ARTICLE
249
Baby entertainer, bumbling assistant and line manager: discourses of fatherhood in parentcraft texts
JANE SUNDERLAND
LANCASTER UNIVERSITY
Discourse & Society
Copyright ©2000
SAGE Publications
(London,
Thousand Oaks, CA
and New Delhi)
Vol 11(2): 249-274
[0957-9265
(200004) 11:2;
249-274;011450]
abstract. In this paper I report on an investigation of discoursal asymmetry in parentcraft texts, in terms of the ways in which the father is represented and backgrounded. In particular, I suggest that it is possible to see one dominant, overarching discourse: 'Part-time father/mother as main parenť. This dominant discourse can be seen as being 'shored up' (as well as, to an extent, challenged) by other, usually complementary, discourses: 'father as baby entertainer', 'father as mother's bumbling assistant', 'father as line manager', 'mother as manager of the father's role in childcare', and 'mother as wife/partner'. These discourses are characterized by recurring and nonrecurring linguistic presences - and, importantly, absences (Van Leeuwen, 1995, 1996). Looking in particular at the following linguistic items from three different semantic fields - mother/father/wife/husband/partner; play/fun/help; and share -1 illustrate how different discourses, with their salient linguistic presences and absences, can organize a text in supporting and potentially destabilizing ways.
key words: discourses, fatherhood, gender, textual organization
Introduction
The textual construction of fatherhood, one dimension of masculinity, has hitherto been underexplored. In this article I aim to illustrate several fatherhood discourses, as well as discourses of motherhood, which are evident in parentcraft texts. I suggest that these discourses can be seen as a hierarchy, in which supporting 'specific' discourses 'shore up' a dominant one. Most of the specific discourses are 'companion' ones, but there are also 'conflicting' specific discourses, which can create tension and potential instability in the whole discoursal structure.
Gendered discourses in parentcraft texts may be shaped by parenting practices, but they may also contribute to the shaping of parenthood - both fatherhood and
250      Discourse & Society 11(2)
motherhood. This is what Dorothy Smith refers to as 'textually-mediated social organisation' (Smith, 1990). And although numerous different readings of a given text are possible, the potential of a text to 'shape' needs to be recognized. As Smith notes:
The text's capacity to transcend the essentially transitory character of social processes and to remain uniform across separate and diverse local settings is key to the distinctive social organisation and relations they make possible. (Smith, 1990: 168)
This study of gender in relation to language, as it moves away from the investigation of 'gender differences' towards the study of the construction of a range of masculinities and femininities through a range of gendered discourses in a range of topically-related texts, is thus in accordance with current language as well as gender scholarship.
If gender is constructed, and if gendered practices are reproduced inside written texts as well as outside them, I suggest that both those written texts which are related to the act of sex in some way (but which move beyond the merely biological), and those related to heterosexual relations in a wider sense, provide particularly valuable data for the study of gender. Since parentcraft literature arguably falls into both camps, it is an important epistemological site for feminist linguists.1 And given the universality of childbirth, and the universal necessity for childcare (in some form, by someone, for almost all children), parentcraft texts may have a particular relationship with the actual practices of Smith's 'social organisation and relations they make possible'.2
There are, of course, many famous 'classic' texts on childcare, such as those by Hugh Jolly (1975, 1977) and Benjamin Spock (many editions, from 1946 onwards, the most recent (co-authored with Stephen Parker) being 1998). On the subject of 'How a Husband Can Help', Jolly (1977) offers the following advice:
Your husband can help by being useful with the baby and sympathetic about chaos in the house; by not expecting you to feel like a mother overnight when your past life has been spent in a totally different way; and by taking the initiative sometimes to prevent you from becoming a drudge who never wants to leave the house. When a woman does need help from her doctor in adjusting to the new life, it is disastrous for her husband to scoff. His role is to try to understand the stresses she feels when she suddenly becomes completely responsible for their child, and to offer her practical sympathy. Arranging for domestic help, buying her a washing machine, taking her out regularly -allcan help. (19 77: 169; emphasis added)
And, from Spock:
Some fathers have been brought up to think that the care of babies and children is the mother's job entirely. But a man can be a warm father and a real man at the same time. (19 78:41)
At the time of their publication the modern Women's Movement was well underway, but it seems to have had little impact on these texts (though Spock seemed to be trying, however oddly, to move away from traditional asymmetries). However, given the continuing changes in gender relations and gendered social
Sunderland: Discourses of fatherhood      251
practices at all sorts of levels, the evidence that Anglo-American men's involvement in childcare and housework is increasing (e.g. Parke, 1996), and the high profile that gender relations as a topic now enjoys, it might be expected that more recent parentcraft literature would manifest more symmetrical representations of motherhood and fatherhood. And, indeed, this seems to be the case. Diane Richardson, in Women, Mothering and Childrearing (1993: 51) observes that 'By the mid-1980's ... most authors of childrearing advice books encouraged fathers to become more involved in the care of their children.' The following analysis, however, suggests that shared parenting still has some way to go.
The data
I became a mother some 5 years ago, and hence found myself in a good position to conduct a small-scale study of what contemporary parentcraft literature has to say to, and about, fathers and mothers.3 I chose data from several source texts (summarized on p. 6, cf. Theo van Leeuwen, 1993). They included those that I was given by my doctor and the local hospital when I was pregnant in 19 9 3, and two my partner and I received as presents, prior to and just after Emily's birth. The literature I was given by members of the medical profession (Pregnancy Book, Birth to Five, The Bounty Babycare Guide, The Bounty Infant Health and Feeding Guide, and Your First Baby) would normally have been given to every pregnant woman at that time who had informed her GP of her pregnancy and was regularly visiting him or her, and/or a local hospital. The booklet 'It's Your Baby Too: a Guide for Fathers' was also free and widely available, but may not have been given to or picked up by every mother and/or father.
Having chosen the source parentcraft texts based on how I had encountered them as a parent, I chose the texts to be analysed for their representation of fatherhood and motherhood according to specific social practices, rather than looking at the whole of one source text, or at random passages of the same length (as in some other corpus studies).
I was thus not interested in constructing an argument around the statistical distribution of lexical items. The consideration of, say, the distribution of pronouns in one source text, or over random or particular texts, might obscure very striking differences in relation to specific practices. For example, sections on babies' ailments which refer to 'parents' and, when a pronoun is needed, make use of the word you, might balance out other texts on other practices in their overall statistics, but this would not mean that the source texts or corpora were close to a gender-symmetrical representation of practices.
Neither was I interested in constructing a 'text type' or a genre of 'parentcraft literature'. Treating parentcraft literature as a genre, and the practice of parenthood as an undifferentiated whole, might have meant missing crucial asymmetries. Parentcraft texts, being themselves so-named, and use of the termparent, 'so as not to distinguish between fathers and mothers', can be seen as implying that parenthood is shared equally' (Deborah Lupton and Lesley Barclay, 199 7: 88,
252      Discourse & Society 11(2)
91). However, although this can be seen as constructive and encouraging as regards equal parenting, it can also be seen as being gender-blind, inaccurate and disingenuous, glossing over a clear case of gender asymmetry.
In a textual analysis based on social practices, the practices themselves provide microcontexts for the texts. I decided to look at the treatment of two particular areas of practice: (a) what fathers are represented as doing, and (b) the practices associated with expressing breast milk.4 Texts on fathers are interesting because the care of new-born children can logically be written about (and sometimes are) with no mention of the father at all. (By contrast, because of the possibility of breastfeeding, it is hard to imagine one which does not refer explicitly to the mother.) This means that references to the father - whether in passing, or in sections addressed to him and entitled something like 'Fathers', 'The Role of the Father', or 'For the Father' - are marked, and therefore deserving of investigation.5 I chose to look at texts on expressing breast milk for a similar reason. Since one reason for a mother expressing milk is that the father can then feed the baby - to give him pleasure, make him feel more involved, ensure he has some responsibility, allow him to 'bond', or allow the breastfeeding mother to have a few hours when she can choose not to be with the baby - from a gender equality perspective, I would regard it as salient if such a text did not make some reference to the father when dealing with this topic.
table 1. Texts on what fathers are represented as doing
	Text	Source text	Text addressed to?	Distribution
1	'Home from hospital' and 'Helping with baby'	It's Your Baby Too	fathers	free
2	'Dad Chat'	Your First Baby	fathers	free
3	'The modern	Ľr Miriam	mothers and	commercial
	father'	Stoppard's New Babycare Book	fathers (?)	publication (book)
4	'How do you	Nanny Knows Best;	'... mothers, or for	commercial
	determine when	How to Bring Up a	nannies of course,	publication (book)
	bedtime should	Happy Child	but mainly ...	
