"Trust is like Jell-O. Men and women don't know what it's [referring to trust] made of, but can make it whatever they need it to be to justify a hook-up, stay in a bad relationship, leave what looks like a good one, or protect themselves from relationships altogether. Jell-O is jiggly. It comes in many colors. You can make it into many shapes, like Playdoh, and you can mix it with other ingredients like carrots, fruit, other Jell-Os, and Cool Whip and make it into something that doesn't look like the Jell-O you started with. Trust is jiggly, and it can look like something it is not. You can't trust trust, and you can't trust Jell-O. That's what I've learned from working with [low-income couples]."

This Jell-O/trust analogy was presented to us by Ms. Renee Jenkins (a pseudonym), a key informant for the Three-City Study ethnography. Ms. Jenkins directed a church sponsored Healthy Marriage Program for low-income singles and couples in one of the study's Chicago neighborhoods. She was responding to questions we posed to her about a conundrum in our data. Within our ethnographic sample of 256 low-income mothers of young children, 96% of them avidly voiced a general distrust of men, with statements like, "Don't trust a man any farther than you can throw him;" or "En los hombres no se puede confiar [you can never trust a man]." Several existing studies have suggested that these general attitudes about gender distrust may, in part, be responsible for declines in marriage among the poor (Carlson, McLanahan, & England 2004; Edin & Kefalas, 2005). That distrust, however, did not deter mothers in our study from becoming involved in or serially seeking marital and cohabiting unions with men. In fact, 82% of the mothers were "married," cohabiting, or in a series of transitory unions with men. Some mothers moved in with romantic partners after knowing them for less than a week. Our questions of our data and to Ms. Jenkins were: If broad-based distrust of men is indeed a deterrent to marriage and other forms of intimate unions, why are so many of the mothers in our study declaring their strong distrust of men, yet entering into so many relationships, many with the hope of forming sustainable marriages? Why doesn't the pervasive sense of distrust lead them to delay rapidly-paced cohabiting relationships, for example, until they have cautiously accumulated evidence of a partner's trustworthiness?

We believed that Ms. Jenkins' Jell-O analogy coupled with the theoretical work of Herbert Blumer (1955) offered the appropriate conceptual guides to address these questions in our data analysis. Ms. Jenkins' description of trust brought an important point to the forefront: People enact their own brands of trust in relationships that may be distinct from the attitudes about trust that they espouse in public arenas. Blumer's (1955) theory of social action suggested that these variants of trust are principally shaped by individuals' personal situations and that situated forms of trust likely trump general attitudes about trust in shaping individuals' choices about intimate unions. Thus, in our data, we could expect low-income mothers to exhibit a variety of trust-placing approaches toward their romantic partners based on a range of situationally-specific conditions (e.g., economic uncertainty) or what Ms. Jenkins referred to as "ingredients." Was the logic we adopted for data analysis from Ms. Jenkins and Herbert Blumer correct? Did the mothers in our study, despite their generalized distrust of men, enact variable forms of trust that allowed them to enter into romantic unions?

What did the data show?

Ms. Jenkins and Herbert Blumer were correct. Beyond mothers' public declarations of distrusting men, we identified four forms of situationally-specific trust in the data and labeled them as suspended, compartmentalized, misplaced, and integrated. These forms were clearly shaped by the life circumstances of the mothers, such as their histories of physical or sexual abuse. The defining features of trust that emerged in the data were: (a) mothers' attention to and reliance on evidence of partners' trustworthiness (ranging from minimal to extensive), and (b) partners' interdependence, which was the degree to which mothers' and their partners' lives were intertwined. Although mothers demonstrated several variants of interpersonal trust within and across their relationships, the majority (86%) had a penchant for using one type more frequently than others. That is because mothers tended to select partners with similar characteristics as they moved from one relationship to another. Below, we briefly describe the forms of trust that seemingly-distrusting women used in particular relationships, and we discuss the implications of this phenomenon for marriage programs and policies. For a detailed discussion of this research, please see the article, "The Role of Trust in Low-Income Mothers' Intimate Unions," in the December 2009 issue of the Journal of Marriage and Family (Burton, Cherlin, Winn, Estacion, & Holder-Taylor).

Suspended Trust

Suspended trust was observed among 26% of the mothers in the sample. These mothers expressed a general distrust of men, yet they occasionally entered into arrangements with men that appeared to be quasi-intimate partnerships. These partnerships typically began very quickly after what one mother called, "a short getting-to-know-you hook-up." Suspended trusters devoted little attention to acquiring hard and fast evidence of their partners' trustworthiness. They generally entered relationships with the interest of getting particular needs met and usually under circumstances when a short term crisis (e.g., needing money to pay the rent) arose in their lives. As such, mothers who suspended trust were not initially concerned with whether their unions would survive beyond getting their current needs met and they saw no need to delve deeply into their partners' personal histories. These mothers also did not engage in elaborate fantasies about their relationships. Rather, they voiced realistic appraisals about their partners' capacities to meet their needs, often brashly speaking about their partners' limitations and objectionable character traits. Trust in these situations appeared to be "hanging in the balance."

