‘Circumstantially childless’ women (Cannold 2000) are those who have seen themselves as having a biological child or children at some point in their lives, but have come to or are approaching the end of their reproductive years without giving birth, primarily for social rather than biological reasons. They are in the unusual — but not uncommon — position of being neither ‘voluntarily childless’ (since they would like to have a child), nor ‘involuntarily childless’ (since they are, as far as they know, at least initially physically capable of conceiving and bearing a child, although they may be later affected by age-related infertility). It is impossible to establish statistically how many women fall into this specific category of childlessness, since there is no distinction made in official records between those who are childless (or ‘childfree’) by choice and those who are circumstantially childless, but it does appear to be a growing phenomenon in Western countries (Hadfield et al. 2007, McAllister & Clarke 1998). Academic research in the field of infertility generally is neglected (Van Balen and Inhorn 2002, pp. 5–6). In this particular instance of childlessness it is particularly sparse, and the experience is not well understood.

This article draws on a psychoanalytically informed psychosocial study of the experience of circumstantial childlessness in 26 women in Aotearoa/New Zealand (Tonkin 2014). I used semi structured individual and group interviews, and participant-produced drawing to elicit rich, nuanced accounts that were layered across one another in analysis. In general terms the study’s purpose was to generate a better understanding of this experience. More specifically, it aimed to conceptualise the relationship between fantasies of maternity and a child, and a sense of loss and adaptation to it, both of which were common threads running through participants’ narratives. I have written elsewhere (Tonkin 2012) about the ways in which I conceptualised many of the women in this study as being ‘haunted’ by these fantasies, and in this article I look more closely at these fantasies, through a focus on excerpts from the narrative of one participant, whom I have called ‘Deborah’. My intention is to try to make sense of aspects of Deborah’s experience that seem — on face value — to be nonsensical. I read her words through the frames of Hollway’s work about the development of a maternal subjectivity (Hollway 2015), and of Ettinger’s concept of the ‘matrixial’ (Ettinger 2006b). I conclude by suggesting that ‘fantasy mother’ may be a dimension of motherhood that has an intense personal reality for circumstantially childless women, even though they have not embodied motherhood, and their ‘maternity’ is unseen and unrecognised by others in their social worlds.

Elaborating a fantasy of maternity

At the time of the interview Deborah was 45, and saw the opportunity to have a biological child as something that was no longer possible for her. She spoke of her perception of a rising risk of having a child with disability because, having met in their early 30s, she and her husband were already both in their mid 30s by the time they came to make the decision about children. The primary reason she gave for the choice they had made not to have a child was related to her older sister, whom she described as having ‘complex disability’. Deborah been a carer for her sister throughout her childhood, and in the interview she described feeling that on one hand she ‘could not do that again’ if they had a child with a disability, and on the other that she and her husband had been unwilling for her to become pregnant and have testing for abnormality, because they were ‘unable to deny a child life (once that life had begun) for failing to be our view of “perfect”’.

Deborah’s response in the extract below was prompted by my question about why making the decision not to have children was difficult for her:

[It was] incredibly difficult. And is still difficult. […] It was difficult because [pause] um [pause] I think it was difficult because [pause], I think it was difficult because [long pause] I had never [pause], and in a way still haven’t, um, relinquished a, sort of [pause] sense of myself as a mother [trembling, quiet voice, crying]. Um [pause], [it’s] very hard to describe.

Deborah is a highly articulate woman, whose professional career has trained her to be skilled in using language to express complex ideas. These skills were demonstrated throughout her interview, where she spoke almost entirely with a notable clarity and fluidity, choosing her words very carefully. The tears and faltering hesitations she made here were a significant exception to her talk in the rest of this interview. My analysis took account of both what is said and what is not said, of the tone of speaking, and of what might be indicated in a change in delivery, and these alterations to her delivery provided useful points for analytical scrutiny in themselves (Mazzei & Jackson, 2009). Deborah had spoken a lot about how important her feeling of herself as being ‘nurturing’ was, and how painful it was that others did not recognise this quality in her. When I asked her why it was hard to decide not to have a baby, why did she not continue in the strong fluid style of speaking she had used throughout the interview and simply say ‘it was hard because I am a nurturing person, and it saddens and angers me that I am not recognised as such by others because I am not a mother’? Why did she struggle, and begin to cry, and eventually find the extraordinary phrase ‘the sense of myself as a mother’?