	be?'		mothers' (p. 1)	
5	'The father's role'	The Bounty Babycare Guide	mothers	free
6	Accepting help'	'Coping together',	'the magazine for	commercial
		in M and M	the mum-to-be and new mother'	publication (magazine) but free copies available
Sunderland: Discourses of fatherhood      253
TABLE	2. Texts on expressing	l breast milk		
	Text	Source text	Text addressed to?	Distribution
5	'The father's role'	The Bounty Babycare Guide	mothers	free
7	'Putting the baby to the breast'	Your First Baby	mothers	free
8	'Expressing your milk'	The Bounty Babycare Guide	mothers	free
9	'Coping with breast feeding'	Birth to Five	mothers and fathers (?)	free
10	'Expressing milk'	Pregnancy Book	mothers and fathers (?)	free
11	'How to express breast milk'	The Bounty Infant Health and Feeding Guide	mothers'	free
The result was a corpus focused on specific practices which created the possibility of studying the microcontext of specific textual choices, and the range of representations of these practices - omissions as well as realizations (cf. Van Leeuwen, 1995,1996).Thisgaveme 11 texts, in two groups, listed in Tables 1 and 2 (full references are in the Bibliography). Text 5 appears in both tables since it makes reference to expressing breast milk within a discussion of 'the father's role'.
As indicated in the fourth column of these tables, the implied audience for the source texts is diverse. Even prior to analysis, this diversity can be predicted to have shaped the lexical choices of the selected texts, and the distribution of these choices - and, accordingly, the range of discourses running through the selected texts. Thus, while texts in the commercially-produced books and M and M magazine can address more specific audiences, texts in the booklets distributed free to all mothers might be expected to make fewer assumptions about families' arrangements for childcare.
The unit of analysis, the 'text' in Tables 1 and 2, is therefore a selected extract, However, I have attempted to contextualize the first six of these extracts as far as possible by giving in the Appendix the whole of the short text (usually a clearly marked section of the source text) in which mention of the father occurs. (For reasons of space, I have not included the five additional texts on expressing breast milk; these, however, are less crucial to this paper.)
Discourse analysis applied to parentcraft literature
Drawing on the recognition of Critical Discourse Analysis and of Dorothy Smith (1990), that language has a role to play in actively constructing social practices,
254      Discourse & Society 11(2)
parentcraft texts can be seen as specifically shaping practices surrounding/arfrer-hood and motherhood, and accordingly as constituting both gender identities and gender relations. 1 am regarding the grammar and vocabulary in these texts as systems of choices from which writers can select. 1 am also assuming that these choices have non-equivalent meanings, and that choices, though not necessarily intentional, and almost certainly not conspiratorially motivated, are meaningful (Norman Fairclough, 1992). In doing so I hope to show that these 11 texts, through their language, realize a range of gendered discourses, in the Foucauldian sense of discourses as 'different ways of structuring knowledge and social practice' (Fairclough, 1992: 3), or, more simply, as 'ways of seeing the world'. These discourses thread their way through the texts, positioning fathers and mothers differently, each discourse doing this in a particular way and hence suggesting a particular representation of gender identities and relations. The majority of the co-existing discourses here are, I suggest, mutually supporting, even though, as Wendy Hollway writes (with reference to her study on sexuality):
at a specific moment several co-existing and potentially contradictory discourses ... make available different positions and different powers for men and women. (1984:230)
I will not be looking at what is often dealt with under the heading of 'sexist language' - most obviously, texts referring to the baby as he (which several still do, though others alternate between he and she), and the frequent but often inaccurate use of husband and wife. Since discourses can be realized by non-sexist language items as well as by sexist language items, discourses would seem to be more relevant to the study of gender identities and relations. A pornographic text, for example, may objectify women without using a single linguistic item which of itself degrades, trivializes or defines women. As Deborah Cameron points out, the real problem is meaning, which does not always correlate with linguistic form (1994: 29). She observes that
... sexism in language exists below the surface, so that superficial reforms (like proscribing some finite set of offensive forms or making all texts formally gender neutral) are insufficient to combat it. Many instances of sexism are manifested not in single words or specific constructions but through an accumulation of discursive or textual choices; this kind of sexism will always elude the mechanical application of a standardizing rule. (1994: 32)
The aim of this paper is to show this 'accumulation of discursive or textual choices', not because texts determine audience responses (which are also important; see Cathy Urwin (1985), on interpretation of parentcraft texts), but because they can show what is assumed about the world, what practices are possible, what practices are not thought of, andperhaps what practices seem too obvious to be mentioned. Thus, rather than finding meaning in audience accounts, or solely in the texts, I am attempting to provide a description of the kind of world the texts assume.
The study of gendered parenthood behaviour is not new to linguists (e.g. Jean Berko Gleason, 1973), and as already indicated, Urwin (1985) included the
Sunderland: Discourses of fatherhood      255
interpretation of parentcraft texts in her study of the 'construction of motherhood'. Gendered discourses related to sex have also been identified before. Through discussions and dialogues about sexuality, Hollway (1984) arrived at the 'male sexual drive discourse', the 'have/hold discourse' and the 'permissive discourse'. Harriette Marshall (1991) has identified three main discourses of motherhood in childcare and parenting manuals, which she ironically titled (1) 'Ultimate Fulfilment', i.e. what women find in motherhood; (2) 'Happy Families', i.e. the assumption that the child is living with its biological parents, a contented heterosexual couple; and (3) 'Sharing the Caring', i.e. where mothers and fathers are both involved in childcare. Marshall has also pointed out that 'sharing' usually means that the father is responsible for the most positive aspects of childcare, while the mother does the maintenance work. Marshall's was, however, not a detailed linguistic study. And though Lupton and Barclay (1997) do write on the discourses and experiences of fatherhood in a range of contexts and genres, this is largely from a sociological perspective. The study reported here, in contrast, primarily explores fatherhood through detailed text analysis, and uses booklets, brochures and magazines as sources of data, as well as manuals.
The analytical framework
Hollway writes that
I arrived at [the sexuality discourses] through a combination of my own knowledge and what was suggested by the data (an approach which Glaser and Strauss (1967), call 'grounded theory'). Certainly my assumptions and those of the research participants share a largely common historical production; they will also be recognizable to most readers.
Such an approach, though appealing, and indeed productive, is likely by itself to be problematic to linguists - in particular, what is the relationship between what is 'suggested by the data' and particular linguistic items? To investigate this relationship, i.e. that between any discourses which 'emerge' from the texts and the language which realizes them, a textual analysis is also necessary. 1 have therefore tried to supplement 'grounded theory' with greater consideration of the language of the text itself, so that any discourses which do 'emerge' are an effect of the interface between the analyst's knowledge, and the actual data. The procedure is largely one of 'shuttling between' emerging discourses and relevant linguistic items, making each given discourse both the object and the result of analysis. This means that rather than drawing on assumptions which are shared and recognizable, 1 am also offering features of the text as a common ground between myself and the reader. In this way 1 hope to show, in terms of these features of the text, how one discourse is selected as 'dominant' or hegemonic, and when and how one specific discourse rather than another can be seen as prevailing.
I am also using 'my own knowledge' to identify what is not instantiated in these specific texts (in terms of linguistic items), when my own experience suggested
256      Discourse & Society 11(2)
that it might be. Such an approach allows the researcher's own perceptions (e.g. of which linguistic items are striking in their (in)frequency or absence) and understanding to be drawn on, and seen as valuable for the insights they offer.
In order to see what sort of 'gendered identities' are being constructed through discourses of parenting in these texts, I thus follow CDA in looking at what grammatical and lexical choices have been made from those available, thereby suggesting a certain view of the world, and at what others could have been made, and from an equal opportunities perspective should have been, but were not. In looking at what is not said, as well as what is - at both absences and presences, and at what is made explicit and what left vague - I have found it helpful to draw on parts of van Leeuwen's (1996) framework for analysing representations of 'social actors'. Of interest for gender relations and identities in a text is who is included and who excluded, when, in what contexts, and linguistically how - the 'who' in this case, being the social actors of mother and father (also potentially lexicalized as wife, husband andpartner). As Van Leeuwen observes, some exclusions 'maybe "innocent", details which readers are assumed to know already' (1996: 38). Others, however, may be intentional. Van Leeuwen proposes two 'exclusion' concepts: 'suppression' and 'backgrounding'. These may both be realized similarly in language: classically by passivization and agent deletion, for example, but also by process nouns and nominalizations. Thus, van Leeuwen gives actual contextualized examples of immigration (unaccompanied by an indication of who is immigrating) and support (unaccompanied by an indication of where this comes from). However, whereas suppression is 'radical exclusion' which 'leave[s] no traces in the representation, excluding both the social actors and their activities', in backgrounding
the excluded social actors may not be mentioned in relation to a given activity, but they are mentioned elsewhere in the text, and we can infer with reasonable (though never total) certainty who they are. They are not so much excluded as de-emphasised, pushed into the background. (Van Leeuwen, 1996: 39)
Where fathers are occasionally suppressed as social actors in parentcraft texts, these days I would suggest that they are more likely to be backgrounded.6
Linguistic items drawn on
When reading these texts, certain linguistic items appear to recur, others to occur only once, others to be conspicuously and surprisingly absent. A frequency count of these linguistic items in terms of which social actor each is associated with shows that in Texts 1-6 - the 'What Fathers are Represented as Doing' texts -play(ing) referring to fathers in relation to their children occurs as a noun or verb 7 times, help(ing) referring to fathers in relation to mothers 5 times, and fun 3 times. Share, however, occurs only once, in Text 5.
