988) or receive very little assistance from their partners (Gershuny et al. 1986, 1994). Hence only some women in some circumstances can compete on equal terms with men. Their ability to do so is cited as evidence that equal access exists for those who wish to take advantage of it. In reality, such equality of access is an illusion. In addition to unequal demands made on the time of men and women for domestic responsibilities, a variety of exclusionary mechanisms.
32  Conceptualising equal opportunities
Conceptualising equal opportunities 33
experienced by women as "chill factors1, still operate widely wilhin the professions inhibiting women from reaching the top jobs (see. for example, Rees et al [in press] on women and principal posts in Northern Ireland).
Walby (1990) and Duncan (1996) draw attention to the significance of societal 'gender contracts' between men and women which are underpinned by taxation, welfare, childcare and other institutional arrangements. Gender contracts are defined by Duncan as 'a rough social consensus on what women and men do, think and are' (1996: 415). They inform expectations about the domestic division of labour and power relations which, in turn, shape systems and structures that reinforce those expectations- These combine to exercise a strong influence on economic activity rates and patterns of gender segregation. While Duncan argues that the nature of the gender con trad for any society can change, such patterns tend to be relatively long lasting. Equality policies, such as the provision of childcare, that do not address underlying gender contracts will inevitably be limited in their effects.
This is demonstrated in the review of EO measures which follows, A legal framework providing for equal treatment for men and women is clearly essential; however, this approach focuses on the establishment of rights and procedures rather than on outcomes. One of the main criticisms of the shortcomings of European law on EO is that it focuses on issues of equal treatment of men and women as workers, hence it leaves the inequalities rooted in the gender contract largely untouched. In other words, it addresses the symptoms rather than the causes of inequality. As a consequence, it is clearly limited in its potential effectiveness.
In the UK, equal treatment legislation has clearly had some effect on the position of women in education, training and employment. However, it was the 1944 Education Act, principally designed to provide free and compulsory education for all children aged between five and fourteen as a device to open up higher education opportunities to children from the working class, which enabled girls to have equal access to education for the first time. However, this equal access to education nevertheless still had the effect of reinforcing the gender contract, as the curriculum was designed to gear children to their future roles in the sexual division of labour: girls as homemakers and boys as breadwinners (HOC 1996a).3
EC Directives led to the introduction of EO legislation to make sex discrimination in employment and pay unlawful all in the
Member States in the 1970s; similar legislation had been introduced in the UK by the Labour Government in 1970. However, despite the one-off reduction in the pay differential between men and women following the Equal Pay AcU the impact uf legislation has been highly circumscribed (Gregory 1987, 1992; Kahnand Meehan 1992; Leonard 1987a, 1987b; Rubcry et al. 1994). Patterns of vertical and horizontal segregation remain cemented both within education and training systems and in the labour force. The impact of the original Equal Pay Act (not brought into effect until 1975 to allow employers time to get their house in order) was undermined by patterns of gender segregation which meant women had no male 'comparator' with whom to claim equal pay. The Equal Pay for Work of Equal Value (1984) amendment, introduced to address this issue, has had limited effect because of the cultural barriers attached to valuing work largely done by women in the same terms as that done by men (Kahn and Meehan 1992). Von Proudzynski (1986) points to the way in which 'feminine skills' (such as manual dexterity and speed) are valued less than alleged 'masculine' qualities (such as physical strength) in collective agreements and judicial decisions. The personal and financial cost of bringing action
(inhibits the widespread use of the law to further EO (Leonard 1987b} and the limited resources available to the EOC means that support has to be restricted to cases deemed winnable which further ■f clarify or extend the law.
Equality of access (egalite des chances) clearly does not lead in : practice to equality of outcome (ega/ite des fails). The concept of
EO as a necessary and sufficient measure to eliminate the impact of gender on educational and employment lifechances is therefore found wanting. Equal treatment legislation has not significantly affected either the rigidities of segregation within occupations and professions, nor has it eliminated (although it has clearly diminished) gendered differentials in pay. Indeed, as we shall see in Chapter 7 on the impact of EC action programmes on training for women, equal treatment policies can in some circumstances reinforce the status quo, or, indeed, even increase skill gaps between men and women.
