Growing Points for Coparenting Theory and Research

Abstract

Though the study of coparenting is still in its infancy, an explosion of coparenting research is in the wings. This paper identifies several emerging issues in coparenting theory and research to guide work in the years ahead, including issues in definition, conceptualization, and measurement; the interface between coparenting and adult development; and conceptual benefits that will accrue from studies of interadult coordination across diverse cultures and family systems. We emphasize that theory must lead empirical efforts, that across family systems the strongest coparental alliances are likely to be those in which the coparents both experience and provide support and solidarity for one another’s parenting efforts with the children, and that measurement approaches need to be expanded so as to capture more comprehensively each family’s organizational cooparenting structures and themes.

Throughout history and around the globe, perhaps the most fundamental task of adulthood has been the parenting of young children. And in the overwhelming majority of family systems cross-nationally, such parenting has been collaborative, shared by two or more other adult caregiving figures. Those assuming these shared cocaregiving roles have varied considerably as a function of social class and cultural heritage. In some groups, it has been children’s biological, adoptive, or foster parents who have jointly cared for and parented them. In others, it has been mothers, together with female and/or (much less often) male relatives within their own or their husband’s extended kinship systems assuming the cocaregiving roles. In these latter family systems, fathers have rarely been altogether absent, though in some groups they may have been almost totally inactive in day-to-day decisions concerning the child, or differentially active with children of certain ages or genders. Nonetheless, even when nearly all of children’s regular contact has been with mothers and other female relatives, fathers have often been psychologically significant parenting figureheads in the family.

Oddly, in the face of these realities, most studies of families and development have failed to recognize the powerful and far-reaching impact of coordination (or lack of coordination) between different adult caregivers in their raising of children together. Despite a prodigious and scholarly literature on parenting and its effects (a literature that itself is commonly referred to as “family” influences on child development), most accounts of adult, child, and family development have taken narrow views in which one parent and one child at a time have been the foci. Our knowledge base on children, adults, and families contains very few data on variability among coparenting dynamics either within or across societies, and virtually no data on whether or how distinctive coparental patterns in the family group reflect adult development or influence child adaptation within different cultures.

This circumstance is slowly beginning to change, as attested to by the papers in this volume and by other recent reports of coparenting in diverse family systems. Much important conceptual and empirical work lies ahead, however. This essay identifies important issues in the conceptualization and empirical study of coparenting, highlighting critical topics in need of competent study. Because of space limitations, we cannot address all the important innovations in conceptualization introduced by the initial wave of theoretical papers and coparenting investigations published over the past decade, or all the questions these seminal papers raised. Rather, the intent of this paper is to articulate several important points that we believe will help provide some necessary coherence to this emerging field of study. We focus on the following major points: issues in definition and conceptualization, coparenting and adult development, and interadult coordination across diverse cultures and family systems.

Conceptual and Definitional Issues

What Constitutes the Domain of Coparenting and Coparenting Research?

In the research literature, the term “coparenting” is used frequently as a noun rather than as a verb. But coparenting is an enterprise, one involving the coordination among adults responsible for the care and upbringing of children (Feinberg, 2003; McHale, Khazan, et al., 2002; McHale, Kuersten-Hogan, Lauretti, & Talbot, 2001). It involves far more than just the carrying out of childcare responsibilities (McHale, 1995; McHale & Fivaz-Depeursinge, 1999; McHale, Khazan, et al., 2002; McHale, Lauretti, Talbot, & Pouquette, 2002), proceeding even when fathers never change a diaper, warm a bottle, or get up with a colicky child at night. Indeed, coparenting can proceed in families where the child’s father does not reside with the child’s mother at all (Ahrons, 1981; Black, Dubowitz, & Starr, 1999; Coley & Chase-Lansdale, 1999; Lerman, 1993; Maccoby, Depner, & Mnookin, 1990), as well as in families without fathers (Apfel & Seitz, 1991, 1999; Patterson, 2002; Patterson et al., 2004) or mothers (Hamer & Marchioro, 2002; Hilton & Macari, 1997; Silverstein, 2002). To be sure, the division of childcare labor can provide a very important window into family commerce, as illustrated poignantly by Patterson et al. (2004). But conceiving of coparenting principally as the division of childcare labor provides a very limited perspective on the joint activities of adults responsible for the care and upbringing of children. For whereas women assume the primary burden of child-related tasks in nearly all families across the globe, most men are committed to and involved with children and help cultivate, in very formative ways, children’s adaptive styles. However, in many cultures men’s major contributions in shaping children’s affect regulation, competencies, and adaptive profiles are realized largely in the context of playful, stimulating activities rather than in the context of caregiving, especially with infants and young children (Levy-Shiff & Israelashvili, 1988; McHale & Huston, 1984), and hence their roles as coparents need to be evaluated in ways beyond the amount of childcare responsibility they assume.