I consider Deborah’s words in the light of Lisa Baraitser’s (2006, p. 22) reflections of being a mother. She writes:

(M)aternity is an experience that I maintain is impossible to anticipate in advance, one that unravels, as it proceeds, and that one is always chasing the tail of, never become expert at, or even competent, and that always eludes our attempts to fully understand it.

If Baraitser is right, and maternity can only be experienced as it is lived, then what is this ‘sense of myself as a mother’ that Deborah experienced without becoming a mother; a sense that this otherwise articulate woman struggled to find words for? Her use of a present tense — as opposed to a past perfect one, conditional on her having had a child; the ‘sense of how I might have been as a mother’ — suggests that in some way Deborah experienced herself to be already a mother. Her use of the word ‘relinquish’ suggested that, since she would not now embody her maternity, she saw herself as needing in some way to abandon a maternal aspect of herself that already existed.

Deborah’s ‘sense’ makes no sense logically — she does not have children biologically or socially — but her words were unequivocal, so how might they be interpreted? Quite by accident, I encountered further data that illuminated this question. Some two thirds into the project I wrote a paper (Tonkin 2012) that addressed the ‘haunted’ quality of the relationships participants had with their fantasy child. I quoted Deborah in it, and since she had requested that I send her anything published out of the research, I sent her a copy of the article. She replied:

I do feel the paper gives voice to something I feel. I read it with gentle tears. I can hear the voices of Alisdair James and Elinor Elizabeth too, as though they are ‘down the back’ (the term we used at home when I was a child to mean playing on the farm somewhere distant from the house but never far enough away we couldn’t hear the cowbell signalling it was time to come home for a meal or some other reason). We didn’t ever name my ghost children in the interview; like other women I do name my unborn children too.

Deborah’s response said a lot about the nature of the ‘absent presence’ of her fantasy children in her life. These children whose voices she ‘hears’ had names that appeared to be carefully chosen, but in giving permission for me to use them, Deborah explained that they would not be identifiable since they ‘are personal to me (no one else, not even my husband, knows them)’, highlighting the very private, often secret nature of these relationships. She added:

(I)n some ways using the names gives my ‘down the back’ children a validity or reality I’m pleased to gift them (and myself). […] I realise that I continue to ‘mother’ Alisdair and Elinor because I give them life in my head. As I age I seem to think of them less often but some events certainly bring them back into a ‘ghostly’ (in the sense of ‘hazy’) focus–and interestingly it is more their voices than a visual picture of them.

For Deborah, my use of the names of her ‘unborn children’ gave a ‘validity or reality’ that they had previously lacked when they existed only in the privacy of her imagination; it suggests that for her, these imagined children who already had an existence ‘in my head’ were called into a further existence as part of the social world by my naming them in my work. Perceiving this as a ‘gift’ that she and I could give them is reminiscent of the idea of ‘giving birth’, and she underlined this by speaking of giving them life. Although she could not give these imagined children life through giving physical birth to them as embodied infants, she nevertheless appeared to understand herself to have given them a continuing life by creating them in fantasy, and maintaining their existence in her thoughts. She saw it as a gift to them to make them more real in this way, and to herself to honour the maternal subjectivity she claimed in continuing to ‘mother them’. Although she was ‘childless’, Deborah perceived herself to be in some sense a mother with a relationship to the children of her fantasy.

