Different concepts of representation have shaped Australia's evolution as a parliamentary democracy. Currently there is renewed debate over both principles and practice of representation coinciding with a loss of faith by many citizens that their views are being represented in Australia's parliaments. Some citizens are attracted to forms of plebiscitary democracy, such as Citizen Initiated Referenda or CIR, that sideline the role of the representative and solve in this way the perceived gap between the policy preferences of policy elites and of the public. Others advocate quotas for women, reserved seats for indigenous Australians, increased representation of Australians from non-Anglo backgrounds or of those with disabilities, to make parliaments more representative.
In this chapter, I examine changing concepts of representation, the ambiguity of discursive claims of `under-representation' and dilemmas concerning increased emphasis on the embodiment of diversity within representative institutions. I conclude by discussing how parliamentary institutions such as the Senate may become more responsive to diversity as well as reflective of it. I suggest that the Senate is well positioned to become a guardian, as well as an exemplar, of more inclusive forms of representative democracy.
Changing concepts of representation
The Senate, like other parts of the Australian political system, is formally based on principles of geographical representation. There is an assumption that state boundaries define communities of interest. Australia has little history of representation based on other than territorial principles, apart from brief experiments with university seats in New South Wales and with separate seats for railway employees and public servants in Victoria. We did not follow New Zealand, a pioneer of the separate representation of indigenous people.
Geography has, however, become less relevant to political identity, despite the role of federal political structures in sustaining differences based on it. It was largely supplanted early in the century by the strong party identification that became characteristic of Australian politics. These long-established party loyalties have themselves weakened markedly over the last 20 years. There has been a decline in voter identification with the major parties, with less than one in five having strong partisan attachment.  In other words, we have arrived at a postmodern era where political identities are contingent and fragmented, where social movements have mobilised new identities alongside and cutting across the old party cleavages.
Our parliamentarians themselves articulate the ways in which these multiple identities inflect party allegiance and representational roles. One only has to look at the first speeches of senators Aden Ridgeway, Brian Greig and Tsebin Tchen this year to see how relevant embodiment has become to the way new senators perceive their roles. The Senate has become an important forum where multiple political identities are represented in addition to party and geography. It is striking that senators Ridgeway and Greig barely alluded to their party or their state constituencies in their speeches. Senator Ridgeway told the Senate: `I come to this place as a Gumbayynggir Goori, but also as a reconciliationist.' Senator Greig said: `I stand here today, on this first day of spring and in the dawn of a new Century, as a representative of the last generation of gay and lesbian people who will tolerate the injustices of the past being carried into the future.' 
When I interviewed members of the federal parliament in 1996, about half of those in my sample mentioned the importance of representing a constituency related to some aspect of their identity other than their political party or geographical constituency. Such constituencies related to age, gender, sexuality, ethnicity and so on and were nation-wide rather than restricted to one state. 
Perceptions by the major political parties of the significance of embodiment are nicely illustrated by the case of the NSW state seat of Bligh, which includes the gay heart of Sydney. Both Labor and Liberal parties have run gay candidates in Bligh in an attempt to wrest the seat from the prominent Independent who has held it for many years.
The relevance of embodiment to representation has been a matter of controversy. Political scientists have been inclined to dismiss what they call the mirror or microcosmic view of representation, arguing it has little relevance to how representation is performed. `Standing for' is not the same as `acting for', we are often reminded, and we cannot assume that those who embody group characteristics will necessarily act in the interests of that group. Margaret Thatcher is often used as an example of this—a woman who did not support measures to promote equal opportunity for other women.
On the other hand, some of these arguments concerning the irrelevance of embodiment can clearly be seen as self-interested. As the late Clare Burton pointed out, discrimination research shows that dominant and subordinate groups inhabit different cognitive universes. Dominant groups see equal opportunity and merit-based progression where subordinate groups see patterns of discrimination and bias. 