A further weakness in the equal treatment approach lies in the passive approach to the application of the principle. Equal treatment policies are not always reinforced by any mechanisms to ensure their application or to monitor performance. A statement to the effect that an organisation has an EO policy is clearly not
34   Conceptualising equal opportunities
sufficient to ensure its application. Pro-active procedures, such as gender monitoring, awareness-raising and training at the very least, are necessary to make equal treatment effective within its own limited terms. The work of Cockburn (1991) and Collinson et al (1990) has been extremely important in using qualitative field work to dissect the processes by which organisations reproduce their cultures (including the characteristics of individuals in positions within hierarchies), even when equal treatment policies of varying strengths are in place.
The failure of equal treatment to produce equal outcome lies, then, in the lack of analysis of the causes of segregation, be it along gender, race or any other dimension. The failure to acknowledge the impact of hierarchies such as patriarchal ordering through the gender contract, and racial oppression, limits the scope of equal treatment actions. The equal treatment approach suggests that people should be treated simply as individuals without recognising the impact of group membership in the allocation of positions and the implications of this for cultural reproduction. It is limited, too, by weak application of the principle. Positive action and positive discrimination measures have been introduced in many settings to seek to address this weakness.
POSITIVE ACTION AND POSITIVE DISCRIMINATION Positive action
The concept of positive action (PA) recognises that equal treatment can reinforce existing inequalities. PA measures seek (in the hackneyed phrases) 'to create a level playing field' or Ho untie the hand behind women's backs'. The emphasis shifts from equality of access to creating conditions more likely to result in equality of outcome by equalising starting positions. The notion of merit determining outcome is still in place and the terms by which merit is defined remain unquestioned. The contribution of PA is to assist women to compete more effectively within the existing construction of merit.
PA rests on the notion that membership of groups makes a difference to outcome. PA measures have been based on attempts to identify barriers to women's participation in education, training and employment. They seek to provide mechanisms which will facilitate women's entry on an equal footing by compensating for unequal starting positions. They are predicated on addressing the ways in
Conceptualising equal opportunities 35
which women are different* from men and thereby disadvantaged'. This approach begins, then, with a recognition of the significance of group membership determined by ascriptive characteristics in the allocation of positions. It acknowledges the dominance and oppression of some groups by others. Like equal treatment, however, it is principally concerned with the distribution of positions within hierarchies rather than with challenging the structural status quo which reinforces systems of oppression in those hierarchies.
Women-only management training courses provide an example of PA training for women which in some cases is designed to assist them to behave more like a man or to survive in a male culture. Curricula headings include "developing a killer instinct', 'coping with the office romeo'. 'making yourself heard in meetings'. Such training, while much appreciated by course participants, and deemed effective by them in assisting promotion prospects, seeks to adapt women to a particular model of masculinity dominant in management hierarchies (Pollert and Rees 1992: Tanton 1994). There can be casualties from this approach. Marshall (1995) for example documents the case of women senior managers who drop out not because of the work but because of their discomfiture within a male dominated culture.
One of the best known PA measures in employment is the UK's Opportunity 2000. launched by Business in the Community and backed by the (then) Prime Minister, the Rt Hon Mr John Major. Signatories to Opportunity 2000 are. for the most part, a group of large employers who are publicly committed to improving inter alia the proportion of women in middle management and top jobs through various forms of PA. Many of these employers conducted an analysis of the gender make-up of their workforce, introduced actions such as HO awareness raising, appointed an EO officer, and set goals and targets for proportions of women in certain grades by the year 2000. Progress is being monitored, but annual reports of Business in the Community suggest, unsurprisingly, that change is slow. The recession is identified as causing difficulties in sustaining momentum (Business in the Community 1992. 1993; Hammond 1992).
However, more deep-seated, structural and attitudinal factors may also be at work. Hammond and Holten (1991) suggest that, to be effective. PA measures need commitment from the top. widespread 'ownership* of the policy (whereby all employees and especiallv line manauers are committed to it and share responsibilitv
36  Conceptualising equal opportunities
for it), links with business plans, and adequate resources. This package of support measures is rarely systematically applied. Voluntary PA measures do not on the whole deliver the cultural and organisational change required to combat gender segregation, A number of qualitative studies have documented all too graphically the mechanisms used by line managers to subvert such policies and the way in which male backlash can erupt in organisations seeking to develop PA (Cockburn 1989, 1WU
In training. PA measures which seek to address gender inequalities aim to establish the extent to which gender is an issue, and then To adapt provision to take on board women's 'special position'. This may arise because of their career breaks, their domestic commitments, or because they are under-represented in certain subjects and at some levels. Hence PA in training includes measures to accommodate women's domestic responsibilities, to check for gender biases in course materials, and to provide special supplementary training (for example for women wanting to enter construction industry trades) to make up the 'deficit* of their knowledge compared with that of men before they begin a mainstream training course. Women-only training by women tutors, especially for women returning to work or entering training in male-dominated fields, is also categorised as a PA training measure. Such courses characteristically include assertiveness training for women (see Willis and Daisley 1990). These measures are discussed in more detail later in this chapter.