Her use of the words ‘down the back’ elaborated this fantasy powerfully in several ways. She positioned her fantasy children within her broader family through reference to a term whose relevance and meaning everyone in her family would apparently understand. In this way, she highlighted the existence of the children within her own locus of belonging in her family of origin; for her they were part of her family culture. The term suggests that for Deborah, even if others in her family did not know of the ‘existence’ of Alisdair and Elinor, her fantasy children were not only hers, but also members of her wider family; for her they too had a claim on the meanings of ‘down the back’ for that family, just as her embodied nieces or nephews do, even if no one else in the family knew of their existence.

‘Down the back’ gave a sense of ‘where’ they ‘are’ spatially for Deborah; it is a term that evoked a place that was always intimately connected to ‘home’ and to the day-to-day events of the family, and yet at the same time was distant and unseen (though within earshot). One senses that in Deborah’s family, ‘down the back’ was a place where the children were understood to be mostly happily and safely engaged in playing together while their parents were busy with the other parts of their lives. In envisaging her fantasy children in this place, Deborah elaborated her ‘sense’ of how her children ‘exist’ in relationship to her; co-existent with her life, unseen, but ‘heard’, distant but safely within the bounds of the intimacies and sense of belonging of ‘at home’.

The transition to motherhood

How might the development of this apparently nonsensical ‘sense of myself as a mother’ be understood theoretically? The transition to motherhood is often discussed as a process that begins at conception, and continues as a woman moves through her pregnancy, anticipates the birth of her baby, and accommodates the huge physical, psychological and social changes that being a new mother creates (Miller 2005, Thomson et al 2011). Part of that adjustment, after the birth, is coming to terms with the embodied reality of motherhood and her baby — although of course they have been embodied in a different way throughout the pregnancy — and accommodating the distance between them and the fantasies of maternity and a child that have developed. This is not to imply that embodied ‘reality’ takes the place of fantasy after the birth of a child, because fantasy continues to permeate the ways a woman takes up a maternal subjectivity. When women do not go on to have a child, their maternal fantasies have the potential to continue unchecked by the embodied realities of a child or motherhood. This works both ways, since women who do go on to have a child and find the experience difficult may fantasize (perhaps in Winnicott’s (1971) sense of ‘fantasying’) about how life might have been if they had not had a child; a fantasy that may play itself out in their daughters’ ambivalent maternal fantasies.

In this view of the development of a maternal subjectivity, ‘being’ a mother hinges on the embodiment of a child and, to a lesser extent, the embodied practices of caring for him or her. This defining factor is a little blurry — is one a mother if one’s only baby has died, or if one is an adopting or surrogate mother with no further contact after birth, for example? — but the reality of a baby’s embodiment is still the key factor, even if the definition of ‘motherhood’ is extended beyond ‘biological’ mother to encompass “genetic, gestational, and carer” mothers, as Welldon (2006) has done.

There is no space in these framings of ‘mother’ for Deborah’s “sense of myself as a mother”. It only makes sense if I step beyond these four dimensions of maternity to foreground the role of conscious and unconscious fantasy in the development of maternal subjectivity, and suggest ‘fantasy mother’ as a further dimension of motherhood. In the section that follows I discuss Hollway’s work on the development of a maternal subjectivity and Ettinger’s concept of the matrixial to suggest some theoretical underpinnings for this proposal.

Intersubjectivity, and archaic re-membering

In her psychosocial study of identity development in new mothers, Hollway (2015, 2006) drew on Winnicott’s (1958) emphasis on the dynamic interplay between mothers and their children, and the mutual identity development that occurred through this. Winnicott theorized that the infant first begins to develop and organize its sense of self within the psyche-soma of its mother, intersubjectively. Hollway argued that a new mother identifies both with her baby and her mother through the ‘vestigial experience’ (Hollway 2006, p. 65) of her own infantile and child state in relationship with her mother, even though as an adult she has separated to a greater or lesser extent and developed differentiation from her as she grew. She writes that becoming a mother positions women as a ‘generational pivot’ because they are able to draw on ‘memories’ of this process in their own infanthood and in this identification, ‘both with their mothers (as babies) and their babies (as mothers)’ (Hollway 2009, p. 4). Of this process Stern (1999, p. 181) writes:

(T)he mother’s stored memories or memorial fragments (…) include both sides of her interaction (…) with her own mother when she was young: the parts that she experienced directly as a baby, while interacting with her mother, and the part of her mother’s experience of interacting with her that she experienced empathically (via imitation and primary identification).

