К оглавлению
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 

I became increasingly interested in the function that narratives were

perceived to serve as my research interests unfolded. A key interest

centred on how women strategically constructed and voiced narratives,

drawing on the cultural and social knowledges that constitute ‘meta

nar ratives’ (Somers, 1994). This was par ticular ly app arent in rel ation to

women presenting a particular version of their selves for example as a

‘good’ or ‘coping’ new mother, and the ways in which this was culturally

patterned. Traditionally, narrative has been a concern more readily associated

with philosophy, literary and linguistic traditions. However, a turn

to narrative is now clearly discernible within the social sciences

(Pl umme r, 1995 ). This interest, in par t, can be attri buted to the

increased interest within the social sciences in subjectivity and the

meanings attributed by individuals to their actions. The study of narrative

is one attempt at coming to terms with how social identity and, in turn,

social action, are constituted and guided. This linking of identity and

action, the ontological condition of social life, has challenged earlier

thinking around narrative as merely textual, non-theoretical representation.

It has also contributed to the considerable debates on how selves are

constituted and maintained in late modernity. The focus then has shifted

to take account of the way in which human life is storied. As Lieblich et al.

( 1998 ) have observed, ‘we know or dis cover ourse lves, and reve al ourselv

es to othe rs, by the stories we tell’ (1998:7) . Narrati ve the n can help

us to understand social life and social practices. This is achieved by

bringing together dimensions of narrative, for example social action,

historicity, temporality and relationality, which in the past have been

often overlooked and which a focus on transition to motherhood enables

us to explore. Through the construction and reconstruction of narrative

accou nts, using devi ces such as emplo tment (Corrad i, 1991; Somers,

1994 ), events are bro ught toget her as episode s an d a life is given uni ty

and coherence. In practice, of course, unity and coherence may give way

to (usually) temporary ‘bafflement’ as people struggle with ‘chaos’ in

their lives (Fran k, 1995). But the impo rtant point is that as indiv iduals

we are ‘not only the actor, but also the au thor’ (MacIntyre , 1981:198) as

we travel through and make sense of our lives.

Past philosophical debates around narrative have been concerned with

how human life is storied. Contributing to these debates, Alasdair

MacIntyre comments that ‘human life is composed of discrete actions

which lead nowhere, which have no order; the storyteller imposes on

human events retrospectively an order which they did not have while

they were live d’ (Mac Intyre, 1981:199). Life then is not lived as a neat,

chronologically ordered series of events. Rather, as actors we are able,

through narrative construction and reconstruction, actively to impose

some order, some intelligibility on events, retrospectively. Devices such

as plot enable the individual to weave accounts into continuous and

intellig ible stori es (Ricoe ur, cited in Vald es, 1991 ). Clearly, then, being

able to produce intelligible and culturally recognisable and acceptable

accounts of events is an important feature of the storied human life. We

use these accounts both to make sense of our own experiences and to

present ourselves in particular and strategic ways to others. Yet some life

events and life transitions may challenge our ability to do so: becoming a

mother and early experiences of motherhood are such events.

The more recent claims that narrativity is more than a method of

representation lead us to consider further the links between individuals

and their actions. To talk in terms of the construction and reconstruction

of narrative accounts implies a distancing of individuals from their actions

and does not sufficiently address the notions of an individual’s intention

and accountability. Yet intention refers to the individual as a competent

social actor with ideas and aspirations, who possesses the ability to ‘move

purposi vely in the worl d’ (Step henson, 1999 :114). If we are to understand

an individual’s behaviour and the narratives he/she construct then

these need to be considered within the context of longer- and shorter-term

intentions and ordered both causally and temporally. Accountability for

action is also linked to intention. For example, MacIntyre has argued that

we can ask an individual ‘to give an intelligible narrative account enabling

us to understand how he [sic] could at different times and different places

be one and the same person and yet be so differently characterised’

 (19 81:202). We can unders tand narrative history and ‘narr ative trut h’

(Lie blich et al ., 1998 :8), then, through an unders tanding of intention and

cont ext. Yet huma n life is both fragile and unpre dictabl e. The pote ntial

for the discontinuity of a narrative account, both in terms of making sense

of experiences to oneself, and producing accounts for others, is ever

present. For example, when experiences do not match predicted expectations

or marry with intentions, an individual’s ability to produce and

sus tain a coheren t, culturally reco gnisable and socia lly acc eptable narra -

tive may be challenged. It is through the device of narrative reconstruction

– the revisiting and reordering of past experiences – that long-term

‘bafflement’, or narrative lapse can usually be avoided and continuity be

maintained/regained. Yet self-reflexivity is key here, for as Frank has

pointed out, ‘those who are truly living the chaos cannot tell in words.