Awareness raising is an important aspect of PA measures. The Government- and EOC-backed UK Fail Play initiative, which was started as a grass roots movement in Wales as Chwame Te$ (Weish for 'fair play4), comprises regionally-based consortia of organisations which seek to raise awareness of the barriers to women's participation in work both qualitatively and quantitatively (DfEE/EGC 1997), The initiative has already spread lo Belgium and the Netherlands, and there are plans to set up Fair Play in Italy and a number of other Member States. Much of the emphasis in Fair Play is on raising awareness of the need to develop women's skills, to address the issue of childcare needs, and to recognise and remove barriers to women's recruitment and progression in the workplace.
In Belgium, the Netherlands and Italy, some PA programmes are backed by the force ot law, and financial incentives accrue to employers providing such programmes from the Stale. In Italy, France and Belgium, there are collective agreements between
Conceptualising equal opportunities 37
employers' and trade unions^ organisations on women's training. In parts of the US, PA is enforced by contract compliance. There is very little PA training in the UK., where such actions are voluntary, and, indeed, of all EU Member States, the UK is the only one which still places restrictions on certain categories of women taking up training opportunities for the unemployed because of their ineligibility for unemployment benefit in their own right.4 While it is possible to apply for exemption under the Sex Discrimination Act to provide single-sex training, this facility is under-used as a mechanism to train women in areas where they are under-represented.
Positive discrimination
It can be argued that equal treatment is in itself in effect discriminatory, given the different sets of cultural capital and normative expectations of men and women and the history of discrimination [ and disadvantage. PA can be seen as providing window dressing; it
I facilitates some women in some areas, in particular well qualified,
I middle class women seeking to enter the professions, without
j affecting the statm quo for the vast majority. It does not challenge
\ the culture and structure of organisations which give rise to
i inequalities in the first place. Different treatment according to
different circumstances is needed, the argument runs, to ensure not simply equality of opportunity but equality of outcome The policy implication of this analysis is positive discrimination (PD).
PD seeks to bring about changes to the status quo through mechanisms designed to increase the participation of the under-represented group: it is in effect the application of 'unequal treatment'. In the US, affirmative action measures allow for positive discrimination in some areas of employment in favour of women and members of ethnic minorities. Quotas also operate in Sweden. This approach is seen as a temporary measure to even up the balance of characteristics, such as gender and race, of people in sectors or at levels in the hierarchy which are particularly skewed. PD in employment is illegal in the UK.
PD thus goes furlher than PA. It acknowledges thai discrimination exists and needs to be addressed: it seeks to redress uneven balances. Discrimination can be active or passive. Failure to know about the relationship between characteristics in the distribution of resources or positions is interpreled as passive discrimination. Failure to undertake gender (or ethnic) monitoring has been identified therefore as a
38  Conceptualising equal opportunities
Conceptualising equal opportunities 39
form of passive discrimination. Organisations that are contractually bound to show that they are offering equal treatment to groups m their provision need to monitor their performance of services to be aware oť the outcomes of their actions.
This was shown graphically in an evaluation of the extent to I which the Training and Enterprise Councils in England and Wales, charged in their contracts with providing equaJ treatment for ethnic minorities, were found guilty of passive discrimination by not having instigated ethnic monitoring (Boddy 1995). Boddy argues that key elements in combating discrimination, once monitoring has illustrated the pattern of outputs, arc a diagnosis of causes, and evaluation and feedback linked into subsequent action. This rarely happens, TECs have also been the subject of a formal investigation by the EOC which revealed many had inadequate mechanisms for gender monitoring (EOC 1993).