Raphael-Leff (2009, p. 2) provides a clue to the some of the possible mechanisms of this process when she writes that as a mother of a new baby a woman is pulled back in memory and fantasy because of her prolonged exposure to her baby’s ‘raw feelings’ and her ‘unremitting contact with the smell and feel of primal substances (amniotic fluid, lochia, colostrum, urine, faeces, breast milk, mucus, posset etc) implicitly absorbed from the mother’s own archaic carer’s childbearing body’.

I use the word ‘memory’ in a particular sense here. Stone (2012, p. 6) distinguishes between ‘explicit’ remembering which is ‘mediated by language and the categories it encodes, and by the publicly shared map of time and space’, and the ‘bodily, habitual, emotional remembering’ that is archaic and preverbal; what Klein (1959/1975, p. 180) called ‘having memories in feelings’. The latter are not experienced as memories but rather ‘consist of schemata for behaviour or for certain patterns of emotional reaction to given kinds of situation’, and are re-membered and reproduced in the corporeality of relationships with others; through the raw and visceral’ experience of a mother’s constant bodily closeness to her baby as Raphael-Leff has suggested, for example (Stone 2012, p. 5). Stone writes that in this re-membering as a mother in relation to her baby, a woman is ‘drawn back to re-inhabit her earlier position as one who reciprocates, mimics, and reproduces her own mother’s behaviour’ (2012, p. 11, italics in original).

The narratives of the circumstantially childless women in my study lead me to suggest that women who have not yet had children are also positioned as a generational pivot in the way Hollway discusses, although perhaps not as vividly, since they are not immersed in the living viscerality in the same way as Stone’s words invoke. Perhaps many women’s maternal fantasies are largely unconscious for many years — both in the sense of being in part constituted of archaic memories and therefore unavailable, and also of being not accessed in conscious thought, in a way I would describe as dormant — until they are ignited in some way. This ignition has usually been understood to be linked to a re-immersion in the viscerality of maternal body relations as a new mother that Stone and Raphael-Leff describe. Stone argues that in the usual course of their adult lives today — as opposed to a former time or in other places where people might have more ongoing contact with babies as part of their everyday lives in families and communities — the relational and corporeal aspect of these early ‘memories’ are usually dormant because our adult contexts seldom evoke it. The research materials in my study suggest that, while the intensity of these ‘memories’ evoked by the embodied relationship that a mother has to her newborn baby may not have been experienced, other experiences in her life such as sisters or close friends having a baby, their own contact with babies and small children, or a brief experience of a pregnancy that is miscarried or terminated may quicken the archaic memories of her early time with her own mother.

In Deborah’s case, it seems a number of factors — her growing awareness that she had veered unthinkingly off the ‘imagined pathway’ she thought her life would take, her peers starting to make families, and her fertility running out — may have been the trigger that moved her childhood maternal fantasies to a ‘place’ where she would explore them consciously. As a child, she ‘never doubted’ that she would have children. She had envisioned a trajectory whereby she left New Zealand to spend some time working in the UK before returning to meet a partner and have a family. Once in the UK however, she found herself enjoying life enormously. The time away extended beyond her 20s, and the usual two year ‘OE’ (overseas experience) that many young New Zealanders take. In her early 30s she was still living and working in London, but her circle of friends was beginning to break up, marry and have children. She described a time of feeling ‘very unsettled and uncomfortable’, decided to return to NZ, and within two years had met and married her husband. It was at this point — now in her mid 30s — that they had made the ‘very difficult’ decision not to have a child because of the risk of disability.