Beyond the sharing of childcare labor, three primary, core features of coparental alliances were initially articulated in reports by McHale (1995) and by Belsky, Crnic, and Gable (1995): these included the degree of solidarity and support between the coparental partners, the extent of dissonance and antagonism present in the adults’ coparental strivings, and the extent to which both partners participated actively in engaging with and directing the child. These three coparenting indices (together with the division of childcare labor) have remained at the center of most research and conceptual efforts in the decade since. Feinberg (2003) and Van Egeren et al. (2004) have offered attempts to expand on or otherwise embellish these three core features; Feinberg’s expanded conceptualization of “joint parental management” seems an especially useful one, linked as it is to Minuchin’s seminal views (Minuchin, 1974) of the coparental unit as the family’s “executive subsystem.” Such efforts at clarification can only help improve and organize future work in this field.

Coparenting as a Dyadic and as a Polyadic Construct

In describing the sanctity of the parental “holon,” Minuchin (1974) emphasized that conceptualizations of the coparental unit of two or more individuals must be kept boundaried and distinct from other family and extrafamilial systems (such as older siblings providing care, daycaregivers, or grandparents not involved in day-to-day executive decision making about the child). Along this line, Van Egeren et al. (2004) propose that coparenting might best be viewed as a dyadic construct, given that many coparenting relationships involve only two adults. We concur, in large part, with this notion. Not all adults who care for children share privileged roles as executive decision-makers, and researchers need to expend efforts to ascertain precisely who it is that any given family endorses as its coparental team.

In a different sense, however, coparenting is also an inherently triadic or polyadic construct. The coparenting enterprise is concerned only and always with the guidance and upbringing of one or more children for whom the caregiving individuals share responsibility, and in studies of families with two young children (e.g. McConnell, Lauretti, Khazan, & McHale, 2003; Volling & Elins, 1998), it is often not possible to characterize a coparental alliance as “cooperative” or “competitive” without reference to which of the two children is being coparented. That is, coparenting stances are elicited differentially by the qualities afforded by a particular child, including (but not limited to) child age and infant birth order (McConnell et al., 2003). This is important, because virtually all of the knowledge base in this developing field has been built on studies of parents and one child, even when there have been siblings in the families of the child targeted for study.

A different, but also pertinent issue relevant to this question of coparenting as a dyadic or a polyadic construct is the fact that coparenting has sometimes been construed as a subsidiary of marital partnerships (e.g. Cohen & Weissman, 1984). Yet, as we have developed in great detail elsewhere (McHale, Khazan, et al., 2002), coparenting partners need not be married partners—and in fact, when we look globally, often are not. Hence, it muddies the waters considerably when coparenting and marital functioning are discussed interchangeably (e.g., Doherty, Kouneski, & Erickson, 1998). Mothers and grandmothers or mothers and aunts sharing parenting have no marital dynamics to replicate. Coparenting relationships endure in families even when marriages end. And, perhaps most importantly, even in families where both marital and coparenting systems do coexist, linkages between family domains are not always straightforward (Ablow, 1997; McHale, 1995; Talbot & McHale, 2004).

As a rule of thumb, we advocate that researchers studying coparenting relationships in different family systems keep a clear head about who is responsible and not responsible for the guidance, care, and upbringing of children. It will not benefit the field if conceptualizations of coparenting are limited to husbands and wives, or to couples who share both conjugal relations and coparental responsibilities. Families vary in terms of who they do and do not endorse as agents and executive decision-makers in the long-term enterprise of raising their children, and definitions of the coparenting system need to flexibly accommodate these different configurations. As we have indicated, there is also a need to consider whether coparenting can ever be assessed completely independently of the particular children involved, as studies of multichild families and of stepfamilies (e.g. Crosbie-Burnett & Ahrons, 1985; see also McHale, Khazan, et al., 2002) indicate.

Within nuclear family systems, we advocate retaining distinctions between coparenting and marital relationship systems—effective coparenting partnerships can bond struggling marital partners, whereas ineffective ones can drive a wedge between two people who, on their own, might each be very adept parents (Beitel & Parke, 1998; McHale, Kuersten-Hogan, Lauretti, & Rasmussen, 2000). Studies that have disentangled coparenting and family group processes from husband–wife relationship systems (e.g., Bearss & Eyberg, 1998; Belsky & Hsieh, 1998; Floyd, Costigan, & Gilliom, 1998; Lindahl & Malik, 1999; McBride & Rane, 1998; McHale & Rasmussen, 1998; Schoppe-Sullivan, Mangelsdorf, Frosch, & McHale, 2004) have made fundamental contributions to this emerging field, and we encourage researchers working in this field to continue considering such distinctions. Indeed, in that well-functioning marital partnerships often set the stage for better coordinated coparenting partnerships (Belsky, Crnic, & Gable, 1995; Katz & Gottman, 1996; Katz & Woodin, 2002; Lewis, 1989; Lindahl & Malik, 1999; McHale, 1995; McHale, Kazali, et al., 2004; McHale, Khazan, et al., 2002; Van Egeren, 2003), researchers should look more closely at families that do and do not fit this general trend (e.g. Talbot & McHale, 2004). Ascertaining what is different about families in which partners struggle in their marriage but develop a sound coparenting partnership nonetheless will be of great value from both basic research and clinical vantages.