To turn the chaos into a verbal story is to have some reflexive grasp’

(19 95:98). We will return to explore the contou rs of reflexiv ity late r in

this chapt er and in chapt er 7.

In seeki ng to unde rstand how individu als make se nse of disrupt ive or

transitional events in their lives, the role of narrative retelling has been

recognised as an important enterprise in which ‘individuals actively shape

and accou nt for biographi cal disrupt ion’ (Reissman , 1990 :1196). Yet we

are not isolated actors but make sense of events and construct narrative

accounts in relation to past experiences and future expectations, and

importantly in relation to other social actors. Narratives are, then, interpersonally

and interactionally constructed. As individuals we are guided

in our storytelling by reference to multi-faceted cultural scripts, which

pro vide the cont ours of par ticular way s of knowin g (see chapt er 2 ).

Actors, then, are limited in their narrative construction and reconstruction

and make sense of their experiences from, and within, particular

locations in the social world. Somers points to the important interplay

between meta-narratives, which shape our actions, and the stories that we

produce individually. She points out that ‘stories guide action’ and the

social and cultural contexts in which we live will offer us ‘an ultimately

limited repertoire of available social, public and cultural narratives’

(19 94:614). The implicat ions of this are that par ticular ways of knowin g

may be privileged over others. These may be ‘contested politically and

will depend in large part on the distribution of power’ (ibid.:629). In the

case of reproduction and motherhood the shifts that have occurred in the

management and place of birth and the role of biomedicine in many

Western societies show how particular stories can come to dominate

and shape practices and expectations. Where biological ‘facts’ are also a

facet of the stories that shape practices, as in the example of childbearing,

the power of particular ways of knowing can appear incontrovertible.

Narratives, then, exist at different levels. They are individual stories

emanating from personal experience and reinterpreted and reconstructed

over time and in different contexts. They are also collective stories of

discernible groups in wider society, which provide the contours of the

available and, importantly, acceptable cultural scripts. It is important

then to note the cultural dimensions of narratives. The cultural

embedding of narrative accounts has been emphasised in the work of

such the orists as Ma cIntyre ( 1981 ), Ricoe ur ( 1981 ) and Bar thes ( 1977 ),

and is expl ored further in the nex t chapter. In his work on illness

narra tives, Kle inman ( 1988 ) too has emphas ised the impo rtance of

cultural belief systems that are reflected in narrative accounts and the

need to be sensitive to these.

It is argued, then, that as human beings we are storytelling animals. We

act with intention and purpose and make sense of past experiences,

present and future hopes and expectations, in relation to particular historical,

cultural and social contexts. And it is this ability that ‘provides us

with an identity – a sense of existing through time and of acting purposively

in the worl d’ (Step henson, 1999 :114). Takin g a narrative appro ach

in the study of subjective experience, then, enables the researcher to

access and explore individual identities: the ways in which social actors

actively produce narrative accounts and present their selves to others.

This approach, however, raises further questions about, and throws light

on, the ways in which identities and ‘the self ’ are constituted and maintained,

situated and narrated. Theoretical positions in relation to debates

on the self have ranged across contested terrain, from those grounded in

essentialist arguments of a pre-existent, core self, through liberal humanist

arguments of a disembodied and disembedded self, to postmodernist

positions and the ‘death of the subj ect’ (But ler, 1990 , 1993 ; Griffit hs,

1995 ; Evans, 2003 ). Femin ist cont ributions have no ted the te nuous,

changing and fragmented dimensions of the self and emphasised the

importan ce of situati ng the se lf (Benha bib, 1992; St anley, 1993). Life

events, such as the onset of chronic illness or becoming amother, provide

opportunities to explore the ways in which selves are constituted and

maintained. These enable us to see how individuals make sense of periods

of biographical disruption and personal transition. A focus on the constituents

of a life, narrative, identity and the self, reveals the complex

interplay, fluidity and ultimately precarious nature of these components.