In particular, then, I focus on the following linguistic items:
(a) the verbs play, help and share
(b) the nouns fun, help and play
Sunderland: Discourses of fatherhood      257
(c) mother■/'fatheri'wifei'husband/'partner, as well as substitution forms such as someone else
These items can be seen as coming from three semantic fields: those in (c) from one, play/fun/help (the first and third, whether as nouns or verbs) in another, and share in a third. I am not assuming that the selection of these items is a 'neutral' one which all analysts would automatically arrive at; rather, the above inventory can be seen as a variable construct dependent on the analyst's knowledge and experience.
'Part-time father/Mother as main parent' discourse
I am going to suggest that current parentcraft literature clearly embodies the dominant discourse of what can be called 'Part-time father/Mother as main parent'. I am suggesting that this is a single 'combination' discourse because the two parts are consistent and complementary, with each part implying the other, and because they do not seem to originate from different sources. (The name, and others which follow, are clearly interpretative in themselves.) This discourse is manifested (in different ways and to different extents) in all 11 texts under discussion here. Though it may reflect what is frequently the case in practice, it does not reflect alternative and more egalitarian possibilities for childcare.
The 'Part-time father/Mother as main parent' discourse is clearly related to the fact that most parentcraft publications are, directly or implicitly, addressed to the mother. Apparently this was not the case in the early 19th century, but became so at the start of the 20th century, and largely remains so today (Lupton and Barclay, 1997: 88). The 'Part-time father/Mother as main parent' discourse supports a traditional division of labour, but can also be traced back to Freud and to psychological views of the importance of the mother-child relationship - for example, those advanced by John Bowlby (1969) and since largely discredited, e.g. the importance of mother-child attachment and of the dangers of 'maternal deprivation'. Nevertheless, Urwin (1985: 164) still points to an 'image of an almost totally child-centred mother ... reinforced by contemporary childcare literature'. And Lupton and Barclay identify a 'bonding discourse', which 'privileges the mother over the father as the primary care-giver' (1997: 43).
Through the absences and presences of the lexical items indicated (both), the two complementary parts of this discourse also illustrate the profoundly relational nature of gender. In particular, as we have seen, what fathers do is represented by the transactive verb help. There are instances of help in Text 1, 'Helping [her] with baby': the title, 1.1: 'You can help with the baby's care by changing nappies ....', and 1, 4: 'If your baby is bottle-fed you can help by making up and giving feeds' -the can, I suggest, indicating possibility rather than ability. Who the father is helping is not explicitly stated - this backgrounding of the mother suggesting that her role is too obvious to require mention. Help also appears as a noun in Text 2, 'Dad Chat', in the subtitle 'Your wife still needs lots of support and help with all the chores'; and in Text 6, Accepting help', addressed to mothers: 'If you do need to ask for help, go to your partner first' (1. 16) - the if and do taken together suggest-
258      Discourse & Society 11(2)
ing that the mother who does not take sole responsibility for the baby is the exception that proves the rule. Help applied to the father can be seen as 'vague language', compared to other possible verbs such as bathe, change, wash, cook, as well as those verbs used in reference to the father in Text 2: limit [the number of visitors], and pur [locks on all the low cupboards]. This vagueness is in one sense surprising since these texts are supposed to be useful in a practical sense to uninitiated novice parents. However, this vagueness can also be seen as legitimizing the father's parr-time, essentially supporting and non-responsible role, in that in most of these cases of help, no actual, specific tasks are represented for him. These different uses of help thus suggest that the father has some role to play in childcare, in that he is expected to act in ways which have an effect on the world around him, but that he will not be the one left holding the baby.
Other, non-recurring lexical items which help to realize the 'Part-time father/Mother as main parent' discourse are 'she may need a short, regular break' (Text 2,1. 2 7), and references to the father 'stepping in' (Text 1,1. 20) and 'taking a turn at night-time care' (Text 5,1.21). Stepping in and taking a turn - likewise vague lexical items - are in the same semantic field as help, as are material verbs like that addressed to the father in Text 1, 'Home from Hospital': 'give her a break' (1. 20). Though this give is 'integratively transactive' (van Leeuwen, 1995: 90), in that the 'goal' of give is a human being, her, the father is nevertheless being positioned as someone empowered to 'give' the mother a 'break' (i.e. before she returns to caring for the baby) - or not!
As representations of material action, these verbs are normally transactive (van Leeuwen, 1995: 89) in that they implicitly or explicitly involve more participants than the social actor who 'does the deed' (you would normally help someone, share something with someone, and probably play with someone). However, transactive social action can be de-agentialized or deactivated- and there are cases of each here. Fathers' transactive social actions of assistance and support are represented as de-agentialized, in 'playing with your baby will help him get used to his environment' (Text 1,1. 22), and as de-activated, i.e. 'as though they were entities or qualities rather than dynamic processes', and in play in 'bathtime offers a good opportunity for play [with your baby]' (1. 2 7)) (van Leeuwen, 1995:93,95). Both de-agentialization and deactivation can be seen as forms of backgrounding.
Another manifestation of the 'Part-time father/Mother as main parent' discourse is the recurrent and somewhat mysterious phrase someone else, or its semantic equivalent another person, which occurs frequently in the texts on expressing breast milk, as follows (emphasis added):
Text 5:'... breastfeeding mothers are encouraged to express their milk, so that another
person can sometimes give the feed' (1. 43)
Text 7: '... if you want to go out and let someone else give the baby his feed' (1. 23)
Text 8: 'you can express your milk into a sterile bottle so that someone else can give the
feed' (1. 7)
Text 9: 'you may want to express some milk for someone else to give to your baby in a
bottle' (1. 48)
Sunderland: Discourses of fatherhood      259
Text 9: 'it can be an advantage later when you want to leave your baby with someone else' (1. 58)
In the first three of these the father is clearly 'suppressed' from these texts (though 'backgrounded' if the texts are seen in the context of their wider source texts - see Appendix for Text 5) since he is not even mentioned as a possible 'someone else' - though logically he could have been. In Text 9 he is 'backgrounded', since there is in fact a reference to a 'partner' between the two 'someone elses' of 11. 48 and 58. The father can thus be seen as included in these extracts, and personalized, but indeterminate ('when social actors are represented as unspecified, 'anonymous' individuals or groups ... [this is] typically realised by indefinite pronouns' (Van Leeuwen, 1996: 51)). Van Leeuwen suggests that in anonymizing a social actor, the writer 'treats his or her identity as irrelevant to the reader' (1996: 52) - not, I suggest, an effective way to indicate that one active and responsible parent might be the father. Of course, the writers of these texts may have been trying to be as inclusive as possible - to the extent of including lesbian family relationships - but it is nevertheless striking that there are so few references to the father, even as an example of 'someone else'.
Texts 10 and 11 do refer explicitly to the father ('The odd bottle of expressed milk can also be a chance for a father to have the pleasure of feeding his baby' -1. 11) and partner ('... if your partner would like to give a feed, you can express your milk into a sterilised bottle' -1. 2), respectively. These texts, therefore, do not background the father. Nevertheless, he does not seem to be indispensable: 'a chance', 'to have the pleasure of ...', 'if [he] would like to ...'. In both texts the father's involvement is mediated in such a way that in neither case is he the overt agent of feed (Norman Fairclough, personal communication).