The Interplay Between Theory and Measurement

Efforts to describe the unique features and contributions of coparental functioning within family systems will be most likely to advance our current state of understanding when they are guided by top–down conceptual approaches, rather than bootstrapping from the nascent and limited assessment paradigms and tools currently available to researchers in the field. As an example, Kurrien and Vo’s conceptualization of fluidity of parenting boundaries within Vietnamese American families (Kurrien & Vo, 2004) is an important and innovative addition to the way in which coparenting has typically been conceptualized in two-parent nuclear family systems, and introduces a construct that might well be studied across different cultural and subcultural groups as well as within the culture in which it was defined.

A broader theory/measurement issue concerns the blurring of measures with the constructs that the measures are intended to estimate (Edwards & Bagozzi, 2000). Implicit in the theoretical underpinnings of coparenting research is the notion that in strong coparental alliances, the coparenting adults each provide for and experience from one another solidarity and support. As noted by Van Egeren et al. (2004), most studies of coparenting phenomena to date have attempted only limited assessments of the coparental alliance, estimating the strength of this alliance either by inquiring about felt coparenting support or personal coparenting conduct, or by sampling coparental behavior directly during very brief staged observations. The impression created by different reports using one or the other of these two different assessment strategies can be one of disconnect between the personally held experience of support and validation from coparental partners, and the publicly observable coconstruction of coordinated and supportive (or, miscoordinated and antagonistic) coparental and family group process.

This is an unfortunate misperception, thoroughly missing the key point that an alliance entails both private and public commitments to a common cause. True allies are those who not only agree to cooperate, but actually do so, and strive on multiple fronts to help strengthen and enhance the chances for success of mutually agreed-upon goals and aims. To be sure, both observations of coparenting behavior and self-reports of perceived coparenting support can provide useful windows into the underlying coparenting scripts and structures that organize the family. But the critical question that needs to guide both research and clinical efforts is have we spent enough time with the family and gathered enough pertinent information to be certain that we have accessed these themes and structures from all important angles? If our assumption is that the family is organized and guided by core coparenting scripts and structures, then a multipronged approach in assessing every family is called for, with less emphasis given to research-driven questions such as, “which self-report measure is better?” or “which explains more variance, self-reports or observations?” Such a clinical approach is most elegantly exemplified in the work of J. Lewis and colleagues (Lewis, 1989; Lewis, Owen, & Cox, 1988).

We advocate several changes to better our current assessment strategies:

1. Great care needs to be taken in contextualizing and grounding data drawn from brief family observations, especially when the paradigms and measures being used to estimate a coparenting dynamic have not yet been standardized or validated for families with infants or children of particular ages or stages, or used with families of different cultural groups. Take for example the observation of adults interacting together with their 6-month-old infant in a very short unstructured play or teaching situation (e.g. Van Egeren et al., 2004). Striving to capture something of essence about meaningful coparenting behavior from this kind of interactionmay or may not be a successful route for evoking the core coparenting themes that organize families. The use of brief family interactions with 6-month-old infants did prove a successful strategy for Schoppe-Sullivan et al. (2004) and Frosch, Mangelsdorf, and McHale (2000), as it did for McHale (1995) in a study of older (8- to 11-month-old) infants. In these earlier studies, coparenting behavior evaluated during the staged play interactions was significantly associated with independently assessed marital behavior, mirroring marital-coparenting linkages documented in numerous other studies (e.g. Belsky et al., 1995; Katz&Gottman, 1996; Katz&Woodin, 2002; Kitzmann, 2000; Lewis et al., 1988; Lindahl & Malik, 1999; McHale, Kazali, et al., 2004). But similar linkages were not as clearly in evidence in the Van Egeren et al. report, leading to questions about the viability of the paradigm used, or about the sample studied.

   Hence, if single shot and very brief assessments of interadult play or teaching behavior with child(ren) are to remain a measurement strategy of choice for coparenting researchers, it might be wise to consider capitalizing on the strategy of “intensification” advocated by Minuchin (1974). That is, it might be most effective to evoke key coparenting processes by devising paradigms that present a magnified challenge or emotional press, rather than relying on a brief teaching task. As yet, our field has no analogue to the well-validated “Strange Situation” paradigm used to assess dyadic parent–child attachment (Ainsworth, Blehar, Waters, & Wall, 1978) that could be widely used to optimize the valid evaluation of important coparental behaviors during infancy or toddlerhood. At the moment, Fivaz-Depeursinge and Corboz-Warnery’s Lausanne Trilogue Play (Fivaz-Depeursinge &Corboz-Warnery, 1999) has shown promise for assessing coparenting dynamics in a standardized and challenging situation during the early stages of infancy (McHale, Kazali, et al., 2004), and we are currently experimenting with a modified still-face paradigm to evoke mild strain for both infants and coparents (McHale, Berkman, Kavanaugh, Carleton, & Alberts, 2004). At this point in theory building, we advocate that whenever feasible, investigators using observational methodologies consider assessing families during both stressful and nonstressful circumstances. At minimum, including assessments evoking coparenting dynamics during moments of emotional press or challenge would be of use in helping to locate data drawn from brief play interactions into a broader portrait of the family’s core coparental alliance.