The dominant group in society (from which political scientists have usually been drawn) has always tended to suggest that their gender, race or class is irrelevant to how they perform their roles. Those who bear the markers of difference, on the other hand, have been perceived as lacking in objectivity or in other ways unsuited to acting as representatives of the community as a whole. Their identity as a member of a marginalised group results in their views being deemed inherently particularistic; indeed this expectation means, as Senator Kathy Sullivan discovered, that parliamentary colleagues can save time by not even listening. 
The idea that those who represent difference are unable to transcend their identity in order to pursue the common good clearly serves the interest of the status quo. It suggests that to undertake advocacy from the standpoint of a marginalised identity is a disqualification for public service, which can only be undertaken in a disinterested way by the dominants. 
Representatives who are not from the dominant group are also much more likely to have attention drawn to their embodiment by the media. One would be unlikely to see a newspaper headline about a male-dominated ministerial conference, such as the Premiers' Conference, drawing attention to their parental status. A newspaper story about the Commonwealth-State Ministers' Conference on the Status of Women, however, was headlined: `The mums and grannies who speak for women.' 
Recent political philosophy has placed new emphasis on embodiment and its relationship to representation. These newer approaches are epitomised by Anne Phillips' work on the politics of presence.  These approaches go beyond the idea of the parliament as the `mirror of the nation's mind'—an idea put forward by John Stuart Mill and subsequent supporters of PR—to suggest that reflection of different forms of embodiment and the life experiences associated with them, particularly experiences of subordination, exclusion and denial, are also important.
This is in part an issue of authenticity of representation—that it is difficult for members of the dominant group to represent perspectives arising from quite different life experiences. Who is entitled to speak, for example for people with disabilities? People with disabilities have been denied a democratic voice of their own and have had infantilising stereotypes imposed on them. They argue that while parents, families, carers and service deliverers may articulate a strong disability rights perspective, they cannot themselves `represent' the interests of disabled people.  Reflecting this concern, the National Disability Advisory Council, established in 1996 to advise the federal government, has a requirement that 50 per cent of members be people with disabilities. Similarly the Australian Federation of AIDS Organisations (AFAO) now has a provision requiring 50 per cent of delegates from state AIDS Councils to be HIV positive.
Apart from authenticity there is also an issue of expectations—we expect that those who share our characteristics will be more likely to understand and be responsive to us. Women politicians are more likely to be approached by women's organisations, non-Anglo politicians by ethnic community organisations and so on. There is an expectation that politicians who are different will feel obliged to represent difference, even if this is not always the case. The degree of exposure to such advocacy is in itself likely to increase awareness and sensitivity to group issues. 
Those who argue for the importance of the physical representation of difference are not suggesting that parliaments can mirror all characteristics to be found in the population. Usually they are arguing that characteristics that have been assigned significant social meaning and resulted in significantly different life chances should be present in the legislature if its representative function is to be adequately performed. This is important not only in terms of sensitivity to the distributional effects of decision-making, but also in terms of the politics of recognition, or the acknowledgment of politically salient identities present in the community.
Parliaments will never completely reflect the make-up of society and nor is that proposed. All mirrors distort and the point is not to remove all distortions, only those that produce undesirable effects in terms of representation and responsiveness.  The virtues of proportional representation (PR) as used for the Australian Senate are the scope it gives, both inside and outside the major parties, for the representation of other than territorial constituencies and the reflection of both diversity of opinion and social diversity.
Meanings of `under-representation'
I will now consider briefly the discursive strategies of those who claim to be `under-represented' in our current political system and the meanings that are bound up in the concept of `under-representation'. The concept of `under-representation' is inherently ambiguous, suggesting that the presence of members of a group will also serve the goal of representation of the interests of the group as a collectivity. Political language is often characterised by this kind of ambiguity; it is useful in sending different messages to different audiences and maximising the appeal of the slogan concerned.
I will illustrate how the slogan of `under-representation' operates in this way by reference to women—the same remarks, however, apply to other groups making claims to greater presence in parliaments.