An example of social (though not strictly linguistic) backgrounding comes from Text 4, from 'Nanny Knows Best' (originally a television series). I am not suggesting that this is typical of parentcraft texts in any way; rather, it is an illustration par excellence of the 'Part-time father' discourse. Nanny's response to the question 'How do you determine when bedtime should be?' includes the following:
Nowadays it seems quite usual for a mother to keep her child up so her husband can see him when he comes back late from the office. I would say 'Too bad'. The husband can peep at him whilst he is asleep and play with him at the weekend (italics mine). (1993: 126)
Text 3, 'The modern father', from Dr Miriam Stopparď s New Baby care Book, is the only text here to attempt to represent the father as anything other than a part-timer: 'The modern father is a father who takes responsibility for the general care of his child... the modern father is a full-time parent' (11.1,20). However, as I show later, this does not mean that his tasks are the same as the mother's. Stoppard attempts to construct her full-time father through what van Leeuwen calls 'categorization': the modern father. One way this 'categorization' is realized is through 'identification' (as opposed to functionalization), since the categorized actor is represented in terms of what he 'more or less permanently, or unavoidably' is, and, more specifically, 'relational identification', which 'represents social actors in
260      Discourse & Society 11(2)
terms of their personal, kinship or work relationship to each other'. Van Leeuwen observes that terms such as mother- but not father- are polyvalent in that
'mother' can be used as a functionalisation ('mothering' is not the act of bringing a child into the world, but the act of giving care to a child, while 'fathering' signifies only the act of begetting a child!) (emphasis added)
1 would suggest that Miriam Stoppard is here attempting to extend such poly-valency to the term father, not by using father as a verb, but (less convincingly) through such items as the existential is ('The modern father is active rather than passive') and the naturalised future modal will ('He will participate from day one with the care of the baby'), in characterizing his, 'relational' role. Stoppard is apparently adopting the familiar rhetorical strategy (though unique in this dataset) of making an appeal by assuming a favourable but inaccurate view of the audience, deliberately avoiding any problematicization of such claims (for example that, in most cases, the father of the 1990s is not like this). Presumably she feels that this strategy - like that mentioned earlier of creating texts which deliberately refer to parents - is more likely to help change traditional roles than an accurate portrayal.
The 'Part-time father/Mother as main parent' discourse is characterized by three salient lexical absences. The first is that of the importantly transactive verb share (which from an equal opportunities perspective one might have expected both in texts on fathers and in those on expressing breast milk). As indicated, there is only one occurrence of share, in Text 5: 'To achieve ['bonding'], a father needs to get acquainted with his baby through sharing in feeding, nappy changing, taking a turn at night-time care ...' (1. 18). Notably, the promotion of gender relations based on domestic equality is not a rationale given for such sharing. (This share is in fact share in - arguably another mediation of involvement.) The other two lexical absences have been made salient by my own experience. The first is the term laundry - a tedious and time-consuming but unavoidable aspect of childcare, as everyone knows who has had to deal with all the wet sheets, babygrows and one's own damp clothes that accompany a new baby. The second lexical gap is paternity leave - a curious omission, since several institutions in Britain (for example, universities and local government) do allow fathers to take a week's unpaid leave for this purpose, and did so when these texts were written. Text 2, 'Dad Chat', does not in fact mention taking leave of any sort when the baby is born, nor, surprisingly, does Text 3, Miriam Stoppard's 'The modern father'. As with the grammatical and lexical presences that have been identified, these absences, and the second in particular, can be seen as representing the father's part-time role as fixed, and incapable of being altered even for a week. To that extent they endorse the status quo as regards British social and institutional structures surrounding current childcare practices.
It is also worth looking at absences in the wider, though still discoursal sense. Lupton and Barclay note how in a recent edition of Baby and Child Care (Spock and Rothenberg, 1992), the authors
Sunderland: Discourses of fatherhood      261
emphasize the importance of fathers considering the care of their children to be as important as their careers, taking the time to share with their children and partners, putting family life as their first priority and letting it be known at their workplaces that they take their parental responsibilities very seriously. (Lupton and Barclay, 1997:89)
This, to my mind, is to be welcomed, and the 'letting it be known at their workplaces ...' idea is not one I have come across in other parentcraft literature. However, as Lupton and Barclay point out: 'None of these injunctions is directed at female readers in relation to motherhood. Rather, it is simply assumed that mothers will automatically adopt such priorities and approaches to their role as mothers' (1997: 89). It also seems to be taken for granted that they will not be returning to work after some weeks or months of maternity leave. The absence here is not simply a lexical one, but rather the absence of a whole 'New mother who also works outside the home' discourse. Instead, Urwin cites (admittedly older) references, for example in Penelope Leach's Baby and Child (1977), to the importance of childraising as a job (1985: 194), and to how emphases on the importance of stimulation and enrichment for young children textually constitute motherhood as well as childhood (1985: 196). This absent discourse, arguably waiting to be intertextually inscribed (Fairclough, 1995: 189), is a potentially destabilizing one for the specific companion discourses, identified in the following pages, which intratextually 'shore up' the overarching 'Part-time father/Mother as main parent' discourse.
I suggested earlier that there are in fact several supporting, specific gendered discourses evident in these texts. 'Within' the 'Part-time father/Mother as main parent' overarching discourse, three further fatherhood discourses can be identified. These can be seen as either running in parallel through the different texts or as 'bricked together' within texts. These three I have called 'Father as baby entertainer', 'Father as mother's bumbling assistant' and 'Father as line manager'. All are recognizably 'ways of looking at the world'; each has particular linguistic characteristics which allows it to prevail as a specific gendered discourse. And though at first sight they may seem to be potentially in competition with each other, suggesting, respectively, prepared performance, natural incompetence and strategic overview and detachment (Greg Myers, personal communication), I suggest that in practice they are not - trivially, when they occur in different texts, and, more importantly, because they are all to do with operating on the margins of childcare and therefore complementary at a higher level of abstraction, i.e. as specific discourses which each shore up the same overarching, hegemonic 'Part-time father' discourse.
'Father as baby entertainer' discourse
'Father as baby entertainer' is probably the most prevalent of the three specific fatherhood discourses, frequent lexical exponents being the material and potentially transactive verb play, and the nouns play and fun. In Text 4, from Nanny Knows Best, the father's sole role during the baby's waking hours is to 'play with
262      Discourse & Society 11(2)
him' - at the weekend.7 Nanny Knows Best may be atypical, but this discourse is apparent elsewhere, too. In Text 1, 'Helping with baby', we read about how 'playing with your baby will help get him used to his environment and help with his development' (1. 22), that bathtime 'offers a good opportunity for play' (1. 27), that 'babies are lots of fun so enjoy playing with him' (1. 31), and that the father 'should take every opportunity ... to play with [his] child as this is a vital part of his development' (1. 32). There is also one occurrence of cuddles (1. 3) and another of cuddling (1. 8).8 In Text 2, 'Dad Chat', the subtitle includes 'Enjoy this time together as a family' and ends with the injunction, in bold type, 'Have fun'. The father is reassured in the last paragraph that 'Babies are for loving and enjoying', and that 'Babies can be fun!' - positioning him as someone who might be expected to think they are just boring and entail a lot of hard work with no rewards. And even in Text 3, Miriam Stoppard's 'The modern father', though this father is expected to change nappies and get up at 2 am to feed the baby (1. 15), he will also (from 1. 10) 'spend time playing with [his children], showing them new things, helping them with their hobbies, taking them with him when he enjoys his own', and (from 1. 18) will 'participate with reading stories, playing games and singing songs before bedtime'. Despite Stoppard's claim that 'the modern father is a full-time parent, not a part-time stranger' (1. 20), and the transactive verbs which represent his social action as involvement with the children, the emphasis here is definitely on this involvement being through fun and enjoyment rather than through washing wet sheets. And even amid all the fun and enjoyment, the father's involvement is again covertly mediated: rather than simply doing something, he 'spends time', he 'participates', he has 'opportunities'. Despite her Utopian rhetoric, Stoppard's 'full-time parent' sounds not so very different from the 'part-time father' who inhabits all the other texts.
'Father as mother's bumbling assistant' discourse
The 'Father as mother's bumbling assistant' specific discourse can be seen in Text 1, from 'It's Your Baby Too', in 'Helping with baby', in which the imperative 'Remember' occurs three times: 'Remember never to leave your baby alone with a bottle' (1. 9), 'Remember to always check the temperature of the [bath] water before putting him in' (1. 28), and 'Remember that babies are rather fragile so don't be too rough [when you play with the baby]' (1. 3 6). This use of the cognitive remember represents the father as a thinker as well as a doer, but could be read as patronizing and hectoring, and definitely not as positioning the father as a competent parent. It is certainly hard to imagine a mother being addressed in this way. Since there is little or no mention of fathers doing these traditionally feminine tasks, Texts 2 and 4 do not adopt this same sort of positioning. Text 2 identifies a father's roles in different terms, which I will return to, and as we have seen, in Text 4, 'Nanny Knows Best', the father's role is minimal.9 Text 5, 'The father's role', however, refers to the possibility of the father being made to feel 'less confident' (1. 34). And Text 6, Accepting help', line 19, reads 'They [fathers] are often
Sunderland: Discourses of fatherhood      263
keen to be involved, but are not sure how to go about it' - which is again positioning men as something other than competent carers. Fathers are represented here as people who feel (as well as think and do), but, as in Text 5, the problem they are supposed to have, is here identified as lack of self-confidence, of being unsure - an individual, psychological explanation, rather than one drawing on societal practices, such as paternity leave not being standard, playgroups being referred to as 'Mother and Toddler' rather than 'Parent and Toddler', nappy-changing facilities being located in women's toilets rather than both women's and men's (or in a third area entirely), and fathers facing particular disapproval if they take leave from work to look after sick children.