Debates concerning the self and social interactions have provided an

enduring area of interest in the social sciences. The relationship between

language and the self was first explo red in the cl assic work of Me ad

( 1934 ). Me ad arg ued that the mind and self were not pre-e xistent, but

eme rged through language and in rel ation to interacti ons with othe rs.

Goffm an, too, made a maj or cont ribution to debates on the self. His

deta iled an alysis of socia l int eraction s enabl ed him to make claims

abou t the ways in wh ich the self is so cially cons tructed (Goffma n,

1969 ). In his work, Goffma n revealed the ways in wh ich we are able to

pres ent a particul ar self in particula r settings using ‘impre ssion

man agement’ to manag e performan ces. He drew atte ntion to the

‘all- too-huma n’ task of stagi ng a p erformance and conc luded that the

self ‘is a product of a scen e that comes off, and is not a cau se of it’

(Goffma n, 1969: 245). A m ore recent readi ng of this work draw s attention

to a dua lity of self that is impl icit in Goffma n’s distinct ion betw een an

‘all- too-huma n self ’ and a ‘socialis ed self ’ (Bra naman, 1997). Brana man

arg ues that ‘the al l-too-huma n se lf is the huma n being as a psychob iologic

al organis m with impulse s, moods and vari able energies, but al so is

the self which engag es in the all-too-hum an task of stagi ng a performa nce’

(19 97:x1vi ii). Goff man, then, is conce rned with performa nce and m anageme

nt in terms of bodily actio n rather than engag ing with the pote ntial

mes siness of ‘the sensu al, visceral ’ embo died social actor (Jacks on and

Sco tt, 2001 :11).

The notion of ‘true’ selves and ‘false ’ selves is found in Hochsc hild’s

work and moves beyo nd the performa tive self found in Goffman’ s work .

Hoc hschild focus es on our abili ty to carry out ‘emoti on work ’ through

‘surfa ce display s of the right fee ling’ an d ‘deep’ work (1983). More recent

cont ributions to deb ates on the self have importan tly noted the gende red

and embodi ed nature of identity, that ‘in our society we are never simp ly

selv es, but gen dered selves’ (Cool e, 1995 :123; Bro ok, 1999; Evans ,

2003 ). The conn ecti ons betw een agency and wome n were of course

mad e m uch earlier in the work s of Simon e de Beauvoir (Ev ans, 2003) .

All this has implications for our reading of Goffman’s work, for as Jackson

and Scott point out, ‘while embodied actors are ever present on

Goffman’s social stage, he was always concerned more with bodily action

and performa nce than with the sensual, visc eral body ’ (2001:11) . Indeed ,

the lack of attentio n to ‘emotion al needs’ more gen erally in socio logies of

the se lf has also recently been noted (Elliott , 2001 :44). Clearl y, ta king

account of the embodied, gendered dimensions of the self is important.

This is especially so when focusing on women’s experiences in relation to

their bodies and perceived biological and social roles, together with convent

ional expectati ons of ‘bein g there for othe rs’ (Bail ey, 2001 ; Ad kins,

2002 ).Yet not all theo rists would accept such a readi ng of gen der. The

positions, according to Evans, divide according to ‘those who accept the

givens of bio logy’ and ‘tho se (a nd most infl uentially Judith Butler ) who

argue that all gendered behaviou r is a matter of the int ernalis ation of

socia l expectati ons’ (20 03:57). Inde ed, for Butler it is the performat ive

self that is emphas ised, there being ‘no doers behind the deed’, wh ere

‘the produc tion of self and g ender (are) a discursi ve effec t’ (Ell iott,

2001 :117). But social expec tations and experi ence s arou nd reprod uction

and mothe rhood sit precarious ly withi n both camps . For me it is

necess ary in any analy sis of aspe cts of the social worl d to take acc ount

of the ways in wh ich material ci rcumstan ces and embodim ent shape

expec tations and experi ences. The notion of a clearly gende red se lf,

whic h draw s on the diffe rences in bodil y matte rs between m en an d

women, is, then, an importan t considera tion here (Almond, 1988;

