When Nicolas Baudin left the Normandy port of Le Havre in October 1800 bound on a voyage of discovery to the Southern Lands, he set out in the expectation of attaining the status of James Cook (Horner 1987: 80; Allorge 2003: 589). Four years later, despite all that had been achieved, his expedition would return to France shrouded in disgrace. It was a dramatic reversal of fortune that was all the more remarkable considering the lofty ambitions that had inspired the voyage's conception and planning, the large amount of money that had been invested in the expedition and, last but not least, the considerable esteem in which its leader had been held. The government response to a voyage deemed to have gone wrong was an eloquent silence, but, if the expedition received little by way of official recognition, its memory was nonetheless preserved – mainly in the writings of the expeditioners themselves, who were anxious to justify their own achievements, and in the reports of their defenders in the scientific establishment, whose own interests lay in providing them moral support. In the years that followed the expedition's return, and far into the following century, such self-interested narratives, which knowingly denigrated Baudin or underplayed his achievements in order to enhance those of their authors, were by far the most influential source of information on the voyage. It is precisely the history of these stories that concerns us here and how the misrepresentations upon which they were based became interwoven into the history of the expedition to the Southern Lands.

Introduction: Carol Bacchi and Joan Eveline

This chapter explores the proposition that how gender is conceptualised has implications for the efficacy of gender mainstreaming and gender analysis as change processes. It makes the case that gender is a contested concept, that it can be defined in ways that reproduce male, white, able-bodied privilege, or in other ways that reduce certain inequalities. In particular it develops in some depth our suggestion that both gender analysis and gender mainstreaming be conceptualised as always incomplete, thus ‘unfinished business’, rather than as fixed categories of analysis. The goal here is to shift attention from the idea that we may ‘have’ either a gender or a gender mainstreaming policy/program to the continual effort involved in fixing or ‘doing’ the gendered subject or in giving ‘content’ (meaning) to gender mainstreaming.

The chapter begins with a brief history of ‘gender’ as a political concept within feminist theory. It explains how the theorising of masculinities and the growing attention to differences among women put the utility of the concept in dispute, and how the 1970s idea of a sex/gender distinction was found wanting. We make the case that part of the problem with the category ‘gender’ is the common way in which it is conceptualised as a part of a person rather than as a process that is ongoing, contested and incomplete. Thinking about gender as a verb, or as a gerund (gendering), we suggest, is more likely to capture how gender differentiation is continually ‘done’ through discursively-mediated institutional and organisational processes, including policymaking.

Introduction: Carol Bacchi and Joan Eveline

Previous chapters have made reference to the need to rethink policy as a creative (productive or constitutive) process. The major purpose of this chapter is to clarify what this means and to illustrate the usefulness of this way of thinking about policy for studying gender mainstreaming and gender analysis. The specific focus is ‘gender proofing’ in Ireland and ‘gender impact assessment’ in the Netherlands.

The underlying proposition in thinking about policies as productive, or as constitutive, is that policies and policy proposals give shape and meaning to the ‘problems’ they purport to ‘address’. That is, policy ‘problems’ do not exist ‘out there’ in society, waiting to be ‘solved’ through timely and perspicacious policy interventions. Rather, specific policy proposals ‘imagine’ ‘problems’ in particular ways that have real and meaningful effects. Hence, to understand how policies operate requires that we ask of policy proposals ‘What's the Problem represented to be?’. This question forms the starting place for Bacchi's (1999; 2009a) novel method of policy analysis (elaborated below), captured in the acronym WPR.

The proposition that ‘problems’ do not ‘exist’ ‘out there’ in society does not ignore or downplay the full range of troubling conditions, including the subordination of women, that characterise social relations. Instead, it insists that how ‘problems’ are represented in policies – how they are discursively produced – affects the particular understanding given to those conditions at points in time and space, and that these understandings matter.

Introduction: Carol Bacchi and Joan Eveline

The title to this chapter indicates the major goal of the Gender Analysis Project: to identify the factors that could create gender analysis as a long-term process of emergent changes to the asymmetrical power relations between women and men. The sub-title, ‘testing the water’, indicates that it was written in the early stages of the project. However, it is important to note that the papers are not strictly chronological in their production. The Chief Investigators (Bacchi and Eveline) took turns as ‘lead authors’ and Chapter 4, with Eveline the lead author, was actually completed before this chapter, with Bacchi the chief author. Insights from Chapter 4 are therefore incorporated in this chapter. The resultant analysis represents a cross-fertilisation of ideas, as is the nature of collaboration.