Interdependence among partners with suspended trust was slight at best. Mothers and their partners had very limited and specific roles in each others' lives and were very clear about expectations concerning those roles. Mothers created specific boundaries in these relationships: They did not try to deeply embed their partners in their social networks and they did not expose their own emotional vulnerabilities to them. Instead, mothers appeared to suspend judgments about trust knowing that the emotional risks of the arrangement were low and the duration limited. Overall, mothers who enacted suspended trust saw their trust behaviors as "not the real kind of trust." In viewing trust and relationships in this way, mothers retained governance over every aspect of their lives in these unions and suffered little to no distress when relationships ended. In fact, they often had other potential partners, from whom they were just as emotionally detached, waiting in the wings. The ethnographer's fieldnotes on the case of Angie, a 29-year-old White mother of two daughters illustrate this form of trust:

Angie actively sought partners who could meet her short-term needs for purchasing things for her house and her children. "I ain't looking for love," she declared. "I need a man to help me for a minute, and he's out of my house after that. I get what I need, he gets what he needs, and it's a done deal. I don't need to know nothin' about how he gets what he gets [e.g., acquiring financial resources]. I don't want to know nothin' that particular."

Compartmentalized Trust

Twenty-six percent of the mothers in the ethnographic sample demonstrated patterns of compartmentalized trust. Like mothers who suspended trust, mothers who compartmentalized trust had relationships characterized by limited interdependence. They relied on their partners for very specific things, such as taking care of their children, or in the case of some mothers who had emigrated from Central America, providing financial support for their family "back home." These relationships were often very practical with little emotional base, meaning that when these relationships ended, there was minimal emotional indignation. As such, occasionally, mothers had other partners on the side primarily to meet their own emotional needs.

Mothers who placed compartmentalized trust in their partners differed from those who used suspended trust in that their relationships developed gradually, moving from romantic fantasies to realistic views of the complexities of romantic unions over an extended period of time. Compartmentalized trusters also tended to invest considerable effort in determining the trustworthiness of their partners, but only in specific domains. Mothers usually showed little interest in determining whether their partners were trustworthy across domains that were not critical to theirs or their children's immediate needs. For example, Hortensia, a 28-year-old mother of three sons indicated that she trusted her partner, but only to provide financially for her children and to send money to support her family in Central America. She said, "Our relationship is not love, I do not trust him with my heart. We trust each other only for some things. It's like a business agreement and I can depend on him to do his job."

Misplaced Trust

Twenty-three percent of the mothers exhibited patterns of misplaced trust. Mothers who misplaced trust in their partners put minimal value on gathering evidence about their partners' abilities to be trusted and based their trust on "fantasy," "hope," and self-contrived, often inaccurate stories of their partners' virtues. They typically misrepresented or explained away threatening information about their partners' behaviors, even when there was some recognition of disturbing evidence about the partners by family members, friends, or the ethnographers. Mothers who misplaced trust also tended to focus on sequestering long-term relationships, but frequently without success. They usually escalated a relationship to coresidential status soon after meeting a man even though their knowledge about him was incomplete. Both parties in these relationships tended to have excessive needs: the woman, primarily emotional, and the male partner, material, most often housing. Interdependence within these unions ran high, but the relationships rarely lasted longer than a month or two and frequently were abusive. The nature of the relationships commonly resulted in volatile breakups with mothers experiencing intense feelings of emotional betrayal. More importantly, these arrangements put children at significant risk for witnessing abuse or being abused by the mother's partner. And, these mothers usually did not have another potential partner in the wings for the duration of these relationships.

There was little variability among mothers in the sample who misplaced trust, such that Helena, a 42-year-old African American mother of eight children, exemplifies the characteristics of mothers in this category. She had relatively poor health and suffered from depression and anxiety that she attributed to being sexually abused as a child and not being able to find the "right man" as an adult. Moreover, her partners rarely provided her with any evidence that they were trustworthy. For example, her most recent co-resident boyfriend frequently climbed out of their bedroom window at night to visit another woman down the hall in their housing project. Yet, Helena continued to hope that their relationship would last. She said, "He ain't going nowhere. We need each other." Within one week of making that statement, Helena's boyfriend moved in with the other woman. When asked what she would do if her boyfriend returned, she said, "He can come back. I know he didn't mean to hurt me. I know way down inside, I can trust him."