2. An effectively functioning alliance is one that strives toward achievement of the same, mutually agreed upon aims in both public and private settings. Accordingly, comprehensive assessment of the family’s coparental alliance must extend beyond the observable dynamics of triadic or family commerce to assess parental behavior occurring during dyadic, parent–child moments together (McHale, 1997). For, the solidarity of the coparental alliance is fundamentally affected by parents’ behavior during the private communications they have with their children about the coparenting partner(s) or the family’s operating structure in the absence of those other coparents. That is, parents can and do use time alone with children to either solidify or weaken the parenting alliance by supporting or denigrating the absent coparenting partner to the child (McHale, 1997).

   Relatedly, a parent’s actual strivings to promote coparental solidarity will be less meaningful in families where the coparental partner questions his or her true support, affirmation, and validation. Indeed, it is difficult to imagine circumstances in which effective and coordinated coparenting could be sustained in situations where one partner questions the integrity of the alliance. For this reason, parents’ perceptions of whether they feel validated by their coparental partner (e.g., Abidin & Brunner, 1995; Brody, Flor, & Neubaum, 1998; McBride & Rane, 1998; Van Egeren et al., 2004) are important factors to take into consideration when evaluating the strength and integrity of the coparental alliance. Again, we are not simply addressing marital satisfaction or quality (as illustrated clearly by the data of Bearss & Eyberg, 1998). Moreover, perceived cocaregiving support appears to be as important in coparenting relationships between mothers and grandmothers as it is in relationships between mothers and fathers (Brody et al., 1998).

3. A final, key point pertaining to the refinement of theory-driven measurement is that observed coparental processes must always be understood with respect to parents’ currently held coparenting-related beliefs, hopes, aspirations, and goals. Yet this important step is rarely taken (though see Carmola Hauf, 2004; Carmola Hauf & Bond, 2003). Documenting belief systems is a critical step in estimating the family’s coparental alliance, because parents in virtually every family system hold at least somewhat dissonant ideas about what they would like to see happen with and for their children. In some families, these perspectives may diverge even further over time and come to organize and drive family commerce and inhibit positive development. In other families, dissonance may trigger development, growth, and higher order integration. For these reasons, understanding parents’ private constructions and working models of the coparenting partner(s) and the family’s coparental alliance is a critical step toward contextualizing and interpreting observed dissonance in enacted family practices (McHale, Kazali, et al., 2004; Von Klitzing, Simoni, Amsler, & Buergin, 1999). Understanding family goals and beliefs as organizers of coparental practices provides needed background perspective, and may be of particular importance in cross-cultural investigations (Rao, McHale, & Pearson, 2003).

   To summarize, comprehensive assessments of the beliefs, perceptions, and practices of the family’s key coparenting figures are needed to provide the most clinically meaningful estimate of the strength or tenuousness of the family’s coparental alliance, and of the coparenting themes guiding and organizing the family through time. Of greatest value will be studies providing comprehensive, carefully conceptualized, and developmentally appropriate assessments of the family’s coparenting alliance from the perspectives of both the coparental partners and trained family clinicians (and from children once they are old enough; see McHale, Neugebauer, Asch, & Schwartz, 1999). If not already abundantly clear, we stress that both the “co-” and the “alliance” aspects of the construct must be tended to in future work on this family domain.

Coparenting Relationships and Adult Development

Nearly all conceptualizations of adult development identify parenting as a key process for adult generativity and development (Demick, 2002; Hooker, Fiese, Jenkins, & Morfei, 1996; Leadbeater &Way, 2001; Palkovitz, 1996; Smolack, 1993; Strauss & Goldberg, 1999). Yet nearly 20 years have passed since Weissman and Cohen (1985) proposed that the family’s parenting alliance could be viewed as a telling interpersonal and intrapsychic process in assessing adult development, and few empirical efforts have attempted to test this proposition. Notwithstanding Weissman and Cohen’s limited emphasis on coparenting in family systems where the coparental adults were a husband and wife, the proposition is a sound one. In Weissman and Cohen’s view, adults best able to successfully cooperate and collaborate in the shared raising of children would be those for whom both parents are invested in the child, value one another’s involvement with the child, respect one another’s judgments, and desire to communicate with one another. Very few studies have traced links between adult personality traits and coparenting at all, and those that have (Talbot&McHale, 2004; Van Egeren, 2003) have asked how individual personality traits might influence coparenting. But aside from studies examining how coparenting solidarity affects parenting behavior (e.g., Floyd et al., 1998; Margolin, Gordis, & John, 2001), investigators have not asked the reverse question—how having and raising children together might induce personality change and adult development.

Of currently available data on this topic, Talbot and McHale (2003, 2004) and Van Egeren (2003) have identified as important the assimilative capacity, reactance, and psychological flexibility of parents— facets of individual personality styles likely to have ramifications for how capably the prospective parents will incorporate their child and their partner into an inclusive triadic family dynamic. Whether coparental solidarity ultimately comes to be fostered or hampered by characteristics such as ego resilience or reactance remains an open question, as does the arguably more compelling question of whether coparental solidarity can, over time, lead to enduring changes in adults’ relational comfort and skill (Palm, 1993), or propensities to respond to extrafamilial stressors with greater (or lesser) flexibility or reactance. Recent data have made it clear that coparental functioning can lead directly to alterations in the marital relationship over time (Belsky&Hsieh, 1998; Schoppe-Sullivan et al., 2004), and Grych and Clark (1999) have proposed that felt stress and perceived efficacy in parenting may likewise be affected by the coparental dynamic (see also Feinberg, 2003). Given these indications, it may very well be that that facets of adult adjustment typically seen as more enduring personality traits may likewise transform, for good or ill, as a function of the degree of support, solidarity, and validation experienced within the coparental alliance.