An alternative explanation that is sometimes given for the father's lack of competence is that fathers feel left out because of women's own attitudes and practices, which leads to women being seen as to blame. Marshall cites a recent National Childbirth Trust publication:
Some people think that babycare is 'woman's work' and this can cause problems. It may be that the new mother gets pleasure from feeling indispensable, or that she has old-fashioned ideas about gender roles. By monopolizing the baby she can frustrate her partner's desire to be a tender and caring father, and can hurt him very much. (1987, quoted in Marshall, 1991)
There is also a version of this in Text 5, 'The Father's Role': A mother can unwittingly delay ["father-baby bonding"] by discouraging offers of help' (1. 23). This may lead to her partner 'feel[ing] less confident about offering her a "break" in future'(1. 31).10
Of the six 'To the father' texts, Text 3, 'The modern father', is the only one not to suggest that the father might not be up to scratch in his (fun-oriented) parenting role.
The 'Father as mother's bumbling assistant' discourse can be seen as being related to the idea of 'maternal instinct', and the view that just as 'mother love is natural', so, by extension, is childcare on the part of the mother. This is exemplified by advice to the mother such as 'Do whatever suits you: so long as your baby is loved, fed and washed he will be all right' (from the 1986 edition of Hugh Jolly; quoted in Marshall (1991: 72)), and, here, in Text 6, Accepting help', addressed to mothers, 'Follow your instincts' (1. 43) - a far cry from the 'bumbling assistant' positioning assigned to fathers in Texts 1, 5 and 6. The first part of Text 1, 'Home from hospital', which refers to the father having to 'adjust' (lines 3 and 6 - an individualistic and non-transactive social action), suggests that taking a few days off work will help him 'understand the new routine', and talks about how 'experience' will help him learn why his baby is crying (1. 26) - nothing very instinctive there! Drawing on a poststructuralist perspective, to describe the 'expert discourses' on fatherhood in academic research, Lupton and Barclay note that
The notion of the pre-existing subject who after becoming a father finds himself so distorted and stressed that he can no longer identify his 'real self and must painfully transform into a 'new self is dominant... (199 7: 46)
264      Discourse & Society 11(2)
I would suggest that this same subject is evident in parentcraft texts in the shape of the 'bumbling assistant', and that any such painful transformation mothers might have is represented as muted in comparison.
'Father as mother's bumbling assistant' is complemented by a motherhood discourse which also supports the 'Part-time father/Mother as main parent' discourse: 'Mother as manager of the father's role in childcare'. This discourse was also evident in Text 6, Accepting help', in the claim that men 'are often keen to be involved' (1. 19), and in Text 1 (from 'It's Your Baby Too'), 'Home from hospital': expect to be called upon to step in' (1. 19) - a truncated passive, agentless because, despite the fact that she is backgrounded here, there is no doubt about the identity of the social actor. As 'manager', it is the mother who will 'call upon' the father to help with childcare tasks. In Text 6, Accepting help', the mother is urged to 'try getting your partner to take some of the strain while you recover' (1. 26). Try however suggests that the father may not be as easy to 'manage' as all that, one reading of which may be a warning that his help should not be taken for granted.11 And in Text 4, the implication seems to be that the mother should channel her 'husband's' parenting practices into peeping at the baby while 'he' is asleep and playing with him at the weekend'.
'Father as line manager' discourse
The third specific 'Part-time father discourse', 'Father as line manager', is evident in Text 2, 'Dad Chat'. Here there are several highly transactive imperative verbs which contrast strikingly with the 'you can help with' and 'you can help by' of Text 1, by positioning the father as the protecting, almost patriarchal 'Head of the Family' (the bold typeface is mine):
... limit the number [of visitors] you invite into the house, stop them from disturbing the baby when he is asleep. Ensure that your family routine is protected, (paragraph 2)
And, in paragraph 5, '... plan for the future. Get yourselves a stair-gate, put locks on all the low cupboards, protect electric wires and sockets'. And again, this time as regards the mother: 'Go out together for a treat that is fun for both of you. Let her know that you still enjoy being with her' (paragraph 3). This last invokes the spectre of the logical converse - that new motherhood may in fact make a woman less enjoyable as a companion than she previously was. In the representation of the father's role as 'line manager', the mother herself is backgrounded in the phrases 'Get yourselves a stair-gate' and (though in a less extreme way) 'Go out together for a treat that is fun for both of you.'
'Mother as manager of the father's role in childcare' can be seen as a companion discourse to 'Father as line manager' (as well as to 'Father as mother's bumbling assistant'). However, though there are no actual examples in these texts, there is also potential for conflict here: what mother asks father to do for the baby may not correspond with what father thinks is good for the family, while father's
Sunderland: Discourses of fatherhood      265
way of looking after mother and the baby may conflict with what mother thinks is good for the baby. Importantly, then, specific parenthood discourses can come into intertextual conflict, and can thus create instability in the overarching discourse.
These four discourses of fatherhood (one dominant, three specific) are realized in this analysis of these texts in different ways. Linguistic absences play a role, notably: share, laundry and paternity leave. Secondly, realization sometimes takes place without recurrence: although I have focused on repeated forms, single instances of a linguistic form (e.g. 'men are often keen to be involved ...') are also relevant. However, the main way the different discourses are realized in this analysis is through different recurring linguistic items. Thus, 'Part-time father/Mother as main parent' is realized through the recurrence of help and other verbs in the same semantic field, and of indefinite pronouns such as someone else, which background the father. 'Father as baby entertainer' is realized by recurrences of play, fun and enjoy. 'Father as mother's bumbling assistant' is realized through repetition of the verb remember and other, variously-realized suggestions of paternal incompetence. 'Father as line manager' is realized through a series of imperatives and a degree of backgrounding of the mother. And though other analyses would undoubtedly have identified different fatherhood discourses, they would still, I suggest, need to account for these recurring linguistic features.
The textual Organization of discourse
In any analysis based on discourse identification, the analyst has the problem of not only recognizing a discourse, but having principled ways of (a) distinguishing it from another discourse (with which it may or may not be co-occurring), (b) deciding whether two such discourses are 'combination', being two sides of the same coin, like 'Part-time father' and 'mother as main parent'; 'complementary' and hence running in parallel or even 'bricked together'; 'contradictory' and hence destabilizing; or simply 'unrelated', and (c) identifying discoursal hierarchies, if it is claimed that some specific discourses are shoring up other, overarching discourses. Such an analysis must be a high inference one, always involving interpretation, but if this interpretation comes from the interface between the analyst's knowledge and experience and the actual language of the text, then the analysis will be more principled for not being merely 'suggested' by the data but supported by findings which are 'recognizable'.