Women have, in the 1990s in particular, successfully politicised their absence from parliaments. It is perhaps paradoxical that this has occurred at the same time as widespread questioning within academic feminism of the usefulness of the category `woman'. Despite this questioning of collective identity, and despite the historic ambivalence of the women's movement concerning parliamentary politics, the issue of the representation of women has now been taken up at every level of the political system, whether sub-national, national, regional or international. At the international level, action plans are drawn up by bodies such as the Inter-Parliamentary Union (IPU) as well as by the United Nations (UN) Commission on the Status of Women. 
If these plans were couched simply in terms of a justice argument concerning the equal right of women to participate in public decision-making, that would be relatively straight forward and unambiguous. The right of women to participate in public life on an equal basis with men is set out in a number of United Nations instruments, notably Article 25 of the International Covenant on Civil and Political Rights (ICCPR) and Article 7 of the Convention on Elimination of All Forms of Discrimination against Women (CEDAW).
The justice argument does not rely on women making a difference to public life. It simply assumes, like all equal opportunity arguments, that talent is not confined to one gender and that the absence of women from parliamentary positions is a consequence of direct or indirect discrimination. Such indirect discrimination may include factors such as the electoral system. As Arend Lijphart observes in this volume, there is much evidence showing the advantages of PR in achieving the representation of women and minorities as well as a more consensual style of politics. Discrimination may also be built into the structures of political work and political careers, through failure to accommodate family responsibilities or the privileging of specific gladiatorial styles of politics.
As I have said, the justice argument for women having equal opportunity to participate in public life is relatively straightforward. Like most justice arguments it will, however, need to be supplemented by utility arguments to convert power holders to the cause. Such utility arguments may be in terms of doubling the pool of talent from which legislators are recruited or increasing the electoral appeal of parties. Such utility arguments easily move into the terrain of `making a difference'; that is, providing a new look for parties in the context of voter disenchantment.
Utility arguments also merge into the supposed `civilising effect' of increasing the numbers of women in legislatures—that is, the relevance of embodiment to the way representation is conducted. So although the justice argument does not entail that the greater representation of women (or of any other group) should make a difference or provide partisan advantage, the suggestion is that it will enhance the appeal of the slogan.
Justice arguments are not only garnished with utility arguments, they also slide quickly into arguments about the relevance of embodiment to what interests, values, experiences and perspectives are represented. For example, the Beijing Platform for Action states that: `Women's equal participation in decision-making is not only a demand for simple justice or democracy but can also be seen as a necessary condition for women's interests to be taken into account' (para. 181). This is not a claim that the presence of women is sufficient for women's interests to be taken into account, but it is an argument that the absence of women means that their interests will be ignored or overlooked. This is a relatively difficult argument, although it can be approached through a series of steps relating to the gender roles allocated to women, the specific life experiences arising from these roles, the perspectives and values associated with these roles and finally interests deriving from the social and economic consequences of gender roles.
The suggestion that the presence of women (or of other groups) will enhance deliberative democracy by introducing perspectives derived from distinctive experiences and increasing sensitivity to the distributional impact of decisions (in this case their impact perhaps on those with day-to-day caring responsibilities) is less contentious than the direct link drawn between embodiment and representation of interests.
However, at this stage another element of `under-representation' comes into play, the symbolic effects of presence or absence. In other words, the question of the representativeness of the legislature and the effects of this both on the group concerned and on the political system as a whole. Special arrangements for the representation of indigenous peoples are often seen as an important symbolic recognition of their indigenous status. The presence of women in parliament may also have an important effect on the status of women outside parliament. Whereas the suffragists argued that winning the vote would be sufficient for this purpose, elevating the status of women and increasing respect for them, today the argument has shifted to the need for actual presence in parliament. Such symbolic effects are frequently referred to in terms of the politics of recognition.