One aspect of adult development that is especially likely to be responsive to coparental support and validation is parental development (Demick, 2002). In particular, the strength of the coparental alliance may be especially relevant for the dimension of social–cognitive awareness (Newberger, 1987; Sameroff & Feil, 1985). In coparental alliances where there exist mild dissonance but where partners provide recurring and supportive feedback for one another as they grow attuned to children’s needs and sensibilities, there is an external structure and support to buoy personal advances in perspective. By contrast, where such possibilities for growth are stunted by regular opposition from a critical, dogmatic, or inflexible coparental partner, parents may exhibit greater adherence to subjectively held belief systems about the child and show little movement toward complexity and flexibility of perspective. Related notions concerning development among custodial and coparenting grandparents, highlighting potentialities for later life intellectual change, have been offered by Ehrle (2001).

In summary, an unexplored frontier for researchers concerned with adult development is the manner in which supportive or antagonistic coparental alliances influence not only parental development, but also other relational and cognitive capacities of longstanding. Most of the focus in this field has been on the driving force of parenthood per se rather than on the nature of coparenting processes within families (Demick, 2002), and hence this is a topic calling for sustained inquiry.

Coparenting Relationships, Like Adults, Also Develop

It is not only children and adults who develop—relationship systems do, too. Yet because the study of coparenting is still so new, most of what we have learned about stability and development of coparental alliances has involved families with infants, toddlers, and preschoolers (Fivaz-Depeursinge & Corboz-Warnery, 1999; Gable, Belsky, & Crnic, 1995; McHale, Kazali, et al., 2004; Schoppe-Sullivan et al., 2004). This is certainly not because coparenting of older children or adolescents is any less important—indeed, coparenting solidarity may be particularly important once children reach the teen years (Allen, Hauser, Eickholt, Bell, & O’Connor, 1994; Brody & Flor, 1996; Sroufe, 1991). Rather, it is likely that most research to date has targeted early family life because we do not yet have a conceptual, developmental roadmap for the evolution of the coparenting alliance through time (McHale & Fivaz-Depeursinge, 1999). Hence it remains unclear whether the primary adaptations set in motion during the earliest weeks and months after a baby is born “set the stage” in some fundamental way for subsequent coparental adaptation throughout the family’s child-rearing years. We do not yet know what critical developments during the early postpartum months might set certain coparental and family trajectories in motion, or what major changes in coparental alliances emerge as children and families develop.

Several seminal transition to parenthood studies (e.g. Belsky & Kelly, 1994; Cowan & Cowan, 1992; Entwisle & Doering, 1981; Lewis, 1989) afford empirical and theoretical guidance concerning early developments in families. Although such studies emphasized the determinants of postpartum husband–wife and parent–child relationship quality rather than the development of coparenting alliances per se, they nonetheless provide several key insights into early coparental development. First, though many theories posit a biological advantage to new mothers and emphasize how society “prepares” women but not men for parenthood, new mothers who are on the front line frequently reach out to a network of other women for knowledge and help. It is by so doing that they come to be more knowledgeable and familiar with their babies than the babies’ fathers—who then look to the mothers to figure out how to relate to their baby (Lamb & Oppenheim, 1989; Stern, 1995).

Such paternal emulation of mothers may be especially common among working class families (Entwisle & Doering, 1981), leading to greater interparental consistency in such families’ dealings with babies than is seen among parents in upper-middle class families. In upper-middle class families, by contrast, fathers are sometimes found to show higher overall levels of engagement with their children than working class fathers—but the nature of such involvement is inspired more by father’s own ideas about how to engage with the baby than by what he is learning from mother (Entwisle & Doering, 1981). Although this “dual-pronged approach” may provide more variety in the types of interactions that babies from upper-middle class families have with their two parents, there is also a greater potential for parental miscoordination if coparents each follow their own internal ideas and imperatives about what is optimal for children, and do not communicate well together.

Several studies have shown the early postpartum months to be particularly difficult for many mothers, as disillusionment and disenchantment with fathers’ low levels of participation in the work of childcare mount (Belsky & Kelly, 1994; Cowan & Cowan, 1992). An important question for any longitudinal model of the evolution and trajectory of coparental alliances concerns what happens from the point of disenchantment. Does an unhappy primary caregiver begin systematically dismissing the other coparent, both during family exchanges and when alone with the child, as the literature on maternal gatekeeping might suggest? If so, does the excluded coparent wind up feeling befuddled, and withdraw? Does s/he end up feeling indignant and remain involved, as frequently happens in families exhibiting hostilecompetitive coparenting dynamics? Or, do the coparents draw upon existing personal or relationship resources to move in a positive direction from the point of conflict? And, can we predict which coparenting teams will follow which of these trajectories from prior knowledge of their personality characteristics and other long-standing features of their relationship?