Jackson and Sco tt, 2001 ; Evans, 2003). Almond ‘argu es that the phy sical

facts of men struation, conc eption, preg nanc y, childbirth and menopau se

generat e a series of moral problem s rel ated to ident ity and self conce pt

for wome n which are different to those experi ence d by men who do not

go through such chang es’ (cited in Gri ffiths, 1995 :78). So, as we actively

make sense of experi ences and produce and sustain (a nd re constru ct)

narra tive accounts we do so from embodi ed, gen dered an d une qual

positions within the socia l worl d. These gen dered positions are also

closely bound up with (apparen t) choice s in a life. Commenting on the

actualit y of choices in a life and the way s in which these are linke d to

ident ity, Benhab ib pose s the question, ‘how does this finite embodie d

creature cons titute into a cohe rent na rrative those episodes of choice and

limitati on, agency and suffering, initiative and dependen ce?’ (1992 :161).

The focus on transition to motherhood taken in this book provides an

opportunity to explore the complexities inherent in selves, agency and the

production and maintenance of coherent narratives.

In recent years feminist debates and other theoretically grounded

discussions have also emphasised the importance of contextualising

experi ences and situa ting the se lf (Benha bib, 1992 ; Stan ley, 1993 ;

Griffit hs, 1995 ; Brook , 1999 ). For exampl e, our experi ences can vary

significantly from expectations that have been shaped by available cultural

scripts and particular ways of knowing. In relation to mothering

and motherhood our ideas will be culturally and socially shaped and,

importantly, morally grounded. This is a powerful context in which the

biological, although overlaid with social and culturalmeanings, continues

to shape expectations, modes of authoritative knowledge and professional

practices. For women who need time to develop their mothering skills

and who do not necessarily feel able ‘instinctively’ tomother, even though

they may feel they should instinctively know, this can raise concerns. As

will be seen in later chapters, it may lead to self-silencing and social

isol ation as women try to mak e sense of their early experi ences of

mot hering. The g endered self then is not simply consti tuted through

‘mode rn discour ses and discipli nary techniqu es’ (Cool e, 1995 ). Rath er,

it is constitut ed in the context of particul ar histo rical, social, moral,

polit ical an d mater ial circum stances and embodie d activ ities. Caring for

and havi ng responsi bility for our youn g, depen dent childr en are also

highl y gen dered practices , wh ich are also linke d to the gen dering of publ ic

and pr ivate (Ribbe ns McCart hy and E dwards, 2002 ). All the se –

gen dered selves, pra ctices an d spaces – cl early have implic ations for the

ways in wh ich agency is operatio nalised and mothe rhood experi enced

and narra ted. Debates on the ‘free ing of agency from structure ’ have

large ly failed to take account of the embedd ed and embodi ed dimensions

of agency and the ways in wh ich this is played out in differe nt m aterial

circu mstan ces and differe ntially experi enced (Adkins, 2002 :3).

The moral dime nsions of m otherho od, childrearin g an d assoc iated

pra ctices are hard to escape in an y anal ysis of recent literat ure and

res earch (Gl enn et al ., 1994 ; Arendel l, 2000 ; Chase an d Rogers, 2001 ;

Rib bens-McC arthy at al ., 2000 , 2003 ). The ‘mora l minefield ’ in wh ich

mot herhood is experi ence d clearly shapes the types of acc ounts of

mot herhood we feel able to cons truct and h ow we presen t ourse lves

(Mu rphy, 1999 :187). Indeed , in contra st to Goffman’ s cl aims that our

activity ‘is largely concerned with moral matters, but as performers we do

not have a moral conce rn with them’ (1 969:243), it will be argued

throughout this book that it is the moral context in which women give

birth and become mothers in Western societies that has crucial implications

for both a sense of self and presentation of self as amother. Although

it is noteworthy that debate continues as to whether or not there is ‘a

moral sensibility’ at the core of Goffman’s work, it is clear that his ‘theory

of self says surprisingly little about the emotional or psychosexual

dyna mics of persona l life and soci al rel ationships ’ (Elliott , 2001 :35–6).