The key question is, what should EO policies set out to achieve? Should ihey move beyond equal treatment to counterbalance the effects of past discrimination? In 1996. the British Labour Party decided that, in order to increase the low and apparently declining number of women in the House of Commons, it was necessary to produce all-women short lists for nominated candidates in certain seats where new candidates were being selected, with the agreement of the local party. This produced both considerable backlash and j some vehement support. Comments from those opposed to this I form of PD couched their objections in terms of the desirability to have 'equal treatment' of candidates and not to 'discriminate' on the grounds of gender. Such arguments were highly revealing in showing the models of equality within which the protagonists were operating. The exercise provoked considerable hostility and its legality was successfully contested.
A problematic issue in PD is that of the gendering of value. It is difficult to engender parity of esteem for activities and occupations that are valued differently because of their gendered associations. Campaigns for equal pay tor work of equal value have argued that, while women's jobs may be different from those of men, nevertheless they can be equally demanding, and have an equivalence in level of skill required (Kahn 1992), This moves the debate on from the concept of sameness to recognising and establishing equivalence, and, by implication, the need to seek parity of esteem and reward for different, gendered activities (Kahn and Median 1992). Job evaluation schemes provide an underused opportunity to decon-
struct the skill levels of various occupations irrespective of the usual gender of the occupants,
A further dimension of equality is the concept of indirect discrimination, the most obvious example of which was the situation in the UK where part-time workers were not entitled to employment benefits (such as employment protection, redundancy entitlements and maternity rights) on the same basis as full-time workers. As the vast majority of part-time workers are women, this was eventually recognised as indirect discrimination and, despite considerable vociferous objections from the business world, was eventually addressed through equal treatment legislation enforced on the UK by a European Directive.
Policies towards EO. then, can be permissive, they can include PA measures based on an analysis of the 'barriers1 facing women, or they can be discriminatory - that is, they can positively favour a less represented group in order to achieve a better balance. However, these approaches to EO, which seek to ensure equality of access and which address disadvantages in certain groups, can only go so far in bringing about equality of outcome. Such 'procedural' models (West and Lyon 1995) focus on the elimination of discriminatory practices so that 'merit' will apparently determine who rises to the top. But as Young (1990) has argued, such systems leave unquestioned an hierarchical division of labour and a supposed allocation of people to top places on the grounds of merit. Young asserts that the notion of merit as measured by the applications of impartial measures of technical competence is a myth. Moreover, affirmative (positive) action challenges principles of liberal equality and represents the application of the distributive paradigm of justice;
It defines racial and gender justice in terms of the distribution of privileged positions among groups, and fails to bring into question issues of institutional organization and decision making power.
(Young 1990: 193)
This brings us to what West and Lyon (LQlJ5) describe as the structural model which takes issue with organisational arrangements that are regarded as constructing barriers and impediments to the benefit of one group more than another. The procedural model which underpins positive action focuses on the elimination of informal discriminatory practices so that imbalances in opportunities are redressed and 'merit1 will then determine who rises to the
40   Conceptualising equal opportunities
Conceptualising equal opportunities 41
top. Positive discrimination seeks to 'correct' the distribution of positions by a system of quotas to create a Lfairer balance7. However, the more radical structural model is concerned with effecting outcomes by changing cultures, and thereby systems and structures, rather than perfecting or skewing the rules of the contest. It has the potential to address the gender contract underpinning systems and structures. The next section examines the structural model which I have called mainstreaming.
MAINSTREA MING EQUALITY
The model of EO which underlies mainstreaming policies is based upon the notion of the politics of difference. While the significance of the concept of difference between groups rather than sameness among individuals is now widely accepted, its implications for policies seeking to ensure EO are less well understood. Squires (1994: 7) shows how this is due to a confusion between the importance of difference and domination:
we need to ask of equal opportunities whether it is difference or disadvantage which is our primary concern, for there is a difference between distinguishing people according to group attributes, and explicitly restricting the actions and opportunities of group members. Yet simple difference all too often does translate into perceived inferiority,... Of central importance .., is the attempt to overcome disadvantage without denying difference-How should this be achieved? The politics of difference perspective recognises the androcentricity of organisations and seeks to change it, thus facilitating women's full participation on equal terms. It is a longer-term strategy towards EO than either equal treatment. PA or PD, and recognises, and indeed celebrates, diversity. Mainstreaming policies are those which respect and respond to differences, rather than seeking to assist women to fit into male institutions and cultures by becoming 'more like men' (Cockburn 1991).