Another symbolic effect is the motivational or role model effect—the idea that presence in public life and the visibility that accompanies it will raise the aspirations of other members of the group—the `girls can do anything' effect. For example, one recent writer in the Adelaide Advertiser has described how Senator Natasha Stott Despoja's youthfulness and willingness to engage with media relevant to young people made her `a role model for me, as she is for many young Australians'.  Stott Despoja speaks of how Senator Janine Haines had the same effect on her when she was a student and Senator Haines gave a visiting lecture on issues she could connect with, such as the treatment of sole parents.
A very different kind of symbolic effect, but yet another that can be wrapped up in the slogan of under-representation, relates to institutional legitimacy. The idea is that the legitimacy of parliament will be undermined if significant sections of the community appear to be locked out of it.
Such a threat to legitimacy assumes the mobilisation of group identity, like that around the issue of women's `under-representation'. On the other hand, such mobilisation currently falls short of, for example, denying the legitimacy of laws made by legislatures in which women are largely absent. There is no large-scale campaign of civil disobedience or refusal to pay taxes associated with groups currently said to be under-represented in parliament.
Table 7.1 Meanings of political representation
||collectively mediated experiences
||effects on status of group
||effects on aspirations
||legitimacy of institution
|Equal right to represent
||to participate in public decision-making
||not to be discriminated against by structures of public life
||increase pool of talent
While the rich ambiguity of the slogan of `under-representation' has definite discursive advantages, the element that ties representation of group interests to physical presence of members of the group poses problems. As Anne Phillips has observed, there is the danger that if too much emphasis is placed on the relationship between embodiment and interest representation, this will reduce the pressure on all politicians to be responsive to, and to represent, diversity. It may also restrict the role of those with group characteristics to representing their group or bring undesirable pressure on them always to act as a group representative—the token woman or Black, for example.
It was argued, for example by the Royal Commission on Electoral Reform in New Zealand, that while the Maori seats in New Zealand provided a Maori voice in parliament, they had the effect that only the Maori parliamentarians were seen as responsible for representing Maori interests: `The system of separate representation encouraged the non-Maori majority to regard Maori concerns as the sole preserve of separately elected MPs.'  The same Royal Commission argued that the most effective way of making all parliamentarians responsive to Maori interests was through a form of PR. Responsiveness is enhanced because seats are dependent on maximising votes from all sections of the community.
The stress on the relationship between embodiment and interest representation may have other perverse effects. For example, the image of the Australian Democrats as middle-class teachers or members of the `chattering classes' has posed a barrier to their being seen as representatives of blue-collar workers or of `battlers'. Embodiment gets in the way of the Australian Democrats being seen as the representatives of ordinary workers or ordinary mums and dads, regardless of policies. This is a matter of considerable chagrin for Democrat senators, who feel their economic policies are less directed to the `big end of town' and more concerned with ordinary workers than those of the major parties.
The most effective representatives may indeed be those from outside the group concerned because of the perception discussed earlier that those who bear the marker of difference are inherently self-interested, while the dominants are more disinterested in their advocacy. This has been argued by Dennis Altman in relation to the representation of gay interests.  The political risks of visibility are another consideration for non-heterosexual parliamentarians or for parliamentarians disclosing other intimate details of their lives, such as experience as a single mother. However, while effective advocacy may come from outside the group concerned, if we take this argument too far, we may end up privileging the voice of experts over those with lived experience.
From representation to responsiveness
When we identify `under-representation' as a problem, one of the most important things we are trying to do is to make parliament more responsive to groups in the community that have been ignored or overlooked in the past. While physical presence of members of these groups in parliament may be part of the answer, it is not the whole answer. Structural changes may be at least as important in ensuring that parliamentarians speak to new constituencies, even if they cannot speak for them. Examples of such changes include the standing committees found in European parliaments that help raise awareness of gender issues—for example, committees on women's rights in the Irish, Spanish and European parliaments and on equal opportunities for men and women in the Belgian and Luxembourg parliaments. These committees have varying mandates, including in the case of the Belgian Senate looking inwards at the working of the parliament and issues such as family-friendly sitting hours and the gender balance of expert witnesses. 