Clearly, becoming a mother provides new opportunities for

presentation of self. However, the biological fact of giving birth and the

expectations that surround mothering can render early experiences as

ultimately precarious, overshadowed by concerns that ‘performances’

will be discredited. Because becoming a mother involves a biological

act, even though the context in which mothering is then experienced is

socially located and culturally embedded, being a mother is always more

than ‘pla ying a part’ (Goffman, 1969 :28). It involve s an embodie d

experience, culturally located feelings of responsibility and being able to

meet a baby’s needs. At the same time, it has profound implications for a

woman’s sense of self. As data in subsequent chapters will show, although

wome n might have all the ‘props’ (Goffma n, 1969 ) app arently nece ssary

to be able to give a convincing performance as a mother – a baby, pram

etc. – at times this is not enough. Women claim that they do not feel like

mothers and can express concerns that they are fearful they will be ‘found

out’. Such worries are deeply rooted in perceptions of the moral context

in which mothering occurs. They go beyond concerns with impression

management and highlight the embodied and embedded dimensions of

being a mother. This is not to subscribe to essentialist arguments of there

being a core self, although women coming to motherhood may themselves

have quite essentialist expectations and can experience confusion

when these are found to be misplaced. Rather, my argument is that the

practised, recognisable gendered and embodied self, which makes up our

identity, is challenged by the experiences of first time motherhood: over

time, a new social self as mother has to be learned.

In her recent work ‘Mothering the Self’, Lawler usefully captures

contemporary notions of selfhood which go beyond the static

and unchanging self. She notes that ‘most Euroamericans incorporate

various forms of understanding of the self into an overall schema of

self-understan ding’ ( 2000:57–8). As data p resented in subsequent chapters

will show, a prime concern for new mothers is regaining a sense of a

recognisable and practised (pre-baby) self, which is described in relation

to ‘getting back to normal’. This involves the women, over time, coming

to make sense of the embodied and other changes that accompany

transition to motherhood. It also involves developing and incorporating

an understanding of self-as-mother into their broader schema of

self-understanding. That mothering continues to be regarded by many

as biologically determined only serves to complicate the context in which

women must make sense of their experiences. Not only can it be a moral

minefield, but women are also expected instinctively to know how to

mother. Indeed, their ‘femaleness’ may be questioned if they experience

normal difficulties. Yet any examination of the cultural practices and

ways of knowing about mothering shows the variations that exist:

mothering is not a universally standard experience. Many of the skills

associated with mothering, fathering and parenting are only gradually

acquired. So, whilst the performative nature of self is emphasised in

Goffman’s work, the complexities of putting on and maintaining a

performance clearly need further scrutiny in relation to becoming and

being a mother. For example, different settings can provide different

challenges. The home may be felt to be a safer environment in which to

manage early mothering experiences whilst public settings may be seen

as too risky a stage. Performances are, I suggest, also grounded in an

underlying, recognisable sense of a pre-motherhood self, a self which

encompasses emotions and feelings and perceptions of particular ways

of being in particular circumstances: characteristics which are hard to

escape in relation to mothering.

What all this means is that selves are complex, changing and, at times,

fragile, embodied constructs. Yet a sense of self that can be reflexively

sustained andmaintained, by making sense of experiences is, it is argued,

nece ssary for on tological well- being (Giddens , 1991 ; Holstei n and

Gub rium, 2000 ). This is perhaps esp ecially so in the face of greater

uncertainties thrown up by the transformations that characterise late

modern societies. According to Holstein and Gubrium ‘the self is an

incre asing ly compl ex project of daily living’ (2000 :x). Similarl y, the

importance of being able to ‘make sense’ in daily life has been noted by

Giddens. He observes that ‘how far normal appearances can be carried on

in ways consistent with the individual’s biographical narrative is of vital

impo rtance for fee lings of on tological secu rity’ (199 1:58). Reflexiv ity in

this context concerns self-reflexivity and moves beyond being ‘a defining

chara cteristic of all human acti on’ (Gidde ns, 1994 :89), referring to a

deepe r an d more engaged inn er form of self-prac tice. It beco mes an

active, individual response to modernity: a changed context in which

the move to a post-traditional order apparently necessitates greater selfmonitoring

in order to maintain ontological security. Yet this begs the

question of how far striving for, and sustaining, a sense of ontological

security in this format is a Western- or class-, or ‘race’-bound phenomenon

. Clearl y, the ways in wh ich we experi ence individu al agency are

cont ingent on multipl e factors includin g m aterial an d struct ural circu mstances