Trans-subjectivity and the ‘matrixial borderspace’

A second theoretical resource that adds a further dimension to Deborah’s ‘sense of myself as a mother’ and the maternal fantasies it entails is Ettinger’s (2006a, 2006b) concept of the ‘matrixial borderspace’; a ‘space’ Ettinger sees as a ‘prototype’ of Winnicott’s ‘transitional space’ (Lichtenberg-Ettinger 1997, p. 392).

Winnicott and other post-Kleinian object relations theorists built on Klein’s theorization of the series of relationships between the baby and the maternal body as part-objects to push the origins and development of subjectivity back to the archaic field of intersubjective relations and events between infants and their carers (usually understood to be the mother) in the early months of a baby’s life. These accounts generally begin with the rupture of the baby’s separation from its mother at the trauma of birth (Pollock 2009), followed by further individualization via weaning, motor development, language acquisition and so forth. In Ettinger’s terms this is the binary ‘phallic mode’ that underpins dominant psychoanalytic (and popular) assumptions of human subjectivity and relationships. In these accounts the maternal body is primarily interpreted as a background or container that an individual is required to leave in order to become a self (Stone 2012).

In her concept of the ‘matrixial borderspace’ Ettinger reframes the maternal by extending this psychoanalytic paradigm back further still, to a prenatal, prematernal ‘trans-subjectivity’ (Pollock 2006, p. 14) between a mother and her child; that is, to gestation and the psychic and bodily experience of pregnancy (Boyne, 2004). Here subjectivity is conceived of as a stratum of primordial encounter between subjectivities-to-be that are not known to one another, and yet ‘recognize’ each other and develop in relation to one another. Hollway (2012a, p. 550) notes that the metaphor of a stratum that Ettinger engages ‘provides the insight that the trans-subjective is overlaid but also that it remains the bedrock of subjectivity’. Ettinger refers to this original stratum as the ‘matrix’, and she looks to the encounter of pregnancy as the pattern for an alternative form of relationship which precedes and runs alongside (rather than instead of) the phallic mode in individuals’ day-to-day lives; a prototype for other human situations and processes in which the non-I is not an intruder but a partner in difference, a dimension of subjectivity that is ‘several, joint-in-separateness, distant-in-proximity’ (Pollock 2006, p. 28).

Ettinger’s proposal that aspects of the uterine experience continue to weave together with those that come about through a subject’s experience after birth is grounded in feminist critiques of the notion that separation from the mother is necessary in order for a subject to become an independent and autonomous subject. The understanding of an ongoing porosity between individuals on which this trans-subjective ‘memory’ might be based is in marked contrast to the binaries implicit in the phallic mode. Her trans-subjective concepts of matrixial border linking and border spacing move beyond notions of ‘identification’ and ‘disidentification’ to one of a shared matrixial space in which the subject-becoming-a-mother and the subject-becoming-a-child modify one another.

As I have noted, Ettinger proposes that this originary trans-subjective matrixial stratum does not disappear at birth, but rather is accompanied by other strata which are to do with partial separation and increasing individuation, alongside which it co-exists after birth; in Ettinger’s words ‘it accompanies the phallic subjectivity all along its voyage’ (2004, p. 78), and matrixial processes continue throughout life. She uses the notion of a ‘sense’ herself when she writes that ‘a retroactive matrixial ‘sense’ may reach its archaic roots in relations with the other’ (2009, p. 393). This creates the possibility of identifying events and contexts in a woman’s life that contain traces of a maternal subjectivity where ‘the split [with the mother] collapses, yet difference is maintained’ (Ettinger 2006a, pp. 196–197). From a matrixial perspective then, traces of the archaic trans-subjective relationship a woman has with her mother exist alongside those of the later intersubjective relationship that develops as the daughter grows, and permeate the fantasy dimension of the relational positions that each woman develops.