In some texts in this corpus, only one specific discourse is evident, and in others no specific discourses are. The specific discourse running through Text 3 would appear to be 'Father as baby entertainer', while in Texts 5 and 6 'Father as mother's bumbling assistant' is particularly evident. In Texts 7-11, 'Part-time father' seems to be unaccompanied by any specific fatherhood discourse -only by its 'other face', 'Mother as main parent'. In each of Texts 1, 2 and 4, however, there appear to be two co-occurring specific discourses: in Text 1,
266      Discourse & Society 11(2)
'Father as mother's bumbling assistant' and 'Father as baby entertainer'; in Text 2, 'Father as line manager' and (though rather less saliently) 'Father as baby entertainer'; and in Text 4, 'Father as baby entertainer' and, implicitly, 'Mother as manager of the father's role in childcare'. There does not appear to be any tension between these three pairs of differently 'bricked together' discourses, but rather a complementary intertextuality, the 'synthesis' in Text 1 being that father does not have a problem about being a baby entertainer, he just needs advice about the relevant safety aspects; the synthesis in Text 2 is that these roles are both peripheral to the essentials of childcare, and in Text 4 that mother's management of the father's role includes the appropriate channelling of such interest in his child as there is assumed to be into 'playing' at weekends. The idea of complementary, specific discourses (in this case, 'Father as line manager', 'Father as mother's bumbling assistant', 'Father as baby entertainer' and 'Mother as manager of the father's role in childcare') shoring up others (in this case 'Part-time father/Mother as full-time parent') intratextually in the same set of texts is not unlike Lupton and Barclay's claim that there are 'identifiable hierarchies' of discourse: 'Depending on the context, some discourses are hegemonic over others, taking charge over the definition of what is considered to be "truth" ' (1997: 9). It can also be compared with Bronwen Davies' use of the palimpsest metaphor for textual meaning:
This [palimpsest] metaphor ... is derived from the image of writing on parchment, writing which was only partially erased to make way for new writing, each previous writing, therefore, bumping into and shaping the reading of the next layer of writing. (Davies, 199 7:2 75)
While noting the problematic nature of this (i.e. the implicit suggestion that there is an 'original' version of text on the palimpsest), Davies also observes that 'its power as a conceptual tool requires a focus on the multiple layers of writing ...' (19 9 7: 2 75). And while Davies sees its value in explaining 'the ways in which the subject is written and overwritten through multiple and contradictory discourses' (my italics), these parentcraft texts, with their multiple complementary discourses, which shore up yet another necessarily complementary discourse, and their occasional, potential conflicting discourses (I have identified one source of conflict, between 'Father as line manager' and 'Mother as manager of the father's role in childcare'; there may be others) can also accordingly be seen as a form of palimpsest. Parentcraft texts also 'write and overwrite' the subject, though largely in non-contradictory, and hence largely (but not entirely) non-destabilizing ways.
Fatherhood, motherhood, gender relations and the 'shape' (?) of gender in parentcraft texts
Like masculinity and femininity in the wider sense, I suggest that fatherhood and motherhood are relational and mutual constructs, to the extent that they con-
Sunderland: Discourses of fatherhood      267
struct each other even (or perhaps especially) when one of them is not mentioned.
In addition to the three specific fatherhood discourses which shore up the 'Part-time father/Mother as main parent' discourse, I also identified a 'Mother as manager of the father's role in childcare', which I suggested operated as a companion discourse to 'Father as mother's bumbling assistant', and a companion but potentially conflicting discourse in relation to 'Father as line manager'. So while the mother is (attempting to) manage the father, but only in his role as a father, the father is (attempting to) manage the mother in the wider sense of 'Head of the Family'. Just as they are continually being discoursally shaped, then, the shaping of motherhood and fatherhood can also be seen as continually shaping gender relations. This is also evident in a further motherhood discourse, 'Mother as wife/partner', of which in these extracts there are examples in Texts 5 and 2. Text 5, 'The Father's Role', reminds mothers of the risk that 'your partner may feel "left out" and rejected' (1. 11) - though whether this is read as a father, or as a companion and sexual partner, will depend on the reader. The mother is herself backgrounded in this clause, but a likely inference is that she would be the social actor responsible for such rejection. 'Dad Chat' (Text 2) refers to 'your wife' being 'a mother and wife' (1. 26), and 1. 17 reads 'They [most mothers] need to have times when they are women instead of mothers' (my italics). This binary representation of the same social actor reads incongruously to me as someone who is both, since it denies the idea of positive multiply-gendered identities, and indeed even invokes traditional madonna-whore dichotomies of women. Presumably the writer meant by a woman someone who is sexual, i.e. currently sexually active, or at least sexually interested/interesting, whereas, paradoxically, mothers are not supposed to be categorized in this fashion (or are supposed to be, but in a decidedly more functional way).12
The discourses identified can thus be seen as contributing to the shaping of both gendered parenthood and wider gender relations, and to the 'shape' of gender itself, i.e. the relationship between masculinity and femininity (Sally Johnson, 1997: 22, 23). The word 'shape' entails the possibility of a visual representation of gender (e.g. Sunderland, 1998) as something one has or is. The shape of gender in the dominant parentcraft discourse is an asymmetrical one, 'Mother as full-time parent' femininity occupying more of the 'shape' than 'Part-time father' masculinity. The same can be said for the shape of gender in the case of the companion discourses 'Father as mother's bumbling assistant' and 'Mother as manager of father's role in childcare': in both cases influential, even powerful movement within the 'shape' come from the femininity 'side'. In the potentially conflicting discourses 'Mother as manager of the father's role in childcare' and 'Father as line manager', however, the shape, or rather the 'morphology', is more complex: though once again there is asymmetry (masculinity this time figuratively occupying the more generally influential and powerful part), there is more obviously a potential for two-way conflictual interplay. Arguably, then, in parentcraft texts we can see gender as a dialectical 'bundle' of relational
268      Discourse & Society 11(2)
femininities and masculinities, most being 'companion', mutually supporting ones, others potentially conflictual and destabilizing.
Gender, however, may not be best seen as something one has or is, but that one does, i.e. as performative, 'constituting the identity it is purported to be' (Judith Butler, 1990). The focus for any analysis of parentcraft texts drawing on the idea of performativity rather than identity would then need to be on parenting actions and 'deeds', rather than on 'social actors', and how these 'deeds' themselves constitute what Butler would call the 'expressions' of fatherhood and motherhood (1990: 25). Interestingly, the selection of texts according to the gendered practices they represent - as has been done here - would seem to be consistent with a future conceptualization of gendered parenthood as performance.
Conclusion
My own position is that I would like to see more reference to fathers as central. However, parentcraft text writers are in something of a dilemma. They are writing for a huge and very diverse audience, with culturally very different current childcare practices. This audience includes single mothers (including those who have chosen to bring up a child alone), and lesbian mothers living with partners, as well as mothers from cultures in which there is as yet no discourse of what we can call 'fathers' full involvement in childcare'. The mother is the only (almost) guaranteed common denominator. The Bounty writers, in particular, cannot target an audience, since their booklets are automatically given to all pregnant women and new mothers registered with a GP and a hospital. These writers will, in addition, be well aware of the much quoted statistic that in the UK one in three babies is born out of wedlock and that, of these, half have parents who do not live together. This means that one in six babies does not live with both parents, and the majority of these babies live with their mother rather than their father. It is perhaps not surprising, then, that texts are full of references to mothers rather than fathers. Writers may feel that since mentioning a father who is not around is inappropriate, and may not be well received, the safest thing to do in order not to alienate readers is to explicitly mention only the mother. Some readers will, however, see this as regrettable.
I have suggested that it is possible to see one hegemonic, overarching discourse in these texts - a 'Part-time father/Mother as main parent' discourse. The existence of this discourse is perhaps predictable - though problematic, given ongoing changes in gender relations, not least through changing patterns of employment. What I have also tried to show is how this dominant discourse is shored up by other discourses: 'Father as baby entertainer', 'Father as mother's bumbling assistant', 'Father as line manager', 'Mother as manager of the father's role in childcare' and 'Mother as wife/partner'. Gender in these texts is essentially relational: they suggest that a part-time father can really only be expected to play with his children, and to perform certain tasks inexpertly, under the guidance of
Sunderland: Discourses of fatherhood      269
the mother, whom he is after all supporting, as well as managing the family as a whole. Textual analysis of these parentcraft texts, I suggest, has illustrated this relational nature of gender. It has also offered a rich illustration of the capability of discourses - variously running in parallel, being 'bricked together', and existing in a hierarchy, with their salient linguistic presences and absences - to organize a set of texts on the same social practices in simultaneously supporting and potentially destabilizing ways.
ACKNOWLEDGEMENTS
I would like to thank Roz Ivanic for her enlightening teaching sessions on Theo van Leeuwen's work on the representation of social actors and social action, Norman Fairclough and Greg Myers for reading and making valuable comments on earlier versions of this article, participants of the Critical Discourse Analysis meeting at Birmingham University (April 1999) for their many important insights, observations and perspectives, and an anonymous reviewer for further interesting and constructive observations on this paper.
NOTES
1.  Hollway writes 'I consider that heterosexual relations are the primary site where gender difference is re-produced' (1984: 228). I am suggesting that texts pertaining to the sexual side of heterosexual relations provide a site for the study of gender difference par excellence, through the language of these texts. Almost any text (including its visuals) on any of the following topics would be a fruitful source of data on gender identities and relations, in that it is likely to contain several gendered discourses: sex education; puberty, including the onset of menstruation; sex itself (i.e. the whole range of sexual practices); sexuality (Hollway, 1984); HIV, AIDS and other sexually transmitted diseases (Jewitt, 199 7); conception; contraception; abortion; pregnancy; childbirth; and of course news reports and editorials on sexual scandals, the most salient recent one being the Clinton-Lewinsky saga.