Many feminist writers have moved away from notions of equality and equal treatment to focus increasingly on The androcentricity of institutions and the superstructure (Smith 1987). The task is to challenge and change the male-streamism of organisations, illuminating how employers are not gender-neutral but. for example, privilege criteria for promotion (such as longevity of service and unbroken careers) which automatically and systematically favour
one gender rather than the other. From this perspective, the transformation of institutions becomes the agenda, rather than the continuing attempt to improve women's access to and performance within organisations and their hierarchies as they are.
An example of the internationalisation of the male norm in thinking which is reflected rn discourse and policy comes from the European Commission (EC) and its documentation on employment and employment policy. The term used in the EC to describe
I patterns of work for the most part undertaken by women (such as
part-time and temporary employment) is 'atypical work1 (see Chapter 4; McGiveney 1994; Meulders et ai 1994), The term itself indicates the idea that there is a male norm of working behaviour, which is adopted as a universal, from which many women deviate and against which they are measured. An agenda to transform would seek to normalise the plethora of working arrangements followed by many women and some men, rather than considering
-s them in the context of deviation from a particular pattern followed
by most men.
It is women's own experiences of unease or discomfort when seeking to fit in with the male norm which have prompted this \ approach.  Italian feminist trade unionists, for example, have
j described how, in order to participate in "masculine models of trade
; union militancy1, they had to deny their femininity. To gain equality,
' they faced internal conflicts which they poignantly describe as
; ''lacerazione (lacerations). Becoming integrated entailed feeling 'mutilated1 (Beccalli 1984: 201). They determined that it was better to change the unions, rather than changing themselves in order to seek equality within them. Parallel experiences have been reported by women who set up their own businesses, rather than work as employees in male dominated work cultures which they find alienating (Muir 1997; Rees 1992b, chapter 9). Similarly, there are studies of women in male dominated jobs who report they have to behave like a man' as a survival strategy (Breakwell 1985) and of able senior women managers who drop out mid-career from the stress of gender dissonance (Marshall 1995; Tanton 1994), There are also accounts of what women's education could and should be like (Coafs 1994), and views from women students who are angry about what they see as the masculinist culture of their education and training organisations and who seek alternative styles of provision (Taking Liberties Collective 1989).
But how feasible is it to introduce the concept of diversity into
42  Conceptualising equal opportunities
an organisation or, indeed, into training policy? How possible is it to change the value put on activities which are gendered? The next section looks at approaches to integrating models of EO, including mainstreaming, into training policies.
3 CONCEPTUALISING EQUAL OPPORTUNITIES
1 These linguistic and cultural variations in terms and their meanings are not. of course, confined to EO issues but characterise other transnational debates, such as training. Alternance' has become a universal 'European word to describe a mixture of on- and otf-the-job training in the absence of a convenient alternative in many languages. When 1 attended the meetings of the Social Partners preparing their Joint Opinion on Women and Training in 1992/3 (see Appendix III), some of the differences in understanding of EO and other terms emerged in debate: where terms did not exist in all the (then) Community languages, reference to them was avoided in the text. This raises for English speakers, of course, the intriguing question of what terms and concepts arc available :n other languages but not necessarily in English. In the absence of a term for key concepts in some languages, translators and interpreters may reach for the nearest logical equivalent, which may be wide of the mark. In my own rather unfortunate experience of German simultaneous interpretation in a presentation in Berlin, my use of the term "manpower' (as in "Manpower Services Commission') was rendered 'male virility'.
2 The EOC covers England, Scotland and Wales. There is also an EOC Northern Ireland which was set up under separate legislation.
3 It should be remembered that boys in some parts of the country, such as Manchester, benefited from a quota system designed to ensure that equal numbers of boys and girls had a grammar school education to counterbalance the fact that more girls than boys scored highly in the eleven-plus examination. This seems to have been forgotten in the current furore about 'under-achieving' boys.
4 In the UK. there are restrictions on married women gaining access to certain training provision for the unemployed (see EOC 1993).
5 Tinkering, tailoring and transforming appear to defy translation: they tend to be rendered as 'le tinkering, le tailoring, et le transforming* in EC publications that have been translated.
6 The extent to which gender monitoring is used by Training and Enterprise Councils and training providers in England and Wales is discussed in Chapter 5.