In this light, the ‘ignition’ of their fantasies of maternity that I have suggested these participants experienced might be understood to have brought the trans-subjective stratum closer to the ‘surface’ of their conscious awareness (Hollway 2012b). A number of other participants used the term ‘a sense of’ to capture their experience of the ‘isness’ of their fantasy child or maternity, as Deborah did. It may be that her ‘sense of myself as a mother’ is grounded not simply in an imagining of what it might be to be a mother, but rather in an archaic embodied experience of her porous trans-subjective relationship with her mother in the late uterine period, as well as in the later relationship with her mother, her sister, and other people and resources in her social world in the period after her birth and throughout her childhood.

Untangling ‘mother’, ‘nurturing’, and ‘maternal’

Those social resources include the complex ways in which ‘mother’, ‘nurture’ and ‘maternal’ have been tangled and often conflated in the discourses around women and motherhood in which Deborah would have been immersed as she grew up. As we talked in the interview, she drew on these resources as she tried to make her own meaning of the experience she had struggled to articulate. Her response in the extract I quoted earlier was prompted by a question about why the decision not to have children that she and her husband had made was difficult for her. She went on:

I guess [pause] you know, for example, in my job at the, in my job at the moment as [position], one of the things that I do, um, as part of my job, and I love, is mentoring of, and supervision of [a set of people]. And [pause] in some ways, some of the life skills, personal skills that I apply in those relationships—I mean these are not parental relationships [laughs], it’s entirely inappropriate that they should be—um, and yet there are some skills, some attributes of myself, that I use in those relationships, um, or make available in those relationships, that would also be available and used within a mothering, or a parenting type of relationship. And [pause] I think that, um [pause] and strangely, you know, whether it’s in a relationship with a pet [laughs], or a relationship with a spouse, or a relationship with a sibling, or a relationship with a niece or nephew, […] somehow there are elements of myself, in each of those relationships that [pause] that are, that parts of those things are a sort of a parenting, nurturing sort of part of me. And so it’s a tension, I think, between, you know, if you’re not a mother — you don’t have biological children — then you’re not maternal, or you’re not nurturing, or you’re not something.

Writing of the matrixial Pollock (2009, p. 7) asks, ‘(W)here is maternal subjectivity, and is it confined to a maternal subject? Is it singular?’ Deborah’s words suggest that she had understood and identified the notions of ‘nurturing’ and being ‘maternal’ as ‘elements’ of herself that constituted the sense of herself as a mother. Excluded from identification with the subjectivity of ‘mother’ because she had not had children, Deborah had decoupled ‘nurturing’ from ‘biological mother’ by identifying other relationships in her life in which she embodied ‘nurturing’. In doing so she distinguished between being a mother and being ‘maternal’. Rather than understanding those “those parts of me [that] exist, and are used, and are valued” as something she had not developed because she had not become a mother, she had become engaged in a process of mining her maternal fantasies for the aspects of ‘mother’ that she recognised in herself, and in embodying what she understood to be those maternal dimensions of her subjectivity, in a life without biological children. This is not so much a process of doing other things instead of being a mother, but rather of what she described as being ‘maternal’ finding its embodied expression in her life in other ways. In this way she was able to keep something of her ‘sense of myself as a mother’ — what I am describing as her fantasy maternal subjectivity — intact. In the email she wrote telling me about her ‘down the back’ children Deborah confirms this. She wrote, ‘I feel I am moving on slowly towards a place where I am aware that I am ‘mother’ in ways that are possible and matter (are of value) to me. And I’m starting to feel OK about that’.

It is not surprising that she might understand this dimension of her subjectivity as ‘the sense of myself as a mother’. For many women nurturing is frequently expressed through the birth and care of a baby and child, both because that experience is rooted in their archaic trans-subjective relationship with their own mother, and because it is the route to the expression of nurturing that is still most validated socially, in spite of feminist work to challenge the assumption that women are inherently nurturing. Matrixial theory proposes that women are particularly likely to re-experience and express the matrixial dimension of their subjectivity in having a child. However, as Pollock suggests, when a woman has not ‘reactivated’ the matrixial dimension of her subjectivity in pregnancy and the birth of a biological child, it is potentially expressed in other ways. Ettinger (e.g. 2006a; 2004) has written extensively of its expression in art-making and clinical psychoanalytical practice. For some circumstantially childless women, nurturing might be best understood as another dimension of their subjectivity — running alongside the phallic dimensions that develop after birth — that they draw on in a particularly deliberate way in their relationships with others in their life, as Deborah described.