Integrated Trust

Twenty-four percent of the mothers demonstrated patterns of integrated trust. Mothers who engaged in integrated trust behaviors tended to enter relationships gradually, placing high value on evidence of their partner's ability to be trusted across multiple domains over time. Unlike those mothers who engaged in misplaced trust, these mothers described their relationships as balanced between reality and fantasy. They desired romance but they also understood the realities of what it took to make relationships work for the long haul.

The lives of the mothers who enacted integrated trust were deeply intertwined with those of their partners and represented both emotional and functional orientations about the relationship. Mothers depended on their partners for a broad range of needs including emotional, spiritual, financial, social, parental, and recreational support, and their partners likewise depended on them. For these couples, emotional ties tended to be deeply-rooted and stable such that breakups involving violations of trust were emotionally capricious with mothers struggling for long periods of time to move from interdependence to independence. These mothers rarely had other potential partners on the side, and if they moved on to another relationship after a break-up, their conversations with new partners were characterized by reminiscent statements about emotional connectedness to their former partners.

Shana is a 26-year-old African American mother of two children who exemplifies this form of trust. Despite Shana's socialization for a general distrust of men, she recalled that she decided to pursue a relationship with her husband after "checking him out" for a year. "I watched how he handled his business and treated other peoples' feelings, and if he did what he said he was going to do . . . and he did." Shana and her husband were very mindful of each other's needs and devoted special time in their days to tend to their relationship. She stated, "We trust each other very much. I don't know what we would do without each other."

Implications for policy and marriage programs

In our study, we wanted to move beyond merely attributing the declines in marriage among the poor to generic attitudes about gender distrust. Rather, we wished to understand the kinds of limited trust that poor mothers created for specific, intimate unions; that is, what molds they made their intimate union Jell-O fit. We also discerned how mothers' trusting behaviors were shaped by their personal situations. For example, we found high levels of physical and sexual abuse among mothers who engaged in misplaced trust behaviors but substantially lower levels of abuse among mothers who showed integrated trust. Eighty seven percent of the mothers who engaged in misplaced trust had extensive untreated histories of physical and sexual abuse; whereas 31% of the mothers who enacted integrated trust in their partners had a history of abuse.

We think that our findings may have implications for programs and policies relating to marriage. For example, if gender distrust were indeed at the heart of what some see as the marriage problem, then it would follow that restoring a generalized sense of trust should be a central goal. But while building general trust between women and men is a worthwhile endeavor, it may not be the most effective way to engage the issue of trust in promoting marriage. What is required, in addition, is reducing the use of forms of trust (e.g., misplaced) that often lead to the formation of intimate partnerships which have little chance of becoming lasting, healthy marriages.

Moreover, too often a pro-marriage stance is taken to mean urging single mothers to quickly re-partner and marry. But ultimately, the goal of promoting marriage among the poor and near-poor may be better served by urging them to take their time forming partnerships and to carefully examine how suitable their prospective partners are for lasting intimate relationships. By doing so, they could make better judgments about trustworthiness and choose partners who were indeed worthy candidates for marriage. The problem, in other words, is not just getting women to start trusting men; rather, the problem also is getting them to stop trusting men in ways that are not conducive to stable partnerships and to start using trust in ways that are more likely to lead to lasting, healthy unions.

Acknowledgements

We thank the Administration on Children and Families for supporting our research on marriage and romantic relationships among low-income families through grant 90OJ2020 and the National Science Foundation for their support of our work on intergenerational resource allocation through grant SES-07-03968. Additionally, we acknowledge core support to the Three-City Study from the National Institute of Child Health and Human Development through grants HD36093 and HD25936 as well as the support of many government agencies and private foundations. For a complete list of funders, please see www.threecitystudy.johnshopkins.edu. We also acknowledge the efforts of our co-authors Donna-Marie Winn, Angela Estacion, and Clara Holder-Taylor on the article "The Role of Trust in Low-Income Mothers Intimate Unions" which appeared in the December 2009 issue of the Journal of Marriage and Family. This research report presents findings from that article. Most importantly, we thank Renee Jenkins and the families who graciously participated in the research reported in this report.

References

• Blumer, H. (1955). Attitudes and the social act. Social Problems, 3(2), 59-65.
• Burton, L.M., Cherlin, A., Winn, D.M., Estacion, A., & Holder-Taylor, C. (2009). The role of trust in low-income mothers' intimate unions. Journal of Marriage and Family,71(5), 1107 - 1124.
• Carlson, M., McLanahan, S., & England, P. (2004). Union formation in fragile families. Demography, 41, 237-261.
• Edin, K. & Kefalas, M. J. (2005). Promises I can keep: Why poor women put motherhood before marriage. Berkeley: University of California Press.