Parliaments more responsive must also include broadening community access to the parliamentary process. Sections of the community lacking in direct parliamentary representation, or groups only just evolving a group identity, may thus achieve presence and voice in the deliberative process. As Elaine Thompson and Ian Marsh also discuss in this volume, the Australian Senate has played an important role in broadening the participation in the deliberative process of those whose lives will be affected in specific ways by legislative proposals. It is crucial to ensure that those who will be disproportionately affected by legislation have an opportunity to voice their concerns rather than simply being at the mercy of majoritarian decision-making.
Senate committees have the potential to take the lead in what the OECD calls `strengthening government-citizen connections'. This does not mean the way governments relate to citizens as clients or consumers of services, as in the now fashionable citizen charters, but rather how governments interact with citizens in the development and design of policy. 
Parliamentary committees hold hearings around the country, enabling less mobile sections of the community, including women with family responsibilities, to participate without the costs of travel. When a Senate committee inquired in 1995 into outworking in the garment industry, where most employees are women from non-English speaking backgrounds, advertisements were broadcast on ethnic radio stations and submissions were taken through the telephone interpreter service. This is not to say that the committee system always works as it should. Recently, such as during the inquiries into the new tax system, we have had a number of reported instances of senators speaking at, rather than speaking to, community representatives—that is, engaging in partisan point-scoring rather than community dialogue.
Parliaments will never mirror all elements of the community, particularly given the complexity of modern society, so building the capacity of citizens to represent themselves to parliament is an important element of the practice of representative democracy. The public funding of advocacy organisations, as developed in Australia over the last 20 years, is intended to strengthen weak voices, the voices of those who would otherwise lack the resources to make themselves heard.  We need to take these forms of extra-parliamentary representation, particularly community-based peak bodies, much more seriously as institutions of representative democracy.
This means paying much more attention to issues of representation and accountability within such bodies as well as to the process of community dialogue over policy development. It means more attention to the methodology of consultation and to the accountability of government for the relationship between policy consultation and policy decisions. It also means government acknowledging its responsibility for ensuring that policy consultation does indeed contribute to representative democracy by expanding the range and inclusiveness of deliberative forums.
Gianni Zappal, a former Parliamentary Fellow, has written about how multicultural organisations have served as a bridge to ensure the responsiveness of political representatives (regardless of ethnicity) to the needs of their ethnic constituents.  They also serve to enhance access by ethnic Australians to broader deliberative structures through the representational role they play in policy consultation and committee hearings.
If the funding of community-based peak bodies is to achieve its purpose, of ensuring effective representation of community interests to government and to parliament, governments need to exercise restraint and to tolerate criticism. They also need to tolerate an inconvenient plurality of representative bodies, reflecting diverse perspectives within each community sector. Unfortunately community-based representative bodies have been under increased pressure to conform to government convenience and government agendas rather than to represent their constituents. If governments try to control what is said by such organisations they end up hearing only what they want to hear, not what they need to hear if all elements of the community and of community concerns are to be represented.
There is at present little accountability for how consultation processes are used and how they feed into final decision-making. We do not have a requirement that Cabinet submissions specify views presented during community consultation processes and how they relate to final recommendations. Nor do we require feedback to community groups concerning the use made of their contributions. The failure to set up adequate processes for community dialogue over policy development has led scholars such as John Uhr to suggest that there should be parliamentary oversight to ensure consultation remains meaningful as an institution of representative democracy. 
The Senate should play a leading role both in modelling inclusive forms of deliberation over policy development and in overseeing their adequacy across government. The Senate has been evolving in important ways as an institution of representation and is well positioned for further changes. These must reflect the changing nature of an electorate no longer defined simply by geographic location or party loyalty, but by multiple identities and claims for recognition.