The chapter emphasises the importance of involving policy workers actively in practising gender analysis (that is, in applying gender analysis guidelines), a significant learning outcome for the project (Chapter 12). We also suggest that it is useful to conceptualise and to talk about social change in a different way, as the unpredictable effect of complex and continuous processes, occurring ‘somewhere in the middle’ and therefore always involving ‘unfinished business’ (see Introduction; see also the discussion of the ‘rhizomatic’ in Chapter 6). This approach directs attention to the everyday work practices that reproduce gendering as an always- incomplete relation of inequality.

Introduction: Carol Bacchi and Joan Eveline

In Chapter 5 we identify the ‘project trap’ – subservience to wider policy objectives – as a major constraint on potentially transformative gender analysis processes. There we show, for example, how privatisation of health care (Armstrong 2002) increases the caring work that those marked as ‘women’ will have to do, reinforcing the conventional domestic division of labour. It follows that, in order to be transformative, a gender analysis must be able to scrutinise underlying premises in policy proposals, showing how they can be gendering practices that produce gendered beings and gendered relationships.

A major factor deterring critical analysis of this type is the insider status of those performing gender analysis, since policy workers are obliged to an extent to perform assessment tasks as laid out by the government holding office (Chapter 11). To loosen the ties of this limiting ‘insider’ status and hence to enable policy workers to become more critical of government policies, some theorists emphasise the importance of forms of community involvement as a policy practice (see Chapter 1, p.30). The argument here is that members of the lay public may provide contesting views to perspectives shaped largely by business interests and senior management. As mentioned elsewhere (Chapter 3), one of the chief purposes of Linkage Grant projects is theory testing. Hence, the project organisers in South Australia constructed a qualitative research exercise to consider the extent to which community consultation might encourage the development of more transformative gender analysis processes.

This book is about change and how it happens. It draws upon the research and experiences of its contributors to provide glimpses into the challenges facing those who care to produce more egalitarian relationships between and among women and men, and into the ‘spaces’ found within constraints to advance such an agenda. Its specific topic is gender analysis, a form of policy analysis associated with the equality policy initiative called gender mainstreaming.

The setting for the production of the book involved a large Linkage Grant project funded by the Australian Research Council (ARC), entitled ‘Gendering impact assessment: A new framework for producing gender-inclusive policy’. The authors of this volume, Carol Bacchi and Joan Eveline, located in South Australia and Western Australia respectively, were the Chief Investigators for the project. Our Linkage partners included, in South Australia, the Office for Women (OFW) and, in Western Australia, the Office for Women's Policy (OWP) and the Health Department. A number of other public sector agencies in both states were participants (for details see Chapter 3). The project involved a PhD student, Karen Vincent, and several Research Associates (Jennifer Binns and Susan Harwood in Western Australia; Katy Osborne, Zoe Gordon and Catherine Mackenzie in South Australia), who became co-authors of some papers and reports, some of which are published here.

The goal of the project was to design gender analysis procedures appropriate to the respective contexts of the public service in the two states.

Introduction: Carol Bacchi and Joan Eveline

This chapter applies the concept of ‘doing’ to the practices of feminist researchers. Under scrutiny are the ways in which unexamined presumptions about the main business of gender mainstreaming as gender equality foreclose consideration of the lives and experiences of specific groups of women, here Aboriginal and Torres Strait Islander women. Following from Chapters 10, 11 and 12, the chapter emphasises the critical importance of collaborative spaces to the character and shape of egalitarian politics.

The chapter highlights the continuing dispute in feminist communities about whether or not the term ‘diversity mainstreaming’ better reflects current sensitivities to differences among women, commonly described as an ‘intersectional’ sensitivity. It shows how the South Australian research team in the Gender Analysis Project dealt with this issue, deciding to include ‘race and cultural analysis’ within a gender analysis guide (SAGA).

This decision, as we describe, was not, as might first appear, a compromise position. Rather, non-Aboriginal members of the group accepted that Aboriginal women were best placed to articulate a political vision of use to their communities – a vision based on the identity of those communities. This acceptance compelled those non-Aboriginal members to rethink assumptions about the obviousness of gender as an analytical priority. This rethinking is an outcome of what we describe in the book as the process of reflexivity – finding ways to reflect critically on one's own starting points for thinking (all the while recognising that there are no agentic subjects who can invariably avoid traps of discursive positioning).

Introduction: Joan Eveline and Carol Bacchi

The gender analysis project gathered pace in its last few months. Developments that are noted briefly in earlier chapters, such as the topic of this chapter, the Indigenous Electoral Strategy (Chapter 3), and the inclusion of ‘cultural analysis’ in SAGA (South Australian Gender Analysis; Chapter 3) acquired their full significance as we began to reflect on the project as a whole and on what we had learned (on SAGA and ‘cultural analysis’, see Chapter 13). We also began to reflect critically on the relative ‘success’ of the project and more broadly on how change occurs, or fails to occur (Chapters 11 and 12). In tune with the perspective developed in this book, our contributions on these topics represent our current thinking about the complex interactions we have studied.