and our sense of self. In turn, our experiences and expectations of

‘ontological security’ or well-being will be shaped in relation to these

wider factors. For example, when living and working in Bangladesh (see

chapt er 2) it was app arent to me that the main conce rn for the m ajority of

urban and rural poor was securing enough rice to feed their families. This

was a daily concern for many and could literally mean the difference

between survival and premature death. In such a context the idea that

people were reflexive practitioners, in the way the concept is understood

in the West, maintaining ontological security through reflexivity, is

unsupportable. Yet this is not to say that reflexivity is not a characteristic

of all human action at some intrinsic level, or that certain groups lack the

capacity to be reflexive, but rather to urge caution in the claims we make

in relation to its pract ice (see for examp le Adkin s, 2002 for an exce llent

overview of recent debates on reflexivity). As the example from

Bangladesh shows, although individuals had a clear understanding of the

consequences of the material circumstances in which they lived their

lives – they must generate enough income to buy food in order to survive –

the opportunity and relevance of individual reflection in relation to

ontological well-being would not have been either a major concern, or

relevant in the context in which their lives were lived. Lash has also

pointed to the need to consider the structural conditions of reflexivity

(Lash, 1994 ). Yet in stark cont rast, in the USA and m ore recentl y the

UK, the cultural acceptance and dependence on therapy and counselling

indicates a growing need amongst some groups in more affluent societies

activ ely to work on the ir selves (McL eod, 1997; Birch and Miller, 2000 ).

The recognition of the growth of these practices is not to privilege particular

way s of being in the world. Rath er it is to acknowl edge that in some

(Western) societies, such practices are discernible and have developed

alongsi de, and app arently in res ponse to, the rap id trans formatio ns that

have occurred in the se societ ies (see chapter 3). What on tological se curity

is perceived to be, and the work on the self that might be necessary to

achieve and maintain particular ways of being in the different worlds we

inhabit, clearly differs in different contexts. It is also important to note

here the gendered and embodied dimensions of reflexivity and ontological

secu rity. Refle xivity come s from the experi ence s of real bodie s; it does

not just exist in, and emanate from, the mind. Indeed, Giddens’ conceptualisation

of reflexivity has been criticised for neglecting the ‘fleshy

and sensual aspects of the body in favour of a more cognitively and

reflexiv ely man aged body ’ ( Jackson and Scott, 2001:12). Similarl y,

both Giddens and Beck have been criticised for producing ‘overly cognitive

or rationalistic understanding of the late modern self and human

action’ and failure to take account of issues related to gender identity in

relation to reflexi vity (Ad kins, 2002:36). As will be seen in subseq uent

chapt ers, it is the ‘fleshy body’ that plays a key part in women’s reflexi vity

and their constru ctions of soci al se lves in relation to their experience s of

their bodies and becoming mothers.

Reflexivity as a component of agency is, then, experienced and practised

in different ways in different contexts. In relation to childbearing

and motherhood, women’s experiences are diverse and fragmented:

mediated by socio-cultural factors such as class and race and whether

partnere d or alone (Hill Collins, 1994; Du ncan and Edw ards, 1999).

And the same observations can be made in relation to experiences of

choice , refl exivity an d agency (Lazar us, 1997 ). For some reflexiv ity may

be an important device in helping to make sense of experiences, especially

those that do not resonate with expectations. For others, lives may be

lived in cont exts where ind ividua l, active reflexivit y is no t a major component

of daily life. Clearly, all sorts of factors mediate the ways in which

we operate in the social world. For example, the ways in which women

engage with, or reject, scientific or medical discourses, that is, exercise an

aspect of agency, has been explored in the work of anthropologist Emily

Martin (1990). In her work, Martin considers the ways in which class and

race impact on women’s use of medical discourses to explain menstruation.

She found that middle-class American women’s explanations of

menstruation ‘incline toward the medical view’, whilst, in stark contrast,

‘all other working-class women interviewed, black and white – share an

absolute reluctance to give the medical view of menstruation’ (1990:78).

Martin considers her findings within the context of differing ‘material

forces in society’. She argues that whilst agreeing ‘that science should not

be privileged as a description of ‘‘reality’’ . . . this does not mean that the

discourse of science (indeed any discourse) may not in fact be socially

privileged by its relation to structures of power’ (1990:79). Martin’s

work, then, emphasises the different ways in which agency is operationalised.

It illuminates the differences that might be expected in the narratives

produced by women whose lives are shaped by different material and

structural circumstances. Where reflexivity is a feature of everyday life,

accounting for and making sense of periods of biographical disruption

can become an important device, as will be shown in subsequent