2.  Parentcraft texts are not the only genre concerned with childcare and parenthood, of course. Other written genres include academic articles from the fields of social history, psychology, sociology and anthropology; teaching materials for would-be nursery officers doing an NVQ; medical texts for future and practising paediatricians; documentation from nurseries; paternity leave policies; many news items; profiles of famous fathers; and self-help books for men. Spoken texts include drama, television comedy and adverts (Lazar, 1998), and popular films (Lupton and Barclay, 199 7). And some texts whose discourses may shape motherhood and fatherhood may not focus on childcare at all: self-help manuals, for example, which deal with questions of personal identity and personal growth, and how being a parent may be seen as contributing positively to these, or damaging them.
3.  By parentcraft literature I mean texts on childcare written by professionals, such as doctors or midwives, for parents - mothers, fathers, or both. These texts are clearly written to constitute advice - though since parentcraft literature is written largely for a general audience, but by professionals in the field of health, it may blur the boundaries between popular and 'expert' literature (Lupton and Barclay, 199 7). Parentcraft texts can thus be seen as prescriptive, even if a range of prescriptions is there to be
2 70      Discourse & Society 11(2)
selected from. And parents - especially, I would suggest, first-time parents - will often consult such texts for advice, especially in the early months of a new baby's life.
Though such texts have actually existed since the early 19th century, Lupton and Barclay identify the large current number of such texts as one effect of the increase, since the 1960s, of 'experts' who 'monitor[ed] individuals in virtually every social setting' (1997: 36). Similarly, Urwin sees parentcraft texts as one manifestation of 'normalisation', which, pace Foucault, she observes has become 'a basis for regulating the rest of the [non-deviant] population' (1985: 165).
4.  For those not familiar with this practice, it refers to the mother either manually or by means of a breast pump expressing milk from the breast - with the option of saving this milk to be given to the baby later in a bottle.
5.  It would be instructive to look at texts which specifically address or refer to mothers, as well as those which address or refer to fathers, and, where appropriate, to compare them.
6.  Interestingly, in his own study of the representation of fathers in texts on schooling, van Leeuwen found that they 'were radically excluded [i.e. suppressed] in texts addressing teachers, but included in many children's stories, even if only briefly, during the breakfast preceding the first school day ...'(1996:39).
7.  Marshall (1991) in her examination of motherhood discourses suggests that 'fun' in parentcraft texts is not mentioned as much when the focus is on the mother, since the mother's experience overall is described as positive, and so there is no need to go on at length about 'fun'.
8.  In the latest (at the time of writing) version of 'It's Your Baby Too', entitled 'What you really need to know about becoming a new dad', though fun is still foregrounded, other tasks are explicitly listed, without being seen as 'help': Äs well as changing nappies, bathing the baby, and helping soothe away inexplicable crying fits, being involved also means having fun' (p. 13). And, encouragingly, in a section entitled '0: How can I help with feeding?', we read as part of the answer: 'Practically, you can help with "burping" your baby, as well as feeding him from a bottle if your partner expresses her milk' (p. 14). While still 'help', this practice was not suggested for new fathers in the earlier version of the booklet.
9.  It is in fact minimal throughout Nanny Knows Best. Interestingly, Nanny is not over-concerned about children living with single mothers. If the main carer is Nanny, and the mother is secondary, this displaces the father even further down the line - something which may be seen to apply even when there is no nanny.
10.  It would be illuminating in a different study to find out how claims such as this are actually interpreted and evaluated, by both mothers and fathers.
11.  I similarly remember reading another edition of the Bounty Babycare Guide in which there was a section entitled 'Five Ways to get His Help'.
12.  Ironically, this mother or woman binary, to be operationalized by men, is here supposed to be for the benefit of women.
DATA SOURCES
Bennett, V. R. (ed.) (1990) Your First Baby. London: Maxwell Marketing Services. Kohner, N. (1988) Pregnancy Book. London: Health Education Authority. Kohner, N. (1989) Birth to Five. London: Health Education Authority Manning, M. (ed.) (1993/4) The Bounty Babycare Guide. Diss, Norfolk: Bounty Publications. 'M'. (1993) 'Coping together'. M and M: the Magazine for the Mum-to-be and New Mother. London: HHL Publishing: 48: April/May.
Sunderland: Discourses of fatherhood      2 71
Rodway, A. (ed.) (1993/4) The Bounty Infant Health and Feeding Guide. Diss, Norfolk: Bounty
Publications. SMA Nutrition, (n.d.) It's Your Baby Too: a Guide for Fathers. Maidenhead: SMA Nutrition. Smith, N. and Grunfeld, N. (1993) Nanny Knows Best. London: BBC Books. Stoppard, M. (19 9 0) Dr. Miriam Stoppard's New Baby Care Book. London: Dorling Kindersley
REFERENCES
Butler, J. (1990) Gender Trouble. London: Routledge.
Cameron, D. (1994) 'Problems of Sexist and Non-sexist Language', in J. Sunderland (ed.)
Exploring Gender: Questions and Implications for English Language Education, pp. 26-33,
Hemel Hempstead: Prentice Hall. Davies, B. (199 7) 'The Subject of Post-structuralism: A Reply to Alison Jones', Gender and
Education 9(3): 2 71-83. Fairclough, N. (1992) Discourse and Social Change. Cambridge: Polity Press. Fairclough, N. ( 1995) Critical Discourse Analysis. Harlow: Longman. Gleason, J.B. (19 73) 'Code-switching in Children's Language', in T.E. Moore (ed.) Cognitive
Development and the Acquisition of Language. New York: Academic Press. Hollway, W. (1984) 'Gender Difference and the Production of Subjectivity', inj. Henriques
et al. (eds) Changing the Subject. London: Methuen. Jewitt, C. (199 7) 'Images of Men: Male Sexuality in Sexual Health Leaflets and Posters for
Young People', Sociological Research Online 2(2): 1-18. Johnson, S. (1997) 'Theorizing Language and Masculinity: a Feminist Perspective', in S.
Johnson and U. Meinhof (eds). Language and Masculinity, Oxford: Blackwell. Jolly, H. (19 77) Book of Child Care: The Complete Guide for Today's Parents. London: Sphere. Lazar, M. (1998) 'Then Something Wonderful Happened: Advertising a Lifestyle on Offer',
Centre for Language in Social Life Working Paper Series 9 2. Department of Linguistics and
Modern English Language, Lancaster University. Leach, P. (1977) Baby and Child. London: Michael Joseph. Lupton, D. and Barclay, L. (199 7) Constructing Fatherhood: Discourses and Experiences.
London: Sage. Marshall, H. (1991) 'The Social Construction of Motherhood: An Analysis of Childcare
and Parenting Manuals', in A. Phoenix et al. (eds). Motherhood, pp. 66-85. London:
Sage. Parke, R. (1996) Fatherhood. London: Harvard University Press. Richardson, D. (1993) Women, Mothering and Childrearing. London: Macmillan. Smith, D. (1990) Texts, Facts and Femininity. London: Routledge. Spock, B. (19 78) Baby and Child Care. London: The New English Library. Spock, B. and Parker, S.J. (1998) Dr Spock's Baby and Child Care. New York: Pocket Books. Sunderland, J. (1998) Review of B. Bergvall, J. Bing and A. Freed (eds) 1996. Rethinking
Language and Gender Research: Theory and Practice. Harlow: Addison Wesley Longman.
Sociological Research Online 1998. http://www.socresonline/3/2/sunderland.html Urwin, C. (1985) 'Constructing Motherhood: the Persuasion of Normal Development', in
C.  Steedman et al.  (eds) Language,  Gender and Childhood, pp.  164-202. London:
Routledge. Van Leeuwen, T (1993) 'Language and Representation: the Recontextualization of Participants, Activities and Reactions', Unpublished PhD dissertation, University of Sydney. Van Leeuwen, T (1995) 'Representing Social Action', Discourse & Society 6(1): 81-106. Van Leeuwen, T (1996) 'The Representation of Social Actors in Discourse', in C. Caldas-
Coulthard, and M. Coulthard (eds) Texts and Practices. London: Routledge.
2 72      Discourse & Society 11(2)
jane Sunderland is a Teaching 'Fellow' in the Department of Linguistics and Modern English Language, Lancaster University. In addition to the textual shaping of parenthood, her research interests are gender and classroom interaction, teacher talk around gendered texts, student-to-teacher classroom talk, and the use of e-mail in distance education. address: Department of Linguistics and Modern English Language, Lancaster University, LAI 4YT, UK. [email: j.sunderland@lancs.ac.uk]
Appendix
Home from hospital
All of a sudden there's a new little person at home. You will need time to adjust to this new situation. You may find yourself feeling a little left out as your partner and baby get to know each other. As you adjust and get to know your baby you will become more involved and these feelings will disappear. You might both find it useful if you can take a few days off work when your baby first arrives home. This will also help you to understand the new routine which will need to be set up.