Deborah’s significant shift of pronoun in her last sentence from ‘I’ to ‘you’ suggests that there are clearly socially constructed limits to this process however. In terms of other people’s acknowledgement of her capacity to nurture, Deborah perceived these alternative ways of being ‘maternal’ to be deemed inadequate. This judgment caused her considerable distress; in another part of her interview she commented,

(t)hose are the times, I think, when I feel the pain the most, when I get a sense that a judgment is being passed. When I feel that people have no reason to be making the judgment other than an observation that we don’t have children.

For Deborah this judgment implied that being ‘maternal’ in other ways is popularly understood to be both less adequate than being so as a biological mother — it’s ‘not the same’, as several participants noted — and also unacceptable in some way. Deborah’s two little points of laughter and use of the word ‘strangely’ in the excerpt above were perhaps intended to signal to me that she realized that on a common sense level what she was saying was absurd. They were designed to soften her articulation of the notion that her relationship with others or with pets was in some way enacting her sense of herself as a mother. They were necessary as a narrative device in this way because she perceived that such a notion is understood by others to be laughable at best; at worst objectionable, and unspeakable. She understood it to be ‘entirely inappropriate’ that the relationships were parental, because values such as nurturing and concern are closely associated with the maternal, and expression of the maternal is limited to the care of children by their (preferably biological) mothers in dominant discourse. Although it was her experience that these relationships express her ‘sense of myself as a mother’ she did not feel comfortable in saying this, and disavowed her own maternality in these relationships.

‘And yet’; and yet, even though it did not make sense and was not socially speakable, this was Deborah’s experience and she wanted to be as honest and full in her response as she could be. Even as she disavowed it, she also (tentatively) claimed it. Perhaps it was also important to her to honour her experience of that ‘maternal’ sense of herself, just as it was later when she wrote of my use her fantasy children’s names giving them ‘a validity or reality I’m pleased to gift them (and myself)’. The ‘and yet’ was an expression of the tension she experienced in the contrast between her sense of herself as being ‘maternal’ and the lack of social acknowledgement of this because it had not been expressed in biological motherhood. This ‘tension’ was not fully articulated in her sentence; one side of it is the way in which her being maternal is not recognised socially, but the other side — the way in which she understands herself in some ways to be a mother — was unspoken; perhaps an ‘unthought known’ (Bollas 1987, p. 3). In this aspect too, Deborah’s experience exceeded the capacity for expression in discourses of maternity.

Conclusion: Cutting across the available discourses of maternity

Considering Deborah as already a mother in some way appears nonsensical, if ‘mother’ is exclusively indexed to an embodied relationship between a woman and a child, and a woman therefore either ‘is’ or ‘is-not’ a (biological or social) mother. I have proposed that her words are more intelligible when ‘fantasy mother’ is considered as a further dimension of maternal subjectivity that can be understood to begin before a girl’s birth, and to continue throughout her lifetime.

In proposing this I am not suggesting that all women should have children, nor that all want to have children, and that this desire is somehow thwarted or suppressed. In focusing on this aspect of women’s maternal subjectivity I have left aside the question of how women make the reproductive choices they do. The ambivalent nature of these fantasies, competing desires, and a variety of social factors all play a part in an increasing number of women choosing not to have children.

Nor am I suggesting that all women ‘are’ mothers if they imagine themselves to be. Rather I would like to acknowledge that women’s conscious and unconscious fantasies of maternity and a child or children have a powerful significance — a further reality — that is a precondition of the social reality of a maternal identity, that shapes their experiences of maternity if they go on to have a child, and that has the potential to create material and emotional effects in their lives even without their becoming mothers in genetic, gestational or care-giving terms.