In future longitudinal work on the evolution of coparenting alliances, process models must be driven by normative data concerning stages of child and family development. For example, toddlers’ movement toward greater autonomy and older children’s movement toward sexuality introduce dynamic new presses for existing coparental alliances. Although coparental dynamics are likely to show some coherence over time, they may also undergo some reorganization in the face of major changes in child maturation or of additions to the family, as Kreppner’s work on the integration of a second child into the family revealed (Kreppner, 1988). On this latter point, as we emphasized earlier, virtually all of our current knowledge base concerning coparenting has been built upon studies of triads. Although certain coparental processes and dynamics may operate similarly with different children, a much more likely circumstance is that different sets of coparenting dynamics will blossom for different children as studies of differential treatment suggest. McConnell et al.’s documentation of systematic differences in coparenting of first- and second-born children in the same family is a case in point (McConnell et al., 2003); coparenting adults from nuclear family systems witnessed greater mutual parental engagement and direction of older than of younger children. This appeared to be a birthorder effect, in that similar differences in coparental engagement replicated in comparisons of first- and second-born infants interacting with the two coparents in a triadic context. That is—second-born infants received less intensive joint parental attention and direction from the coparental pair than did their firstborn age-mates.

In short, we have discovered only the tip of the iceberg in our preliminary studies of early coparenting development. Apart from a handful of timelimited longitudinal investigations of coparenting during infancy, toddlerhood, and preschool, virtually nothing is yet known about coherence and change in coparenting alliances over time or across changes in family membership. Here again, we advocate that subsequent research on this topic be driven by clearly articulated theories of child, adult, and family development (see McHale, Khazan, et al., 2002; McHale, Lauretti, et al., 2002).

Coparenting in Diverse Family Systems

As alluded to at various points throughout this paper, aside from the now extensive literature on postdivorce coparenting, most research on coparenting has involved nuclear, two-parent families. In following up on the clinical and theoretical contributions of S. Minuchin, and expanding notions of coparenting beyond divorced families, investigators needed to start somewhere—and so jump-starting the field by concentrating on two-parent families made some sense. However, as we begin getting a firmer handle on the interpersonal dynamics of nuclear family threesomes, we must challenge ourselves with the question: are we studying coparental processes that have only limited generalizability and applicability in other types of family systems (see, for example, McLoyd, Cauce, Takeuchi, & Wilson, 2001)? Fortunately, such inquiries have begun, and work to date is summarized in greater detail by McHale, Khazan, et al. (2002). Here, we highlight several important growing points.

The first issue concerns a coparental dynamics that most researchers agree is problematic—antagonism between the adults raising the child. It seems reasonable to ask whether such antagonism reflects a facet of the family process possessing broad cross-cultural relevance and generalizability. In posing this question, we are not simply talking about the mother–father–child triangle; as we indicated at the outset, there are many cultural groups (for example, many traditional African and Middle-Eastern societies) where fathers are not involved at all in the day-to-day lives of infant and very young children, and female relatives serve as the salient cocaregiving figures. Rather, we are referring specifically to the “co”-ness or mutuality of support between decision-making adult caregiving figures. Although there is not yet a critical mass of cross-cultural data on this topic, there are relevant data that illustrate the complexity of even very basic questions such as the one we have posed here.

In coparenting-related investigations with families of color in the United States, Lindahl and Malik (1999) have documented that greater interparental and family harmony can be linked to better child adjustment among Hispanic families in southern Florida, whereas Brody has reported related findings concerning the salience of perceived coparenting support in both nuclear, two-parent families and in three-generational, mother–grandmother coparenting pairs in rural Georgia (e.g. Brody et al., 1998). A particularly important study of coparenting in extended family systems has been reported by Apfel and Seitz (1991, 1999). These researchers have documented longitudinal sequela of supportive and nonsupportive origin family involvement with single, low-income African American adolescent mothers in New Haven, CT. Apfel and Seitz identified five models of early family support following the adolescent’s transition to parenthood: parental replacement (“I am raising your child for you”), parental supplement (“we are all raising this child”), parental apprentice (“I will act as your mentor as you learn how to raise your child”), supported primary parent (“this is your child, and he’s your responsibility to raise”), and abandonment (no origin family caregiver available). Their data revealed that teen mothers whose parents either intervened too much (parental replacement) or too little (supported primary parent; abandonment) were more likely to have a second child within 30 months of delivering their firstborn, and subsequently less likely to be rearing their firstborns by the time these children were 12 years of age. One interpretation of these data is that the greatest benefits accrue when early family support is offered in a moderate and balanced way to teen mothers by grandmothers. In an important side note, Apfel and Seitz documented only a modest correlation between coresidence and family support, indicating that living together and receiving help should not be viewed as one in the same.