Of course your partner may wish to take a break from this routine from time to time, especially if she has had a difficult day with the baby crying. So you should expect to be called upon to step in and give her a break whenever it's necessary.
Babies cry for different reasons. They can be too hot, too cold, still hungry, wet or soiled or sometimes just want some attention or a cuddle - do not panic. Experience will help you to know which.
25
Helping with baby
You can help with the baby's care by changing nappies, giving baths and giving cuddles. If your baby is bottlefed you can help by making up and giving feeds. During
5          a feed your baby may swallow air and
require winding. You can encourage him to burp by cuddling him over your shoulder or on your lap then rubbing his back gently. Remember never to leave your baby alone
10        with a bottle as he could choke.
Sometime between 3 and 6 months your baby will begin on solid food as well as his milk feeds. This is a time for experimentation, not only for the baby but 15         also for you, as you both get used to the
foods he likes and dislikes. This is enjoyable but can be messy, so arm yourself with bibs.
Playing with your baby will help get him used to his environment and help with his development. It will also help you to get to know each other. Bathtime offers a good opportunity for play using different toys in the bath. Remember to always check the temperature of the water before putting him
Babies enjoy lots of fun so enjoy playing with him. In fact you should take every opportunity you can to play with your child as this is a vital part of his development. But remember that babies are rather fragile so don't be too rough.
25
30
35
Text 1
figure 1. Extract from It's Your Baby Too: A Guide for Fathers. Reproduced by kind permission of SMA Nutrition.
Sunderland: Discourses of fatherhood      2 73
Dad Chat
Now that your baby is born, life will take on a different tempo. Enjoy this time together as a family. Your wife still needs lots of support andhelp with all the chores.   Have  fun.
During the first few weeks most babies waken for at least one feed or maybe two at night. Parents get desperately tired at this time. Life can seem one long round of feeds and bleary eyes. Take heart! The baby will settle into a routine which fits in with family life and your night life should return to normal.
Everyone loves babies, especially new ones. Visitors will descend on your home from all corners of the world. Fathers need to be very firm; limit the number you invite into the house, stop them from disturbing the baby when he is asleep. Ensure that your family routine is protected. Babies are not happy to be handled by too many people, it makes them fretful.
Most mothers need to feel precious and special to their partners. They need to have times when they are women instead of mothers. Go out together for a treat that is fun for both of you. Let her know that you still enjoy being with her.
If your wife returns to work after the birth of the baby do discuss with her just how you feel about this. It may be that there is a financial or social reason for her to 25 work, but she may need a short regular break from the baby, which may enable her to be a better mother and wife.
Babies turn into toddlers very quickly and toddlers are brilliant at finding all the unsafe areas in your house.      30 So even  at this stage,   plan for the future.  Get yourselves a stair-gate,   put locks on all the  low cupboards, protect electric wires and sockets.
Babies are for loving and enjoying. They are not so fragile as they look; they love to be touched and     35 spoken to, especially by Mum and Dad. Babies can be fun!
figure 2. Ex tract from Your First Baby, published by the Royal College of Midwives.
figure 3. Extract from New Parent by Miriam Stoppard (1998). Reproduced by kind permission of Dorling Kindersley.
The modern father
The modern father is a father who takes responsibility for the general care of his child. Fewer fathers nowadays are prepared to be strangers to their children,
5             missing out on all the good times in the
family and, most important, missing out on their children growing up. The modern father is active rather than passive. He will arrange his day to come home early
10           from work to see his children; he will
spend time playing with them, showing them new things, helping them with their hobbies, taking them with him when he enjoys his own. He will participate from
15           day one with the care of the baby, with
nappy changing, with getting up in the middle of the night, doing the two a.m. feed, helping with bath times, reading stories, playing games and singing songs
20           before bedtime. The modern father is a
full-time parent, not a part-time stranger, and everyone in the family benefits from this. A father who has a high interest in the
25           pregnancy generally stays interested after
the baby is born. Interest is positively related to how much he holds the baby in the first six weeks of life and also whether he goes to the baby when he or
30           she cries. Not unexpectedly, his attitude
affects his wife's enjoyment of pregnancy and motherhood. The happier he is about the pregnancy and the more he looks forward to fatherhood, the more she
35           enjoys the first few weeks of her baby's
life and, of course, the better the start to the baby's life. The better the father is at playing his role, the more important he becomes. As a woman I'm surprised that
40           men aren't more prepared to manipulate
the situation in their favour.
Text 2
Text 3
2 74      Discourse & Society 11(2)
3. How do you determine when bedtime should be?
My children have always gone to bed at some time between 6.30 and 7 p.m. from                              Text 4
the day they were born until they were about seven, when they went to bed half an hour later, having had their bath at 6.30 p.m. At eight years they had their bath at 7 p.m. and bed followed. It was always bathtime that was prompt and 5               bedtime that was a little bit flexible. I put my children to bed early because an
early bedtime was thought to be best for growing children.
Nowadays it seems quite usual for a mother to keep her child up so her husband can see him when he comes back late from the office. I would say Too bad'. The husband can peep at him whilst he is asleep and play with him at the weekend. 10             I do not see any point in keeping children up. A young parent may feel it is
charming, having the new baby around all evening to fuss over, but it does over-stimulate the child. It also means that when the parents one day want their evenings to themselves again the routine they have established is hard to break and their child may be very reluctant to be put to bed (see question 4).
FIGURE 4. Extract from Nanny Knows Best by Nanny Smith and Nina Grunfeld (1993), Reproduced by kind permission of BBC Worldwide Ltd, © Jean Smith and Nina Grunfeld, 1993,
Text 5
THE FATHER'S ROLE
INEVITABLY, the responsibilities of parenthood bring with them changes in your own and your partner's relationship and lifestyle. Instead of being a couple with shared interests of a varied nature, you have become the parents of a tiny infant whose round-the-clock needs call for priority. And with so much of your time taken up with meeting those needs, there is a risk that your partner may feel "left out" and rejected.
The more he is involved in baby care, the less likely this is to happen. With so much emphasis on "bonding" in the mother and baby relationship, we tend to forget that bonding is also necessary for fathers. To achieve this, a father needs to get acquainted with his baby through sharing in feeding, nappy changing, taking a turn at night-time care, and being aware of the general routine.
A mother can unwittingly delay this process by discouraging offers of help. Perhaps because of her personal anxiety about her baby's well-being, she may find it difficult to trust anyone else to care for her baby. As a result, she may be depriving herself of an opportunity for a much-needed rest or change of scene, while her partner may feel less confident about offering her a "break" in future.
In the past, breast feeding tended to be seen as a barrier to sharing in feeding. As a consequence, mothers felt that they could not be separated from their babies for more than a few hours. But with today's more enlightened approach, breast feeding mothers are encouraged to express their milk, so that another person can sometimes give the feed. (See chapter Breast feeding your baby.)
figure 5. Extract from Bounty Babycare Guide. Reproducedby kind permission of Bounty Services Ltd,
ACCEPTING HELP
If you are used to doing everything yourself, it may be hard to ask for help. But one of the most important things you can do after a difficult birth is to
5 learn how to ask for help and not see it as a sign of your own inability to cope. You will be surprised how many willing hands there are amongst your friends and neighbours.
10 However, if you feel that the problems facing you require more than a helping hand or two, don't be afraid to ask for professional counselling. Ask your health visitor to put you in touch with
15      someone.
If you do need to ask for help, go to your partner first. Many men feel totally left out during the early part of a baby's life.   They   are   often   keen   to   be
20 involved, but are not sure how to go about it. In those early days when baby really only recognises mum, the real truth is that many women want their partner to direct his parenting abilities
25      at them.
Your baby is probably getting all the attention she needs but, if you aren't, try getting your partner to take some of the strain whilst you recover.
30         We live in a society that currently
believes in mastering the body. We whip ourselves into shape with exercise and deny ourselves the foods we like. When you are pregnant - and
35      especially when you are in labour - you
learn the truth. Our bodies have a will
of their own and are not there simply to
be mastered but to be worked with.
Just as it is true that tension during
40      labour can increase the pain, so can
fighting your natural inclinations after
the   baby    is   born   prolong   both
emotional and physical ailments.
Follow your instincts   -  even if that
45 means neglecting the laundry, having a nap, remaining celibate for a while or having that second helping of pudding. Before you know it, things will begin to get easier. M
figure 6. Extract from Coping Together, M&M Magazine.
Text 6