Ian McAllister and Clive Bean, `Long-term electoral trends and the 1996 election', in Clive Bean et al., eds, The Politics of Retribution: the 1996 Federal Election, Sydney, Allen & Unwin, 1997.
First speech of Senator Aden Ridgeway, Commonwealth Parliamentary Debates (CPD), 25 August 1999; first speech of Senator Brian Greig, CPD, 1 September 1999.
See Marian Sawer, `Mirrors, mouthpieces, mandates and men of judgement: concepts of representation in the Australian Federal Parliament', Papers on Parliament, no. 31, June 1998.
Clare Burton, Gender Equity in Australian University Staffing, Canberra, AGPS, 1997, pp. 3–5.
`The dawning realisation came to me slowly and painfully: many of my male colleagues had not actually heard what I had said in the previous 12 years, whether in Parliament, in the joint party room or in committee meetings. Nonetheless, they all thought they knew what I had said …', Kathy Martin Sullivan MP, `Women in Parliament—Yes! But what's it really like?' Papers on Parliament, no. 22, February 1994, pp. 22–23.
See Linda Trimble, ` “Good Enough Citizens”: Canadian women and representation in constitutional deliberations', International Journal of Canadian Studies, 17, Spring 1998, p. 149. For a recent Australian example of this kind of thinking see the claim in a recent letter to the Canberra Times that an Aboriginal candidate (Mick Dodson) would be unsuitable as the president of Australia because he would only be representing `a very small sectional interest' making up less than 3 per cent of Australia's population (letter to the Editor, 17 July 1999).
Michelle Gunn and Megan Saunders, `The mums and grannies who speak for women', Australian, 27 March 1998, p. 5.
Anne Phillips, Politics of Presence, Oxford, Eng., Clarendon Press, 1995. See also Melissa Williams, Voice, Trust and Memory: Marginal Groups and the Failings of Liberal Representation, Princeton, NJ, Princeton University Press, 1998.
C. Newell, `The disability rights movement in Australia: a note from the trenches', Disability and Society, vol. 11, no. 3, pp. 429–432.
See Manon Tremblay, `Do female MPs substantively represent women? A study of legislative behaviour in Canada's 35th Parliament', Canadian Journal of Political Science, vol. 31, no. 3, 1998.
I am indebted to Barry Hindess for this point.
See Fourth UN World Conference on Women, `Platform for Action', New York, UN Division for the Advancement of Women, 1995, Section G; The Interparliamentary Union, `Plan of Action to Correct Present Imbalances in the Participation of Men and Women in Political Life', Geneva, IPU, 1994.
Advertiser (Adelaide), 15 June 1999, p. 18.
Towards a Better Democracy: Report of the Royal Commission on the Electoral System, Wellington, Electoral Commission, 1986, pp. 90–93.
Dennis Altman, `Representation, public policy and AIDS', Paper to Academy of Social Sciences/ Reshaping Australian Institutions Workshop on Representation, Canberra, Australian National University, December 1998.
See the Repertory of Parliamentary Committees Responsible for Equal Opportunities for Women and Men in the EU Member States and the European Parliament, published by the Advisory Committee on Equal Opportunities for Women and Men of the Belgian Senate, 1997.
See the Summary Record of the OECD Working Group Meeting on `Strengthening Government-Citizen Connections', Paris, 17–18 June 1999.
Another thing that can serve to enhance responsiveness of government to `have nots', and that has long been part of the Australian political system, is compulsory voting. Where voting is voluntary there is unequal turnout that is `systematically biased against less well-to-do citizens'. See Arend Lijphart, `Unequal participation: democracy's unresolved dilemmas', Presidential Address, American Political Science Association, 1996, in American Political Science Review, vol. 91, no. 1, pp. 1–14.
Gianni Zappal, Four Weddings, a Funeral and a Family Reunion: Ethnicity and Representation in Australian Federal Politics, Canberra, AGPS, 1997.
John Uhr, Deliberative Democracy in Australia: the Changing Place of Parliament, Melbourne, Cambridge University Press, 1998.