Also pertinent to the topic of parents and grandparents raising children together is the research of Goodman and Silverstein (2002), whose work is also germane to the issues of adult development discussed earlier in this paper. Studying grandparents raising their school-aged grandchildren either in custodial or coparenting arrangements with their daughters, Goodman and Silverstein documented salient differences in grandparental well-being as a function of ethnicity. After controlling for stresses related to the parents’ problems, African American grandmothers were found to experience the greatest well-being in custodial families. By contrast, Latino grandmothers had the greatest well-being in coparenting families, whereas no differences in well-being distinguished Caucasian custodial and coparenting grandmothers (aside from custodial White grandmothers reporting somewhat higher levels of both positive and negative affect than coparenting White grandmothers). The authors posit that the ways in which grandparenthood is socially constructed in different cultures influences women’s adaptation to different coparenting structures.

Perhaps the clearest examples of how the solidarity of extended family cocaregiving alliances can affect adult development and adjustment will come from cases in which children’s natal parents are forcibly separated from their children, reuniting later after a period in which the children are raised by others in the kinship network. Such a circumstance is commonplace in situations where grandmothers take on the role of rearing their incarcerated sons’ and daughters’ children (Hairston, 1991; Mumola, 2000). In a study of the experiences of addicted, incarcerated parents whose children are being cared for by relatives, Smith, Krisman, Strozier, and Marley (2004) reported that none of the parents in their sample believed that their children were faring better in their absence. And although most did express appreciation for the sacrifices being made by their relatives, ambivalence about the resumption of cocaregiving relationships upon parole was apparent in comments such as “my mom might throw it back in my face,” “I’m scared they won’t respect me and may use this against me,” and “my aunt does not want me to take my daughter back from her.” It seems likely that the most successful postincarceration adjustment, and perhaps also the most significant developmental gains, would be evinced among parents for whom the solidarity of the coparental alliance had been bolstered rather than deconstructed by the coparental relative (c.f. McHale, 1997).

Turning to cross-national data, the work of McHale, Rao, and their colleagues (Liu, Sarin, Fan, Rao, & McHale, 1999; McHale, Rao, & Krasnow, 2000) has examined the concommitants of parent-reported coparental conflict in both North American and Asian samples. The early socialization experiences of young children in Confucian-heritage societies such as China can differ markedly from those of children in the United States. Early socialization of Chinese children is guided by indigenous principles including filial piety and child training (Ho, 1986; Wu, 1986), with emphasis placed on child obedience, minimization of conflict, and family harmony. At the same time, there is evidence that Chinese parents are less affectionately demonstrative than are Western parents, and that Chinese fathers tend to be more emotionally distant from their children than are fathers in the United States (Lin & Fu, 1990).

In a culture where the basic family milieu is so different from that of North American families, is there any relevance of constructs such as interparental conflict or disparagement of the coparenting partner? Two sets of findings are especially relevant. First, the self-reports of Chinese and North American parents reveal marked differences in base levels of coparental conflict across the two cultures, with Chinese parents indicating far less disputatiousness than American parents (Liu et al., 1999). However, within the People’s Republic of China, the correlates of coparental conflict are very similar to those reported in the United States; preschool children in Beijing whose mothers report higher levels of coparenting conflict and disparagement are more likely to be rated as sad/anxious and as demonstrating conduct problems than preschoolers whose mothers report low-coparenting conflict (McHale, Rao, et al., 2000). These data suggest that coparenting conflict may have similar disruptive effects for children in cultures emphasizing collectivism as it does in cultures emphasizing individualism.

Despite interesting parallels between mother-cocaregiver coordination and conflict in different cultural groups and family systems, important differences also define different family situations. For example, when mothers accumulate knowledge about how to parent infants from their contacts with other women during the early transition to parenthood, fathers often find themselves in a “one-down” position concerning parenting knowledge. And when a more knowledgeable mother chooses to behave as “gatekeeper” (Beitel & Parke, 1998; Carmola Hauf & Bond, 2003), father may ultimately withdraw, eventuating in a family dynamic characterized by marked discrepancies in coparental engagement. But what happens in a mother–grandmother family system when it is the cocaregiving grandmother who is more knowledgeable? Are there similar gatekeeping processes? If so, who serves as the gatekeeper, and with what outcomes? Cohen (1999) reported that never-married, low-income African American mothers are extremely sensitive to even relatively small differences between themselves and different cocaregiving family figures concerning perspectives about parenting. It does not seem surprising that mothers’ sensitivity to parenting differences would be amplified in family systems where they possess less parenting knowledge than their cocaregivers (compared with those in which the mothers possess more). Not as clear are the ramifications of such knowledge differences for the resulting coparental and family process, though Apfel and Seitz’s investigations provide the beginning outlines for such a roadmap (Apfel & Seitz, 1991, 1999).