In analysing the Indigenous Electoral Strategy, the chapter discusses further the issue raised in Chapter 9, that any understanding of gender used in our policies incorporates a particular cultural base. Arguments for ‘gender mainstreaming’, for example, rest on the assumption that highlighting ‘gender’ as the primary category will have similar effects in differing cultures, a contentious claim, as we proceed to discuss.

The chapter examines a Western Australian project in which Aboriginal policymakers challenged this supposition of cultural neutrality. They argued that the understanding of gender used in western societies has privileged white women's interests over those of Aboriginal people. Consequently, they refused ‘gender equity’ as a term to use in their project of increasing Indigenous participation in local government.

Introduction: Joan Eveline and Carol Bacchi

There is considerable research showing that organisations, including government agencies and the policies they produce, consider gender irrelevant to their core business. The gender mainstreaming of policy is designed to challenge such an assumption, using the argument that mainstreaming gender can ‘transform’ the ubiquity of genderblind policies. Various countries, as this book (among others) shows, have developed particular methods and tool kits for transforming the outcomes of their policies, in order that those policies take seriously the relevance of gender.

This chapter underscores the question of gender (ir)relevance, and how it is produced by complex organisational practices. It turns to feminist organisational theory to analyse the institutionalised practices that construct and organise policy priorities (see Chapter 6). In agreement with Benschop and Verloo (2006), we argue that effective gender mainstreaming cannot be achieved without attention to the specific organisational sites in which policy is developed and implemented. The chapter draws on the ‘turn to practice’ in organisational studies, and feminist strategies of ‘sudden seeing’, to consider what our insights from the gender analysis study might offer future interventionist projects.

Western Australia furnishes most of the examples used in the chapter, although general descriptions of the project's aims and challenges apply also to South Australia. The starting point for the chapter is our finding in Western Australia that it was only through doing the gender analysis in their organisational contexts that policy actors came to see the relevance of gender to policy.

Introduction: Carol Bacchi and Joan Eveline

This chapter offers additional reflections on the ‘learnings’ that emerged from the Gender Analysis Project. With a particular focus on the South Australian experience, it outlines how the shared practice of collaborative discussion within the project's reference group (which consisted of the university research team; representatives of the industry partner, Office for Women; and representatives of the three participating agencies) encouraged reflexivity among participants. Reflexivity here refers to an ability and willingness to examine one's own presuppositions and to take on board novel perspectives. Becoming reflexive, we argue throughout, is a subjectivising effect of the practices in which we engage. Practices that focus on shared, interpersonal exchange and discussion promote the production of reflexive modes of being and thinking. That is, practices that foster a heuristic approach (learning by doing) in tough interactions with similarly committed but questioning colleagues, can promote reflexivity.

The Gender Analysis Project in South Australia brought together feminist researchers and policymakers, mainly women, who shared a commitment to redressing gender inequality, although not everyone would have agreed about what exactly this entailed. The concept of ‘mining the space’ in the title refers to the determination of group members to work through differing perspectives and to overcome blockages within an institutionally sanctioned space.

The chapter describes how the regular meetings of the reference group set up to oversee the project created the space and time required to examine and debate the contested meanings of gender and gender relations.

Introduction: Joan Eveline and Carol Bacchi

This chapter examines the primary organising processes which produce the meanings of policy statements. It outlines and inspects the social power circumscribing these policy statements, the relations of power and resistance involved in such statements, and their effects on those subject to the policy.

The previous chapter outlined how the WPR approach concentrates on the constitutive effects of existing or proposed policies, showing how the characterising of policy ‘problems’ within those policies or proposals (what the ‘problem’ is represented to be) ‘shapes’ (or constitutes) people as particular kinds of subjects. This chapter pursues the point that policies elicit subjectivities, rather than determine them (Dean 1999: 32). It highlights the always-incomplete nature of subjectification processes, emphasising that the subjects of policy are always more than the products of policy regulation, whether explicit or implicit, as is the case in problem representations. In this view political subjects, both those who ‘do’ policy and those to whom it is ‘done’, are both subjected and resistant to policy discourses. A particular focus of this chapter and of the book is how ‘doing’ policy both produces and enables the subjectivities of those who analyse and develop it, including ourselves as researchers.

The influences of feminist poststructuralism and recent organisational theory shape the propositions in this chapter. It was prompted by our wish as authors to fill a gap in our earlier and later chapters.