Hollway has suggested that caring aspects of maternal subjectivities might become known and available to individuals regardless of whether or not they have children. She writes (2006, p. 80) that ‘equivalent capacities to care could be precipitated by other caring relationships and responsibilities, not necessarily by one’s own baby’. Deborah’s narrative — and those of other women in this study — supports this idea. A matrixial reading makes it possible to conceptualise these aspects of maternal subjectivity as being available to her precisely through the same intersubjective and trans-subjective processes that came into play in developing her fantasies of maternity. That is, that she has drawn on her psychosocial knowing/experiencing of maternal subjectivity as a child and in relationship with her mother in the womb, in embodying aspects of her fantasies of maternity in other nurturing relationships in her adult life.

Stone (2011 p. 167) writes that mothers’ experiences and development are organized in a ‘particular relational or psychical position — a distinctive subject-position’. My suggestion that ‘fantasy mother’ is understood as a powerful dimension of some childless women’s subjectivity entails extending theorization of this ‘position’. It hinges on the understanding that the development of the fantasy element of maternal subjectivity can be conceived as a process that does not begin at birth when a woman ‘becomes’ a mother, nor even in pregnancy in relation to an embodied foetus, but rather begins in her relationship with her mother in the womb, and continues throughout her life. Indeed, given the way participants talked about their fantasies of becoming a grandmother later in their life, and the impact of their mother’s fantasies on the development of their own — even before their conception — I suggest that a woman’s fantasies of maternity can be understood to be unlimited temporally (since they are linked to her mother’s own fantasies, and logically, to her mother’s and so forth), socially, and relationally.

Why did Deborah find it so hard to ‘relinquish’ these fantasies, when the physical capacity to embody them had passed? Why do these fantasies persist rather than falling away as these women age and cease to be capable of conception? One explanation might be that despite 40 years of feminist work to dismantle the ideologies that link womanhood with motherhood, femininity, and nurturing, for many women, these understandings of gendered identity persist. Girls still grow up immersed in a culture that teaches them that becoming a mother is a desirable goal for women, and they subsequently value those ‘maternal’ aspects of themselves and want to embody them. Many of these participants’ narratives support this argument; Deborah’s ‘tension’ that ‘if you’re not a mother — you don’t have biological children — then you’re not maternal, or you’re not nurturing, or you’re not something’ for example, suggests that her perception is that, as far as other people are concerned, she is not understood to be ‘maternal’ or ‘nurturing’ because she has not had a child.

However, for me this is not a wholly satisfactory explanation. On the face of it, the idea that nurturing is intrinsic to her would suggest that Deborah was conflating ‘woman’ and ‘nurturing’ in an essentialist way, but I do not think this is an adequate analysis. I argue rather that part of her difficulty in talking about these feelings is that she was very anxious not to adopt an essentialist approach, but she found it hard to talk about her responses without appearing to do so. It was clear that Deborah herself did not understand the issue in that way, and as I worked with the research materials over time it struck me how many of the women I interviewed were like her; highly articulate, educated, self-reflexive people who had grown up with mothers who identified as feminists, and did so themselves. These participants had a complex understanding of and engagement with the themes and issues that have dominated feminist debate throughout their lifetimes. In spite of this, they did not understand, refer to or explain their experience in these terms. The ways in which many of these participants demonstrated a sophisticated level of acuity around the feminist politics of gender discourse adds weight to my analysis that there is something about their experience that these discourses and feminist discourses of biological determinism do not account for; something that the research materials suggest was as an active part of their lives, cutting across the available discourses of maternity, and somehow eluding their capacity to speak of it.

It appears that the creative and satisfying ways in which Deborah and many of the circumstantially childless participants in this study have enacted their maternal fantasies evade available discourses of the maternal. This interpretation is consistent with their perception that their expression of what they experience to be the maternal aspects of their subjectivity in forms other than being ‘mother’ are always understood by others in their social worlds to be in some way inferior to the gold standard of ongoing biological mothering; at best as a ‘way of coping with childlessness’ (Letherby 2012, p. 14), at worst illegitimate and laughable.