A related point here is that understandings of family process must be driven by an informed understanding of family adaptation in its social and cultural context (Kurrien & Vo, 2004). Developing concepts in majority cultures and applying them indiscriminantly across different cultural or socioeconomic groups without thorough knowledge of family adaptation within such groups is ultimately a doomed enterprise (Burton, 1995; McLoyd et al., 2001; Stack, 1974). Yet in moving beyond nuclear, middle-class family households, researchers serious about studying cocaregiving and family group dynamics can find themselves challenged to accurately define the core coparental unit for study. For example, within the sizeable underclass in the United States today, both blood and fictive kin frequently organize flexibly to provide a broad family base for young children (Crosbie-Burnett & Lewis, 1999), promoting continuous responsibility for dependent children even across changes in relationships among adults. In such families, the relevant coparenting “unit” may involve many or even all of those adults involved in the nurturance and support of an identified child regardless of household membership, demanding that researchers be especially attentive in defining the functional family group (Brooks-Gunn &Furstenberg, 1986; Crosbie-Burnett&Lewis, 1999; Roschelle, 1997; Stack, 1974).

Although cross-cultural coparenting research has begun branching out to accommodate the family process of systems beyond nuclear, two-parent families, much still remains to be learned about cultural differences among two-parent family systems. Take for example the work of Feldman and colleagues (e.g., Feldman & Masalha, 1999; Feldman, Masalha, & Nadem, 2001), who documented significant differences in the coparenting and family patterns of Israeli Jewish, Israeli Arab, and Palestinian Arab families. In Feldman’s samples, the greatest degree of family autonomy and child-centeredness, both features emphasized in Western societies, were observed among Israeli Jewish families. In stark contrast were the interactions of traditional Palestinian Arab families, which revealed little autonomy and were parent-led, in accordance with an emphasis on family hierarchy and submission to elders. Palestinian families also showed lower levels of cooperation—but not higher levels of intrusiveness or competition—together with less expressed affect and mutual gaze than did Israeli Jewish families. And Arab families made little use of toys, perhaps because they did not view family interaction as a context for “teaching” infants. With respect to the Israeli Arab families, although their beliefs concerning education, women’s rights, and religious commitment were more modernized than Palestinian Arabs, their coparenting and family interaction patterns mirrored more closely those of Palestinian families in several regards. Yet two facets of Israeli Arab family interactions did more closely resemble Israeli Jewish family interactions—autonomy and infant-led interactions. Feldman speculated that these features of coparental and family process might have reflected recent cultural shifts in family orientation toward greater autonomy, whereas affective components (activity level, affect, gaze) reflected more entrenched rules of interaction or cultural temperaments less amenable to change.

In summary, although replication studies are needed to substantiate basic findings concerning coparenting and family commerce in nuclear, middleclass, two-parent families in North American and European samples, our understanding of coparenting processes in families will also profit greatly from investigations of coparental processes in diverse cultures and family systems. Such work will be of greatest value when guided by indigenous or “emic” approaches guided by native researchers. Also of critical value will be within-culture studies of the concommitants and sequelea of different coparenting patterns and dynamics for adult, child, and family development.

CONCLUDING COMMENTS

As is true in any newly evolving field of inquiry, coparenting theory and research have each experienced significant growing pains over the past decade. Efforts toward elucidating a comprehensive theory of coparenting relevant to the diversity of modern family systems have sprouted, still riding firmly on the shoulders of Minuchin’s seminal ideas (Minuchin, 1974), as research evidence has continued to accumulate substantiating the distinctiveness of coparental functioning within a variety of different family systems. At the same time, work to date has been dominated by single-method assessments of coparental functioning, sometimes leading to an unfortunate distinction between perceived support and enacted behavior in families.

We have argued that coparenting studies must be conceptually rather than method driven, and guided by well-articulated premises concerning core attributes of the coparental structure within families. At base, strong coparental alliances involve shared goals, the personal experiences of support, affirmation, and confidence in the conduct of the coparental partner(s), and consistency between the coparents’ jointly articulated goals and their conduct in both private (parent–child) and shared parenting venues. All coparental systems can be described in terms of the degree of solidarity and support between the adults who share responsibility for the family’s children and their upbringing; the degree of oppositionality and undermining between these individuals; the particulars of how childcare roles and labor are shared among them; and the extent to which these different individuals actively engage with organizing and managing the daily lives of their children and the decisions affecting them.

There are likely to be other salient features of coparental systems that emerge from culturally attuned investigations of coparental alliances within diverse cultures and family arrangements. It is important that any such work proceed from cultural definitions of shared parenting, rather than importing constructs from the cultures in which coparenting theory and research originated. We also advocate conceptualizations of coparenting that are sensitive to the characteristics of the children being coparented, as there may be greater or lesser levels of solidarity witnessed by the same coparents of different children in the same family. The field would also profit from a theory of coparental development sensitive to developmental stages and changes in children being coparented and to changes in family composition over time, and from investigations of how adult development is prompted by different features and dynamics of coparental functioning.

Perhaps our strongest recommendation is that all future work on coparenting consider first the features of the family’s coparental structure and scripts most relevant to the phenomena that the investigators hope to elucidate, and then to mount a comprehensive strategy for evaluating those core, underlying themes. Although short questionnaires and brief interaction episodes have helped breathe life into this field of inquiry, the next important conceptual and applied advances are likely to come from thorough, clinically sensitive, multimethod evaluations that accurately capture the essence of the family’s coparental alliance along a small number of relevant dimensions considered simultaneously. The dim outlines of this work have begun to take shape, and we look forward with anticipation to the second decade of research on this topic that is proving so relevant to so many family systems.